Hansen v. Country Mut. Ins. Co.

Full title: CHAD HANSEN and MELISSA HANSEN, on behalf of themselves and all others…

Court: United States District Court, Northern District of Illinois

Date published: Sep 25, 2023

18 C 244 (N.D. Ill. Sep. 25, 2023)

Opinion

18 C 244

09-25-2023

CHAD HANSEN and MELISSA HANSEN, on behalf of themselves and all others similarly situated, Plaintiffs, v. COUNTRY MUTUAL INSURANCE CO. d/b/a COUNTRY FINANCIAL, Defendant.

REBECCA R. PALLMEYER United States District Judge

MEMORANDUM OPINION AND ORDER

REBECCA R. PALLMEYER United States District Judge

Plaintiffs Chad and Melissa Hansen allege that their homeowner's insurer, Defendant Country Mutual Insurance Co. (“CMIC”), underpaid them for structural and personal property damage resulting from a tornado. Alleging that CMIC systematically underestimated and underpaid insurance claims, the Hansens ask, for a second time, that the court certify a class consisting of nearly every one of CMIC's insureds who made a claim between 2008 and the present. Before the court is Plaintiffs' amended motion for class certification [320], together with both sides' motions to exclude certain evidence [333, 334, 342, 343]. For the reasons discussed below, Plaintiffs' amended motion for class certification is denied. As part of this ruling, the court grants Defendant's motion to exclude portions of a new declaration by Chad Hansen [334] and denies Plaintiffs' motion to exclude two exhibits (and certain assertions they support) from Defendant's response brief [343]. Plaintiffs' renewed motion to exclude Defendant's experts is denied as moot [342], as is Defendant's motion to exclude certain exhibits attached to Plaintiffs' class-certification memorandum [333].

BACKGROUND

In denying Plaintiffs' first motion for class certification, the court observed that Plaintiffs had “provided little factual information about the events at issue in this case.” (Mem. Op. and Order (hereinafter “First Class Certification Denial”) [309] at 1.¹) Plaintiffs now again seek class certification, this time submitting a 103-page brief and more than 100 exhibits spanning 14,000 pages. (See generally Redacted Mem. in Supp. of Am. Mot. to Certify Class (hereinafter “Redacted Second Class Certification Mem.”) [326].) Frustratingly, numerous factual questions remain unanswered. For example, the court has been unable to locate, among the 100+ exhibits, the Hansens' own, entire insurance claim file, evidence the court would have believed central to their claims. The account set forth below reflects the court's best understanding from a review of Plaintiffs' and Defendant's new, voluminous filings, as well as those submitted previously.

1 The record in this case contains some documents with ECF numbers and others without. The court refers to ECF-numbered pages where they occur. In documents missing ECF numbers, the court refers to documents' internal page numbers. Finally, for convenience of finding parties' statements in depositions, the court refers to their transcripts' provided page and line-numbering. The court uses Bates numbers, where they appear, as a last resort.

I. Factual Background

Chad and Melissa Hansen suffered significant property damage when a tornado swept through their home on June 22, 2015. (Melissa Hansen Dep. Tr. Excerpts, Ex. Z to Redacted Second Class Certification Mem. (hereinafter “Pl.'s excerpts: Melissa Hansen Dep.”) [325-11] at 175:19-176:3.) Once they had a “usable telephone,” the Hansens contacted their insurer, Defendant Country Mutual Insurance Company (hereinafter “CMIC”), to claim losses resulting from the storm. (Id. at 176:14-18.) CMIC had been their insurance provider since January 12, 2008. (Am. Compl. [16] ¶ 7.)

A. The Hansen's Insurance Policy with CMIC

CMIC offers two tiers of homeowner's coverage, referred to here as: Replacement Cost (“RC”) and Actual Cash Value (“ACV”). (Certified Copy of Decl. Page and Policy Booklet in Effect at Time of Hansen Loss, Ex. H to Redacted Second Class Certification Mem. (hereinafter “Hansen Insurance Policy”) [325-4] at 6, 8.) ACV coverage provides that in the event of loss, CMIC will pay:

a. For buildings or structures the lesser of the following, as determined by [CMIC]:

(1) The cost actually and necessarily incurred to repair or replace the damaged property using standard new construction materials of like kind and quality and standard new construction techniques, less depreciation; or

(2) Fair market value.

b. For property other than buildings and structures the lesser of the following, as determined by [CMIC]:

(1) The cost to repair or replace the damaged property using materials of like kind and quality, less depreciation; or

(2) Fair market value.

(Id. at 6.) Depreciation is the deterioration in an item's condition-and thus value-over time. For example, if an item of the insured's property was in “better than average condition” at the time of a loss, the depreciation withheld from the ACV would be “less than the norm,” and vice versa. (Id. at 26.)

A higher level of coverage than ACV, CMIC's RC policy provides that, in the event of loss, CMIC will pay:

[T]he lesser of the following, as determined by [CMIC]:

(1) The cost to repair the damaged property using materials of like kind and quality; or

(2) The cost to replace the damaged property with a new article identical to it. When the identical article is no longer available, or cannot be legally manufactured or constructed, “replacement cost” means the cost of a new article similar to the one damaged which is of comparable quality and usefulness.

(Id. at 8.) In other words, RC entitles insureds to the ACV without deduction for depreciation. This higher coverage, however, comes with a catch: when an insured having RC coverage initially submits a claim, CMIC will pay only the ACV; in order to receive the RC amount, the policyholder must first attempt to repair or replace her lost property within a year of her claim arising. (Id. at 26.) If the insured with RC coverage timely moves to repair or replace her lost items, CMIC then makes up the depreciation it has deducted from the ACV. RC coverage is thus limited to circumstances in which the insured takes steps to replace what has been lost or damaged, as opposed to simply cashing out the policy.

According to its policy, CMIC determines property's depreciation by assessing “wear and tear, deterioration, obsolescence, age, physical condition and reduced market value ....” (Id. At 6.) For insureds' personal property, or “contents,” adjusters have numerous tools at their disposal to make this assessment: they can send information about insureds' property-a television, say-to “outside third party vendor[s] for pricing” (Oct. 28, 2020 Robert Bieber Dep. Tr. Excerpts, Ex. 4 to Sealed Response in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Def.'s Excerpts: Bieber Dep. II”) [332-4] at 69:17-25); they receive detailed lists from insureds “documenting] the make, model, serial number, any - as much information as they possibly can come up with” about their lost property (id. at 9:13-17); they search the Internet for items' prices (id. at 10:1-5); and they can (and “on many claims” do) use a pricing estimation program called Xactimate (LaPrade Report, Ex. L to Sealed Mem. in Supp. of Am. Mot. to Certify Class (hereinafter “LaPrade Report”) [326-10] at 24.) Xactimate (along with a related program called “XactContents”) allows adjusters to estimate the value of items and structures by editing various inputs, like an item's age and condition, referred to as its “standard” or other “grade.” (Redacted Second Class Certification Mem. at 19.)

According to Plaintiffs' analysis of Xactimate data, for the vast majority of contents-claim estimates surfaced during discovery, CMIC adjusters defaulted to assuming each item of claimants' property was in merely “average” condition. (Summ. Table of Xactimate Data for Contents, Ex. CX to Sealed Second Class Certification Mem. (hereinafter “Contents Summary Table”) [326-85] at 2.) Adjusters occasionally deviated from that default, presumably only when pressed by insureds; in such cases upgrades to “above average” condition were about five times more common than downgrades to “below average.” (See id.) Plaintiffs' contents-claim expert, Donald Stafford, spit-balled similar figures in a deposition, confirming that insureds who believed their property had been undervalued would have to push adjusters upward and that “given an opportunity, . . . insureds would express that maybe 25 percent of the[ir] items are in better than average condition,” and that “[f]or less than average condition, if you got 5 percent, that would be high.” (Donald Stafford Dep. Tr., Ex. AZ to Redacted Second Class Certification Mem. (hereinafter “Stafford Dep.”) [325-32] at 143:9-15.)

As to so-called “structure claims”-damage to an insured's dwelling, for instance- adjusters engage in a varied set of practices to estimate loss, depreciation, and the “like kind and quality” of replacement materials. For example, adjusters physically inspect the property, noting the extent of any damage they see (Jan. 24, 2020 Bradley Keck Dep. Tr. Excerpts, Ex. 6 to Sealed Response in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Keck Dep.”) [332-5] at 124:8-125:10); they “work with the insured . . . and their contractor” (May 7, 2019 Robert Bieber Dep. Tr. Excerpts, Ex. 2 to Sealed Response in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Def.'s Excerpts: Bieber Dep. I”) [332-2] at 80:23-24); and they consult home improvement stores like “Home Depot, Lowe's, Menard's . . . to identify the type of material that [they]'re dealing with.” (Def.'s Excerpts: Bieber Dep. II at 8:24-9:2.)

Significantly, CMIC's estimates are not set in stone. As Robert Bieber, the head adjuster of CMIC's Large Property Loss Unit (LPLU), explained in a deposition:

I think it's very rare across the insurance industry that a large claim is never going to change. It's just - there's too many - there's unforeseen. You know, adjusters and contractors, we're not going to write - they're not going to write for the unforeseen. So I think there's always a very high probability that there may be additional damage, you know, supplements, however you want to term that, that come back . . . It's an estimate.

(Oct. 2020 Dep. of Robert Bieber Excerpts, Ex. N to Sealed Second Class Certification Mem. (hereinafter “Pl.'s Excerpts: Bieber Dep. II”) [326-11] at 83:23-84:19.) In another deposition, Bieber echoed these sentiments, explaining that “we [adjusters] write what we can physically see is damaged and, from there, the information is . . . given to - I mean, typically we're working with a general contractor.... And once they get into the particular job, they open it up.” (May 2019 Dep. of Robert Bieber, Ex. W to Sealed Second Class Certification Mem. (hereinafter “Pl.'s Excerpts: Bieber Dep. I”) [326-23] at 133:11-19.)

CMIC, like all insurers, keeps records of its policyholders' claim files, including loss estimates from contractors. (See generally, Sample Set Excerpts Part One, Ex. 12 to Sealed Opp'n. to Am. Class Certification (hereinafter “Exhibit 12 Part One”) [332-10]; see also Claim File Record Retention Requirements in 19 States CMIC Does Business, Ex. X to Redacted Second Class Certification Mem. (hereinafter “Insurance Law Table” [325-9]).) As reflected in the agenda for a supervisors meeting, CMIC instructs its adjusters that, for structure claims, “[w]e should also see our own estimate in the [policyholder's claim] file and not that of a contractor, as our estimates will reflect the proper Actual Cash Value and replacement cost.” (Dec. 16, 2009 Supervisors Meeting Agenda, Ex. C to Sealed Second Class Certification Mem. [hereinafter “Agenda”] [3263] at 9).) Plaintiffs' structure-claims expert, Russell Thomas, expressed in a deposition that he would also “expect to see a lot of documentation from contractors in typical insurance companies' claim files.” (Russell Thomas Dep. Tr. Excerpts, Ex. Q to Sealed Second Class Certification Mem. (hereinafter “Thomas Dep.”) [326-13] at 215:13-17.) According to Plaintiffs' analysis of the 313 claim files CMIC produced during discovery (and on which the parties agreed to base their arguments), 63 policyholders' files contain contractor estimates.²(Ex. BA at 8; see also Sealed Resp. in Opp'n to Class Certification [245] at 15 n.14 (noting the parties' agreement that their experts would perform analyses on the 313 claim-file sample set that CMIC produced).)

2 Defendant claims that, in fact, 65 claim files include contractor bids, citing to individual claim numbers in Plaintiffs' Exhibit BA. (Redacted Opp'n to Am. Class Certification at 39 (noting that, in addition to the 17 files Plaintiffs analyze in their Exhibit A, “48 additional structure claim files in the Sample Set . . . contain contractor bids”).) The difference is immaterial for purposes of deciding class certification, so the court simply relies on Plaintiffs' summary table, which lists 63 claims as having contained contractors' bids. (See Summ. Table of Claims Where Contractor Est./Bid/Invoice Exists and the Percentage of the Contractor's Invoice Paid, Ex. BA to Sealed Second Class Certification Mem. (hereinafter “Ex. BA”) [326-34] at 8.)

As to claims handling more broadly, CMIC unsurprisingly seeks to avoid “leakage,” which, in insurance terminology, means overpaying claims. (Sealed Second Class Certification Mem. at 10 n.2.) CMIC directly encourages its claims adjusters to reduce leakage. For example, in one claims adjuster's performance review, attached as an exhibit to Plaintiffs' class-certification memorandum, the adjuster's supervisor notes with approval that “Joe controls leakage by conducting a through [sic] investigation of all losses assigned to him. This includes determining coverage, accurately measuring, scoping, diagramming, and estimating of the loss.” (Performance Reviews Discussing Leakage and Severity, Ex. E to Second Class Certification Mem. (hereinafter “Performance Reviews”) [326-5] at 12.) CMIC training documents also announce “goal[s]” of keeping leakage at low percentages, for the stated reason of avoiding overpayment to insureds and resulting loss of profit. (Internal Country Newsletters Focused on Leakage, Ex. D to Second Class Certification Mem. [326-4] at 2.)

The parties dispute whether CMIC pays its policyholders the full extent of their losses. Plaintiffs analyzed the 313 claim-file sample set, comparing contractor bids with CMIC payments to determine whether CMIC underpaid claims. (See Redacted Second Class Certification Mem. at 10 n.1; Ex. BA at 8.) They conclude that CMIC paid insureds adequately (i.e., 100% or more of contractors bid totals) in just 44% of the 63 claims in which those bids appeared. (Ex. BA at 8.) As counted by the court, the same tables also show CMIC's final estimates matching or exceeding contractors' bids in 21 of the claim files for which such bids appear, or a third of the total. (Id. at 1-8.) For example, the first claim summarized in Exhibit BA-Claim Number 102-0039147-shows CMIC's “[f]inal [e]stimate [a]mount” to be $3,166.13; the policy's deductible to be $500.00; the contractor's bid to be $2,897.63; and the “[a]mount [p]aid” to be $2,397.63 (the contractor's bid minus the deductible). (See id. at 2.)

The Hansens had RC coverage-CMIC's more expansive option. (Redacted Second Class Certification Mem. at 18.) As explained above, CMIC's obligations to the Hansens as RC policy holders were twofold: first, CMIC was required to pay the actual cost that the Hansens incurred to repair their home, minus depreciation as determined by CMIC (i.e., the ACV); and second, if the Hansens undertook repairs within a year, CMIC would pay additional amounts to make up the difference between their incurred repair costs and the depreciation that CMIC had withheld (i.e., the RC).³

3 Once CMIC calculates the ACV, the insured is entitled to it regardless of whether she makes repairs or whether those repairs end up exceeding the ACV. (John LaPrade Dep. Tr., Ex. DZ to Pl.'s Consol. Reply Mem. of Law and Fact in Supp. of Pl.'s Am. Mot. for Class Certification, Resp. to CMIC's Evidentiary Mots. and Pl.'s Mot. to Exclude (hereinafter “LaPrade Dep.”) [355-3] at 2l7:13-218:15.) In other words, in the (likely rare) event that an insured's costs to repair are less than the ACV she has already received, the insured is entitled to the latter, greater sum.

B. The Hansens' Experience

The parties have differing accounts about what happened, what was estimated, and what was paid in the months following the tornado damage to Plaintiffs' property. The court notes those differences where they occur, relying principally on sources cited in the parties' competing briefs.

1. Structure Claims

Within a week of the storm, a CMIC adjuster examined the Hansens' damaged home and shed and estimated the cost of needed structural repairs at $111,585.58 in RC and $88,096.04 in ACV. (June 26, 2015 Est., Ex. M to Redacted Second Class Certification Mem. (hereinafter “June 26, 2015 Estimate”) [325-6] at 20 (Country Mutual-000727.))⁴ CMIC made an initial lumpsum payment of $84,517.14 on June 28 to enable the Hansens to begin paying for repairs. (LaPrade Report at 58.) According to Chad Hansen, the CMIC adjuster who visited their home directly after the storm conducted no more than a “superficial inspection.” (Decl. of Chad Hansen, Ex. O to Redacted Second Class Certification Mem. (hereinafter “Hansen Declaration”) [325-7] ¶ 5.) For example, CMIC's adjuster did not “open[] and clos[e] the windows” when surveying their damaged home, and CMIC originally omitted line items detailing window repair from its repair estimates as a result. (Id.) Dissatisfied with CMIC's estimates, the Hansens requested that a second CMIC adjuster inspect their property the next month; that adjuster's estimate came in slightly higher: $114,230.43 in RC and $90,440.89 in ACV. (July 29, 2015 Est., Ex. AE to Redacted Second Class Certification Mem. [325-14] at 25 (Country Mutual-001145.))

4 Plaintiffs' exhibit has no ECF numbers, so the court counted pages and, for added clarity, refers to the document's Bates number (“Country Mutual-000727”).

Soon after these estimates, CMIC appears to have fielded a bid for the Hansens' structural repairs from a contractor, Elite Construction Company (“Elite”), which offered to repair the Plaintiffs' home for $140,713.16. (Elite Construction Co. Survey of Damages, Ex. AF to Redacted Second Class Certification Mem. [325-15] at 17 (Country Mutual-031286.)) The Hansens rejected this bid. (Melissa Hansen Dep. Tr. Excerpts, Ex. 9 to Redacted Response in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Def.'s Excerpts: Melissa Hansen Dep.”) [347-4] at 55.)⁵Instead, in July, Plaintiffs hired their own contractor, DNR Carpentry and General Contracting (“DNR”), to repair their roof damage and make other repairs throughout their home. (Id. at 55:2356:17; LaPrade Report at 58.) They chose DNR because Melissa Hansen “and one of the owner's wife [sic] went to school together in high school.” (Chad Hansen Dep. Tr. Excerpts, Ex. 1 to Redacted Resp. in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Chad Hansen Dep.”) [347-2] at 137:9-14). From here, the parties' accounts differ.

5 At this stage, neither party discusses why the Hansens rejected Elite's bid. From what the court can deduce, the Hansens were unhappy with how Elite conducted its inspection and an “emergency board-up” of their home. (See Def.'s Excerpts: Melissa Hansen Dep. at 55:1217 (“We didn't like the way that the emergency board-up occurred, and so Chad and I both agreed that we would not be hiring them, and we told them that on multiple occasions.”); see also Pl.'s Excerpts: Melissa Hansen Dep. at 138:5-140:12 (claiming that Elite returned their refrigerator, washer, and dryer machines “moldy,” and that Elite “scratched” their stairs' handrail).) Originally a co-defendant to this action, the Plaintiffs' claims against Elite were dismissed after the company dissolved in 2020. (See 12/22/2022 Minute Entry [350].)

CMIC focuses on the Hansens' repair efforts. The Hansens paid DNR $12,600 to commence repairs on their roof. (Wells Fargo Property Loss Claim File, Ex. 10 to Sealed Mem. in Supp. of Am. Mot. to Certify Class (hereinafter “Wells Fargo Docs”) [332-8] at 3; see also Def.'s Excerpts: Melissa Hansen Dep. at 77:7-11 (noting the Hansens' “advance payment” to DNR for roof repair).) DNR repaired the roof, but before DNR completed any other repairs, the Hansens withdrew from their contract with DNR. This decision appears to have been in part because DNR planned to charge the Hansens “$17,000 higher” than CMIC's estimate for repairs to other parts of their home (Chad Hansen Dep. at 136:23-137:8), and in part due to other disagreements, including Melissa Hansen's belief that DNR was improperly controlling the process such that there was “an issue with us not being able to pick out anything that went into our home ....” (Def.'s

Excerpts: Melissa Hansen Dep. at 73:24, 74:1-4.) DNR eventually refunded to the Hansens the portion of the $12,600 that DNR had not actually spent when fixing the roof, which according to Melissa Hansen ended up being “around 3,000” dollars. (Def.'s Excerpts: Melissa Hansen Dep. at 78:13.) CMIC compares these figures-CMIC's $12,647.32 ACV payment and DNR's $9,600-or-so repair of the Hansens' roof-and conclude that “Country Mutual paid Plaintiffs $3,000 more than the actual repair cost.” (Redacted Resp. in Opp'n. to Pl.'s Am. Mot. for Class Certification (hereinafter “Redacted Opp'n. to Am. Class Certification”) [347] at 19.) When Melissa Hansen was asked whether there were “any issues from your perspective with how [DNR] completed their work on the roof,” she answered “No. Not that I can recall, no.” (See Def.'s Excerpts: Melissa Hansen Dep. at 56:11-14.)

Plaintiffs, on the other hand, ignore DNR's work and its cost and focus instead on CMIC's estimation practices. According to Plaintiffs, CMIC's adjusters used a variety of “common cheats” to underestimate the RC and ACV of repairing the Plaintiffs' roof, including: using “the wrong labor rate” in calculating roof removal expenses; using “Xactimate's generic code for shingle removal” instead of adding a “haul off (ARMVN) [estimate] using a labor efficiency of one-third greater than Xactimate's calculated expense for shingle disposal”; excluding the cost of “removing flashing and vents”; using a below-industry-practice estimate of shingle “waste”; and so on. (Redacted Second Class Certification Mem. at 22-23.)⁶Plaintiffs allege, citing an estimate from their expert Russell Thomas, that through these practices “CMIC underestimated the replacement cost of Plaintiffs' roof by $7,810.21 or 61%.” (Id. at 24). Thomas' 2021 estimation of the cost to repair the roof was $20,597.58 in RC and $20,427.15 in ACV (Thomas Estimate at 23.); CMIC's estimate for the same, in December 2015, was $12,818.28 RC and $12,647.32 ACV (Dec. 15, 2015 Est., Ex. AG to Redacted Second Class Certification Mem. (hereinafter “Dec. 15, 2015 Estimate”) [32516] at 19 (Country Mutual-001282)).⁷

6 In support of these allegations, Plaintiffs cite to numerous exhibits composed of what appears to be Plaintiffs' own analysis of-and speculation from-sets of unauthenticated Xactimate data, as well as computations by their expert Russell Thomas whose work, in part, this court excluded in its previous order. (See, e.g., Thomas Re-Written Est. for Pl.'s Claim, Ex. AQ to Redacted Second Class Certification Mem. (hereinafter “Thomas Estimate”) [325-24] at 2223.) The court shares Defendant's concerns about what appears to be expert opinion offered by counsel without support and notes confusion about such expressions as “haul off (ARMVN) [estimate].” Because the conclusions Plaintiffs draw in the exhibits do not change the outcome of their amended class certification motion, however, the court assumes their admissibility.

7 Simple subtraction does not result in Plaintiffs' $7,810.21 figure, nor do Plaintiffs explain how the $7,810.21 is a 61% reduction in the estimate. As the court does the math- calculating the percentage reduction from $20,597.58 to $12,818.28-Plaintiffs' expert meant that CMIC underestimated the cost of repairing Plaintiffs' roof by around 38%. The numeric confusion is compounded by the fact that the record contains myriad different estimates and other inconsistencies. For instance, though the court found no discussion of this in the briefs, CMIC's earliest estimate for roof repair appears to have been notably higher than its later one, totaling $18,644.41 in RC and $16,552.35 in ACV. (June 26, 2015 Estimate at 15 (Country Mutual-000722.))

Soon after their withdrawal from DNR's contract, Plaintiffs acted as their own general contractor to reduce costs for the remaining (non-roof) repairs to their home. (Chad Hansen Dep. at 136:2-5). As Plaintiffs received-and CMIC “accepted”-bids from subcontractors for various repairs they sought, the estimates climbed, reaching $269,305.60 in RC and $225,762.50 in ACV by December 2015. (See Redacted Second Class Certification Mem. at 20; Dec. 15, 2015 Estimate at 34 (Country Mutual-001297.)) According to John D. LaPrade, Defendant's expert, at least some of these upward adjustments resulted from CMIC's capitulations to the Hansens' demands when there were disputes about coverage. (LaPrade Report at 59-60.) One dispute, for example, concerned CMIC's initial reluctance to pay about $20,000 in “contractor overhead,” which CMIC argued it did not owe given the Hansens were not contractors. (Id. at 57, 59-60.)⁸

8 CMIC appears to have eventually capitulated and offered the Plaintiffs a general contractor overhead amounting to about $20,000. (See LaPrade Report at 60.) So far as the court can tell, the correspondence undergirding LaPrade's recounting is not in the record.

By December 2015, CMIC had paid the Hansens about $225,000 in ACV for structural repairs. (Compare Chad Hansen Dep. at 9:16-24, 10:1 (admitting the figure of $233,750.02 “sound[s] about right”), with LaPrade Report at 60 (noting CMIC paid “over $225,000 in ACV.”)) The depreciation holdback-about $43,500-would later have been paid out as well, upon completion and documentation of the Hansens's remaining repairs. (Dec. 15, 2015 Estimate at 23 (Country Mutual-001286.))

But the Hansens never got that far. Again, accounts differ as to what happened. According to Plaintiffs, CMIC halted negotiations when the Hansens “continued asking questions about disappearing line items and filed a complaint with the Illinois Department of Insurance” to recover a larger sum that they accused CMIC of withholding. (Redacted Second Class Certification Mem. at 12.) According to Defendant, “Plaintiffs never submitted the necessary documentation to verify the final repairs were made.” (Redacted Opp'n. to Am. Class Certification at 10.) In support, Defendant cites deposition testimony of a CMIC representative alleging that the Hansens did not respond to “multiple letters requesting” such documentation. (Def.'s Excerpts: Bieber Dep. I at 293:5-8.) Whatever the cause, the parties reached an “impasse” after the Hansens filed their complaint with the Illinois Department of Insurance, and CMIC turned the case over to its outside counsel prior to the Hansens' filing this suit. (Dec. 7, 2015 Internal CMIC Management Emails, Ex. AJ to Redacted Second Class Certification Mem. [325-19] at 4 (Country Mutual - 016494).)

2. Contents Claims

The tornado also damaged much of the Hansens' personal property. They claimed losses on 707 items of clothing, electronics, and the like. (Stafford Rep. Ex. B, Ex. 11 to Sealed Opp'n. to Am. Class Certification (hereinafter “Stafford Report”) [332-9] at 47.) For this claim, the Hansens used a tool provided by CMIC to make an inventory of their lost items, which they submitted on July 31, 2015; CMIC calculated the ACV and, according to Defendant's expert, “promptly” paid it. (LaPrade Report at 61.) Over the coming months, Plaintiffs submitted receipts to CMIC in bulk as they replaced lost items, and, as required by the RC policy, CMIC reimbursed Plaintiffs for these items to make up any withheld depreciation. (Id. at 61.)

Once again, the parties' accounts diverge. Both sides' experts agree that CMIC classified every individual property item in Plaintiffs' home as being in “average” condition when CMIC entered the Plaintiffs' inventory into its estimating software. (See Redacted Second Class Certification Mem. at 21; LaPrade Report at 61.) Relying on their expert's estimates, Plaintiffs allege that through this practice CMIC “initially underpa[id] the ACV on the Plaintiffs' claim by $10,894.41.” (Redacted Second Class Certification Mem. at 21.)⁹

9 As this court noted in its earlier denial of Plaintiffs' class certification motion, Stafford “does not attempt to justify his methodology” in arriving at this number; he relies on the Hansens' self-reporting. (First Class Certification Denial at 7 n.2.) The court further pointed out that “while [the Hansens] list many of their items as having been in ‘above average' condition, they do not list a single item-in a list of more than 700-as having been ‘below average.'” (Id.)

Defendant draws different conclusions. First, CMIC's expert speculates that classifying all 707 of Plaintiffs' items as being in “average” condition at the outset of a claim, rather than investigating each item's pre-loss condition, “expedite[d] the adjuster's ability to issue the ACV payment and allow the insured to begin the replacement process,” after which CMIC would reimburse any upward difference under the Hansens' RC plan. (LaPrade Report at 62-63.)¹⁰

10 LaPrade notes that, for insureds with only ACV coverage, CMIC's estimation of “depreciation appears to be conservative,” perhaps to account for the lack of RC coverage. (LaPrade Report at 123.) In this sense, CMIC's expert appears to hint that CMIC understood its use of average condition in estimating property depreciation might lead to inaccuracy, which, at least for RC policyholders, CMIC could later rectify when it reimbursed those policyholders for the difference between the estimated value of their belongings and the ultimate cost of replacement.

Second, Defendant points out that CMIC “increased the price” on numerous items “because the items Plaintiffs purchased as replacements were more expensive than the amount initially claimed by Plaintiffs.” (Redacted Opp'n. to Am. Class Certification at 10.) CMIC's corporate representative estimated that “50 some items were overrode price-wise” and that “200 some items were just overrode in general on the [X]act contents,” explaining that the “price . . . the Hansens had placed on the [X]act content sheet and the - what they actually purchased was greater than the amount that was originally allowed and was an override.” (Def.'s Excerpts: Bieber Dep. I at 58:9-24.) According to Defendant's expert, this includes a television “claimed at $2059.13 by the insureds on the inventory and replaced at $2480.30, which was paid by COUNTRY with no question”; Bose speakers “claimed at $2125 . . . and replaced at $2674.99, also paid without question”; and “other items that appear to be significant upgrades.” (LaPrade Report at 61-62.)¹¹

11 LaPrade claims these estimates come from the Plaintiffs' claim file, but since that file appears not to have been attached as an exhibit to these motions, the court was unable to corroborate them. The court remains puzzled by the parties' apparent failure to cite to the Hansens' own claim file. There are snippets among the tens of thousands of pages deposited in the court record. (See, e.g., Ex. H to Donald Stafford Report, Ex. DI to Redacted Second Class Certification Mem. [325-49] (listing the Hansens' 707 personal property items' ACV and RC); Dec. 15, 2015 Estimate (showing CMIC's estimate of structural repairs); Wells Fargo Docs.) But neither party appears to have assembled or presented the Hansens' claim file in its entirety. Or, if it is there, neither side has provided the court with information that might assist in locating it.

In the end, Plaintiffs received RC benefits on 309 items in their 707-item inventory. (Id. at 122.) CMIC paid ACV for other lost contents, as it appears the Hansens did not opt to replace them. The RC payments totaled $115,566.02 (id. at 122-23), which, compared with the $141,394.41 RC estimate for all 707 items, represents about 82% of the RC value of Plaintiffs' claim. (Stafford Report at 30.)

II. Legal Background

Though the court avoids recounting this case's history in depth, a few landmarks bear mentioning. Plaintiffs filed their class complaint on January 12, 2018. (Compl. [1].) In the ensuing years, the parties engaged in extensive, long-running discovery, during which CMIC claimed to have produced

nearly 55,000 pages of documents, responded to more than 100 requests for production, answered 25 interrogatories with multiple subparts, conducted extensive meet and confers with plaintiffs (involving dozens of emails, letters, and teleconferences), produced more than 300 sample claims files, produced two Rule 30(b)(6) witnesses on 21 topics, and worked extensively with one [of] its vendors, Verisk, to provide information requested by plaintiffs.

(Mem. Op. and Ord. [144] at 1.)¹²As part of the discovery process, the parties agreed to rely on about 300 sample claims files that CMIC would produce “similar . . . [to] what was produced in this case in regard to the Hansens.” (Email Chain re Sample Set of Claims, Ex. 5 to Mot. to Exclude Op. Test. of Donald Stafford [251-3] at 2; see also Sealed Opp'n. to Am. Class Certification [245] at 4 n.14 (noting the experts were to rely on the 313 claim files CMIC produced).)

12 By the end of discovery, the number appears to have climbed to at least 70,000 pages. (See Pl.'s Reply in Supp. of Pl.'s Mot. to Exclude the Op. Test. of John D. LaPrade and Stephen Prowse [282] at 30.)

Discovery disputes were bitter, protracted, and unusually frequent. (See, e.g., Docket Entries [75], [144], [154], [163], [178], [184], [223], [243].) Moreover, the conduct of Plaintiffs' attorney was often unhelpful. For example, the magistrate judge handling discovery noted, rejected, and criticized Plaintiffs' repeated accusations of CMIC's “bad faith” and “half-truths,” stressing that “[t]his type of disfavored advocacy has muddled rather than clarified the nature of the parties' multi-faceted dispute and has made the Court's effort to sort things out more difficult.” (Mem. Op. and Ord. [144] at 2.)

A. First Class Certification Attempt

The Plaintiffs filed their first class-certification motion on July 27, 2021. (See Mot. to Certify Class [217] at 1.) Their proposed class was based on alleged uniform underpayments to all of CMIC's policyholders for losses from any disaster from 2008 onward. (Id. at 2.) Plaintiffs defined the proposed class as follows:

All policyholders of Defendant Country Mutual or its subsidiaries who at any point between January 12, 2008 and the present, had their insurance claim(s) accepted and paid, and those payments were based on Defendant Country Mutual Insurance Company's employees[‘] or agents[‘] use of estimating guidelines and practices that were established or designed by Defendant Country Mutual Insurance Company.

(Id. ¶ 7). As the court noted in its order denying Plaintiffs' motion, this proposed class “effectively include[d] every person who made a claim against CMIC and was paid, from 2008 to now.” (First Class Certification Denial at 6.)

The proposed class of some 400,000 individuals easily satisfied the numerosity requirement. (Id. at 11.) But Plaintiffs utterly failed to prove commonality and typicality. They did not even “attempt to show that any particular conduct by CMIC caused all 400,000 proposed class members to suffer the same injury.” (Id. at 12.) Plaintiffs proposed two alternative class definitions, but the court rejected those, as well, noting that they simply “slic[ed] off a sliver of the original group” without “unit[ing] the remaining proposed class members.” (Id.) The Plaintiffs had ventured “19 ‘common' questions,” alleging myriad different ways in which CMIC was allegedly underestimating or underpaying claims; but Plaintiffs cited “no evidence whatsoever” that the alleged conduct was actually occurring. (Id. at 13.)

As to the contents claims, Plaintiffs claimed that CMIC underestimated the ACV by assuming “average” condition when entering items into Xactimate. (Id. at 14.) But Plaintiffs' expert Donald Stafford, on whose opinion Plaintiffs relied in proving their contents claims, failed a Daubert challenge, having relied on what appeared to be pure speculation in determining the extent to which CMIC's assumptions resulted in underestimation of claims on a class-wide basis. (Id. at 15-21.) Stafford's calculations relied on claimants' self-reporting of personal property's condition; he had never adequately accounted for the possibility that, even if CMIC might have been incentivized to under-estimate items' condition, insureds would be equally incentivized to inflate their property's condition when self-reporting. (Id. at 18.) In any event, Stafford's classwide estimates of damages was not relevant at the class-certification stage, since they concerned damages resulting from alleged conduct Plaintiffs had not proven was common to the class. (Id. at 21.)

The court also rejected Plaintiffs' attempt to use a table which purported to show that eighty percent of insureds in the sample set with RC coverage never recovered the depreciation withheld from CMIC's ACV payments. (Id. at 21.) Plaintiffs had “not established that ACV payments are too low” and had failed to offer a foundation for admission of the table under either Federal Rule of Evidence 901 or 1006. (Id. at 22-23.)

As to structure claims, the court noted that Plaintiffs had not “include[d] any evidentiary citation for the vast majority of the allegations in their structure section.” (Id. at 23.) And the evidence Plaintiffs did cite-extrapolation from single insurance claims-was “grossly insufficient to support class treatment.” (Id.) Plaintiffs also cited “certain CMIC materials” describing claims practices-such as allegedly instructing adjusters to not use so-called “Eagleview” reports to assess roof damage-as evidence of underpayment. But such evidence, the court pointed out, “d[id] not show that CMIC has a common practice of underpaying roof claims” without evidence that CMIC had failed to use other means to assess roof damage. (Id. at 24.) The court also stressed the absence of proof that “CMIC's initial estimates were too low.” (Id.) Finally, the court excluded testimony from Plaintiffs' expert Russell Thomas for reasons similar to those that doomed Stafford's report: Thomas did not use reliable methodology in calculating average underpayment of structure claims, and his report mainly calculated damages without proving commonality-that is, “whether (and how) common conduct by CMIC results in breach of the insurance contracts.” (Id. at 25.)

Having denied class certification for these reasons, the court declined to address adequacy or predominance at any length, but pointed out that, as Plaintiffs failed to establish commonality, “there is no possibility that common questions predominate.” (Id. at 29.)

C. Second Class Certification Attempt

In their renewed motion for class certification, Plaintiffs now propose to define the class as follows:

All individuals or entities who submitted a claim under a homeowners, condominium, or renters insurance policy issued by Country Mutual Insurance Company; the date of loss is between January 12, 2008, and the present; Country Mutual Insurance Company accepted and paid the claim; and Country Mutual Insurance Company's employees or agents prepared an estimate using Xactimate and/or XactContents in connection with estimating the claim. Excluded from the class are any claims where: (1) only personal property was involved and no depreciation was withheld from the contents; and/or (2) the claim file contains a full release of all claims.

(Redacted Second Class Certification Mem. at 32.) According to Plaintiffs, this class definition, like the last one, constitutes “over 400,000 putative class members.” (Id. at 34.)

Plaintiffs allege, broadly, that CMIC “underpays the average claim by thousands of dollars” through “the use of common tactics that are applied to almost every claim in the proposed class.” (Id. at 11.) In support of this contention, Plaintiffs identify numerous ways in which CMIC purportedly underpays claims, often, according to Plaintiffs, through underestimation of claimants' losses. Plaintiffs propose twenty-four “common questions,” which they label “a” through “x,” most of them identifying estimation tactics that CMIC employs. (Id. at 35-44.) They identify questions “a”-“c” and “o” as supporting the class definition. (Id. at 46.) As best the court can tell, then, Plaintiffs have zeroed in on four policies they claim are universal to everyone in the class and result in underpayment of class members. Plaintiffs' descriptions of those policies are unfortunately larded with editorializing:

a. How much CMIC underpays claims by conducting superficial inspections that ignore easily discoverable damage on the assumption a contractor will point out the deficiencies, when CMIC's estimates omit necessary steps required to complete repairs such that most contractors will know that CMIC has no interest in

paying the true cost of repair and therefore not waste their time trying to get CMIC to correct its estimate;

b. Whether CMIC underpays claims, and by how much, when CMIC instructs its adjusters that the estimates written according to CMIC's standards ‘will reflect the proper Actual Cash Value and replacement cost,' and to reject contractor proposals that do not match CMIC's estimate, where CMIC's own ‘expert' testified that the cost of repair is ‘[w]hatever the final cost that is a bargain between the insured and the contractor,' and the developer of the software CMIC's [sic] uses to write its estimates explicitly states that its price data is just an estimate and must be verified;

c. How much CMIC underpays claims through CMIC's use of unit prices that are lower than those provided by Xactimate, that CMIC lowered without evidence to support modification of the unit price provided by Xactimate's developer of the estimating software, and CMIC's insistence that it will only pay those prices when CMIC not only lacks evidence justifying price reductions; but has evidence that the software's pricing database's prices are below the amount being charged by contractors; [and]

o. Whether CMIC underpays contents claims by only considering the age of an item, and none of the other five factors listed in its insurance policies (wear and tear, deterioration, obsolescence, physical condition and reduced market value of the property), and then applying a depreciation factor that is based on assumptions regarding average use for an entire category of items (e.g., CMIC assumes that the life expectancy of a t-shirt is the same as a dress shirt as both are depreciated as “All Other Clothing”), and the amount of CMIC's underpayment from this estimating tactic.

(Id. at 47, 50, 56, 76.)

Most of Plaintiffs' twenty other “common questions”-questions “d”-“n” and “p”-“q”-deal with CMIC's practices for estimating certain types of covered loss: roofing (“d”-“f'); shingle and carpet “waste” calculations (“g”); various kinds of repair techniques (“h”); protection or coverage for undamaged property impacted by repair efforts (“i”-“k”); “general contractor overhead” for insureds who take control of repairs (“l”); and various other practices regarding structure claims (“m”-“q”). (See id. at 37-42.) Beyond estimation tactics, question “r” asks “the appropriate types and measures of damages” in the case, and questions “s”-“x” each concern the application of Illinois' vexatious and unreasonable delay statute, 215 ILCS 5/155. (See id. at 42-44.) Without recounting each question in detail, the court pauses to highlight a few that are relevant to resolution of Plaintiffs' motion.

Plaintiffs' question “d” asks “whether CMIC underpays claims by using the general demolition labor rate instead of the higher roofer rate [when estimating the cost of] roof removals.” (Redacted Second Class Certification Mem. at 37.) According to Plaintiffs, the Xactimate data produced during discovery reveals that CMIC used the “general demolition” labor rate-and not the higher roofer rate-in “100% of shingle roof claims” and, all told, 97% of roof claims.¹³(Sealed Second Class Certification Mem. at 37 n. 151 (citing Summ. Chart of Roof Claims, Ex. AN to Sealed Second Class Certification Mem. (hereinafter “Roof Claims Summary”) [326-27]; for the 97% figure, see Sealed Second Class Certification Mem. at 57.) In his deposition, Defendant's expert LaPrade explained that “primarily worker's compensation and other insurance and other fees” account for Xactimate's estimating higher costs for roof-removal labor, since “[a]ccessing a roof does have risk associated with it.” (LaPrade Dep. at 236:13-237:8.) Plaintiffs point to this admission as evidence that CMIC's use of the general demolition labor rate would always underestimate the labor costs associated with roof repair. (Redacted Second Class Certification Mem. at 37 n.151.)

13 The court, looking at Exhibit AN (which Plaintiffs cite for this proposition), is unable to corroborate Plaintiffs' percentages. The summary at the beginning of the exhibit does not appear to state them, and the court is not inclined to sift through the enormous spreadsheet on the exhibit's other pages. Because it proves irrelevant to class certification, the court assumes Plaintiffs' figures are accurate.

Plaintiffs' question “e” asks “how much CMIC underpays claims by failing to include steep and/or height surcharges on roof claims involving steep or higher roofs, and even where CMIC includes these surcharges CMIC uses the wrong labor rate for the removal steep and height surcharges.” (Id. at 37.) As evidence of this, Plaintiffs point to their own analysis of Xactimate data, which shows that “CMIC's adjusters failed to include steep and/or height surcharges on at least 15% of roof claims in the sample set.” (Sealed Second Class Certification Mem. at 37 n. 152 (citing Roof Claims Summary).) By Plaintiffs' own count, then, this proposed common question applies only to a subset of all policyholders who had roofs repaired, and, at that, only 15% of that smaller group.

Plaintiffs' question “g” asks, in part, “whether CMIC underpays claims by not allowing for sufficient scrap/waste material based on industry standards [in its estimates] (e.g., only allowing for five perfect [sic] shingle waste when industry standards are to allow for 10-15% waste depending on roof type[)].”¹⁴ (Redacted Second Class Certification Mem. at 38.) Plaintiffs point to audio files from CMIC adjuster training videos saying things like “For my shingles I do five percent waste, that's the way we're going as a company.” (Adjuster Training Video, Ex. BG to Sealed Second Class Certification Mem. [326-41] at 2:44-2:50.) According to Plaintiffs' summary of its analysis of Xactimate data, CMIC adjusters estimated shingle waste at five percent somewhere between “89.29%” and 94.05% of the time. (See Summ. Table of Shingle Waste Calculations, Ex. BG to Sealed Second Class Certification Mem. (hereinafter “Shingle Waste Table”) [326-42] at 4.)

14 This proposed question also includes allegations related to carpeting estimates, which the court ignores for reasons discussed more fully below.

Thus, by Plaintiffs' own report, these myriad “common questions” are not common to the proposed class as a whole. Yet Plaintiffs firmly resist the possibility of certification of a narrower class:

Contrary to the Court's suggestion, Plaintiffs are not ‘hoping the court will certify a narrower class based on one or more of the 19 ‘common questions listed in their motion.' Plaintiffs included many of these common questions to show that there are common cheats being employed by CMIC (i.e., that there the class is more cohesive than in Wal-Mart where the only commonality was putative class members were women who worked at Wal-Mart). Certifying a class based on one or more of these cheats would: result in too narrow of a class; reward CMIC's grossly deficient production [in discovery], result in an underpayment of damages to the class; and be inconsistent with precedent such that CMIC might actually prevail on appeal.

(Redacted Class Certification Mem. at 46 n.187 (citations omitted).)

DISCUSSION

The court discusses the parties' evidentiary disputes first, and then turns to the issue of class certification. At each step, the court limits its discussion to the parties' briefing and those briefs' attached exhibits.

I. Evidentiary Motions

In this Circuit, expert evidence in support of class-certification motions is limited to that which will be admissible at trial. See Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 812-13 (7th Cir. 2012) (where expert testimony was “critical” to class-certification decision, requiring analysis under Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993)). The Seventh Circuit has not explicitly addressed whether other evidentiary rules apply in full force to pre-trial motions, and other circuits are split as to how demanding admissibility standards should be at the class-certification stage. Compare Unger v. Amedisys Inc., 401 F.3d 316, 319 (5th Cir. 2005) (requiring the court's findings be “based on adequate admissible evidence to justify classcertification), with Sali v. Corona Reg. Med. Ctr., 909 F.3d 996, 1004 (9th Cir. 2018) (“Inadmissibility alone is not a proper basis to reject evidence submitted in support of class certification.”). The Seventh Circuit's holding in Messner appears to be consistent with the former approach; this court therefore limits its analysis to the parties' briefs and evidence, admissible under the Federal Rules of Evidence, that is attached as exhibits to those briefs. (See First Class Certification Denial at 11.)

A. The Second Chad Hansen Declaration

Defendant has moved to exclude certain portions of a new declaration from Chad Hansen, submitted in support of Plaintiffs' renewed motion for class certification. (See generally, Hansen Declaration.)¹⁵ Defendant argues, first, that parts of the declaration contradict the Hansens' prior sworn deposition testimony and interrogatory responses; second, that the declaration consists of inadmissible opinion testimony; and third, that portions of the declaration are simply additional legal briefing disguised as testimony. (See Def.'s Mem. in Supp. of Mot. to Exclude Decl. of Chad Hansen [335] at 4, 7, 11.) As explained below, each of these arguments has traction.

15 Specifically, Paragraphs 3, 5, 8, 11-13, 15, 22-24, 26-29, 31-36, 38, 39, 44, 46, and 48. (Def.'s Mem. in Supp. of Mot. to Exclude Dec. of Chad Hansen (hereinafter “Hansen Exclusion Motion”) [335] at 4.)

1. Paragraphs 26 & 27

After the close of discovery, “[i]f a party fails to provide information or identify a witness as required by Rule 26(a) or (e), the party is not allowed to use that information or witness to supply evidence on a motion, at a hearing, or at a trial, unless the failure was substantially justified or is harmless.” FED. R. CIV. P. 37(c). Typically, though a district court may reopen discovery, “[u]ntimely disclosures and discovery responses and supplements to them are generally not substantially justified or harmless.” DR Distributors, LLC v. 21 Century Smoking, Inc., 513 F.Supp.3d 839, 960 (N.D. Ill. Jan. 19, 2021). Further, “a deponent may not use an affidavit sworn to after a deposition to contradict deposition testimony without giving a credible explanation for the discrepancies.” Abraham v. Washington Grp. Intern., Inc., 766 F.3d 735, 741 (7th Cir. 2014); see also Bank of Ill. v. Allied Signal Safety Restraint Sys., 75 F.3d 1162, 1168 (7th Cir. 1996) (rejecting Plaintiff's attempt to do so in the summary judgment context).

Defendant contends that two paragraphs of the Hansen Declaration contradict the record:

¶ 26. With respect to our roof, we paid DNR roughly $9,000 to simply remove the old shingles and put new ones on. We had to pay another contractor to finish the roof by lifting up the shingles DNR put on so that the drip edge could be replaced. CMIC's suggestion that DNR's bid for the roof work was lower than CMIC's estimate is incorrect because as shown above, when DNR was forced to provide specific numbers, DNR's numbers increased significantly.

¶ 27. The amount DNR was ultimately paid is not a reflection on what DNR would perform the repairs for, but what DNR was willing to give up to avoid litigation after being contacted by an attorney on our behalf.

(Hansen Declaration at ¶¶ 26-27.) These paragraphs contradict Plaintiffs' prior deposition testimony, Defendant urges, given that it is the “first time that Plaintiffs have claimed that they paid another roofing contractor [beyond DNR] for roof repairs.” (Hansen Exclusion Motion at 6.)

In support of this argument, Defendant cites deposition testimony from both Hansens. First, Chad and Melissa Hansen each claimed in sworn depositions that DNR repaired their roof, and neither of them mentioned any other contractor's involvement. (Chad Hansen Dep. at 137:919; Def.'s Excerpts: Melissa Hansen Dep. at 55:23-56:10.) Second, in a portion of his deposition concerning the roof replacement, Chad Hansen confirmed that “Melissa is the right person . . . to get detailed responses” as to “what [the Hansens]'re claiming that you've been compensated and more important what you're claiming you haven't been compensated for.” (Chad Hansen Dep. 122:20-123:10.) Then, Melissa Hansen testified that DNR had refunded about $3,000 of the amount the Hansens had paid DNR for roof work. (Def.'s Excerpts: Melissa Hansen Dep. at 78:713.) Finally, CMIC propounded a discovery request asking for “a complete itemization of all expenditures made on the home . . . as a result of the tornado which was the subject of Claim No. 102-0062421, including expenditures made for the structure, personal property, other structures on the property and any other expenditure for the home related to the tornado loss.” In response, the Hansens presented a chart in which they listed “12,600” as “amount paid” for the “[r]oof,” and identified DNR as the contractor. (Pl.'s Supp. Answers to Def. Country Mutual's Interrogs., Ex. C to Hansen Exclusion Motion (hereinafter “Interrogatory Responses”) [335-3] at 7, 13.) They mentioned no other contractor or roof work.

As Defendant contends, the Hansen Declaration thus introduces new and seemingly contradictory evidence after the close of discovery. Apart from this Declaration, there is nothing in the record suggesting that a contractor other than DNR performed additional roof work on Plaintiffs' home. Indeed, the Hansens' affirmative statement in their interrogatory response- listing only DNR as their contractor for roof work resulting from the tornado-and their failure to bring up other contracting work or dissatisfaction with the work DNR did for them is at best a puzzling inconsistency that CMIC has had no opportunity to explore.

Paragraph 27 also departs from the Hansens' prior testimony. Nowhere in the deposition testimony that either side cites is there an indication that DNR had incurred greater costs to repair the roof than what the Hansens paid. Instead, Melissa Hansen's deposition testimony states the opposite:

So in order to get the roof fixed - you know, so DNR performed that work. We, then, submitted documents to our escrow with the contract information, and the next process was to get a lump sum for them to get started.

And so upon the monies being released we, you know, signed our - they had to sign and we had to sign, and so we signed the check. And they got all of that money, but they didn't do all of that work for that amount of money. And so they weren't going to refund it. They were just going to keep it.... And so we had to get an attorney involved because they were going to keep money that wasn't owed to them.

(Def.'s Excerpts: Melissa Hansen Dep. at 71:10-22; 77:12-78:3 (emphasis added).) The record evidence is, thus, that the Hansens paid $12,600 for DNR to repair their roof Interrogatory Responses at 13), and that DNR “got all of that money, but they didn't do all of that work” and thus owed the Hansens a refund. (Def.'s Excerpts: Melissa Hansen Dep. at 77:19-20.) Finally, though both Chad and Melissa Hansen did claim in their depositions that an attorney had gotten involved to secure the refund of unused funds from DNR, the Hansen Declaration is the first time any allegation is made that the $3,000-and-change that DNR refunded the Hansens was “not a reflection of what DNR would perform the repairs for.” (See Hansen Declaration at ¶ 27.)

In reaching its conclusion that the Hansen Declaration introduces new evidence, the court notes that if any corroboration for the declaration's new claims did exist in the voluminous record Plaintiffs attached to their class-certification motion, they could easily have pointed it out. They did not. Instead, Plaintiffs' objection to Defendant's motion to exclude the Hansen Declaration is, in its entirety (and without any citation whatsoever), the following:

CMIC's objections to the Hansen Declaration not only demonstrate CMIC's overreach, but also that CMIC's citations to the record usually do not support CMIC's assertions. CMIC fails to show where Chad Hansen contradicted himself and a review of the entire declaration shows there is support for the assertions CMIC challenges.

(Pl.'s Consolidated Reply Mem. of Law & Fact in Supp. of Pl.'s Am. Mot. for Class Certification and Resp. to CMIC's Evidentiary Mots. and Pl.'s Mot. to Exclude (hereinafter “Pl.'s Consolidated Reply”) [355] at 24.) These unsupported assertions are unhelpful-and the court finds nothing at all in the “entire declaration” that shows “support” for the challenged assertions elsewhere in the record. (See Mem. Op. and Ord. [144] at 2.) Moreover, regardless of the extent to which the Hansen Declaration actively contradicts Chad and Melissa Hansens' prior sworn statements, it does inject new information about a new contractor into the dispute long after discovery has closed-something Federal Rule of Civil Procedure 37(c) prohibits. Nor is this new evidence “substantially justified or harmless.” FED. R. CIV. P. 37(c). It springs upon Defendant information that could be relevant to the certification of a potential subclass of insureds whose roofs were repaired and does so long after the discovery period has ended.¹⁶ Accordingly, the court refuses to consider Paragraphs 26 and 27 in its consideration of Plaintiffs' class-certification motion.

16 In the court's prior denial of Plaintiffs' class-certification motion, it noted that Plaintiffs' allegation concerning CMIC's use of an incorrect roof-removal rate “hints at a common question.” (First Class Certification Denial at 28.) Defendant finds it “suspicious” that these new roof-related facts have suddenly surfaced in Chad Hansen's declaration. (Hansen Exclusion Motion at 6.) As the court will exclude these portions of the Hansen declaration under FED. R. CIV. P. 37(c), it does not reach the question as to whether the declaration is a “sham.” Cf. Bank of Ill. v. Allied Signal Safety Restraint Sys., 75 F.3d 1162, 1168-69 (7th Cir. 1996) (noting that courts may exclude affidavits which contradict prior sworn statements.).

2. Paragraphs 3, 5, 8, 11, 13, 24, 27, 29, 44, 46, & 48

Lay witness testimony is limited to observations “rationally based on the witness's perception” that are “helpful to clearly understanding the witness's testimony or to determining a fact in issue” and are “not based on scientific, technical, or other specialized knowledge within the scope of Rule 702.” FED R. EVID. 701(a)-(c). In ruling on class certification, the court thus ignores statements in the Hansen Declaration that extend beyond Chad Hansen's own perceptions.¹⁷Because Plaintiffs' class-certification motion fails regardless of whether the court considers most of the challenged portions of the Hansen Declaration, it will not directly address admissibility of its specific paragraphs, apart from this one:

17 As an example of such a statement, Hansen claims in Paragraph 48 that “[i]n comparing our claim to the typical claimant who likely does not have a significant contents claim which they can shift money from to complete structural repairs, we realized that the majority of claimants probably would be unable to repair their homes with the lowball estimates CMIC was preparing.” (Hansen Declaration ¶ 48.) Nowhere does Chad Hansen explain how he knows the characteristics or behavior of the “typical claimant,” let alone his basis for asserting that CMIC provides “lowball estimates” to claimants other than the Hansens themselves.

¶ 24. Since filing this litigation we received another estimate that was prepared off of the documentation in CMIC's claim file. That estimate shows that properly estimated, the true cost of repair for our house using the Xactimate price list for June 2015 and bids to the extent we were capable of obtaining bids was $304,592.80 RCV or $263,110.46 ACV. As such, regardless of how CMIC wishes to categorize its payments to us, we have not been paid profit or even the minimum owed (ACV) under the insurance policy.

(Hansen Declaration ¶ 24.) Defendant argues that “Chad Hansen was never disclosed as an expert in this case” and questions how he could have “the requisite knowledge, skill, experience, training, or education to render an expert opinion about depreciation and the ‘true cost of repair' of Plaintiffs' house.” (Hansen Exclusion Motion at 11.) CMIC notes, further, that Plaintiffs never explain “the contents of the estimate . . . who created it, and when it was created.” (Id. at 10 n.3.) Second, Defendant suggests that this new estimate either should have been produced during discovery or, if Plaintiffs sought it after discovery closed, implies that it was “procured only for purposes of bolstering class certification.” (Id.)

The court agrees with Defendant on this score. Plaintiffs cannot incorporate an expert's estimates of their insurance claims simply by reference in a lay declaration and without any documentation of the methodology used or qualifications (or even the identity) of the person who made the estimate. This is especially true given that the paragraph asserts that this new estimate “shows” underpayment by CMIC. (Hansen Declaration ¶ 24.) That claim is not “rationally based on” Chad Hansen's own perceptions as required by Federal Rule of Evidence 701(a). The court would add that, without hearing from its unidentified creator, the referenced estimate is inadmissible hearsay as well. See FED. R. EVID. 801(c). Accordingly, the court does not rely on it in assessing Plaintiffs' class-certification motion.

3. Paragraphs 12, 15, 22, 23, 28, 29, 31-36, 38, & 39

In its first denial of Plaintiffs' class-certification motion, the court ignored legal arguments that Plaintiffs injected in exhibits, stressing that “labeling a brief an exhibit does not change the essential nature of the thing.” (First Class Certification Denial at 11 (quoting Slaven v. Great Am. Ins. Co., 83 F.Supp.3d 789, 803 (N.D. Ill. 2015).) For this new motion, the court granted Plaintiffs leave to file a brief of 103 pages, far in excess of the court's 15-page standard. (See Mot. for Leave to File Pl.'s Mem. in Supp. of Pl.'s Am. Mot. for Class Certification and Certain Exhibits Thereto Under Seal, Allow for Mem. in Excess of 15 Pages and Set Briefing Sched. & id. Ex. 1 [321, 321-1]; Minute Order [324].) The Hansen Declaration, too, contains numerous paragraphs that can only be understood as legal argument; as before, the court simply ignores this further briefing that it did not permit.

B. Exhibits 10 and 12 to CMIC's Response

In its response to Plaintiffs' class-certification motion, Defendant appended two exhibits which Plaintiffs seek to exclude. First, Exhibit 10 to Defendant's response memorandum purports to be the Wells Fargo Property Loss Department's documentation of requests Plaintiffs made while contracting with DNR for money “to be released” to them by Wells Fargo, who held the funds in escrow. (See Redacted Opp'n to Am. Class Certification at 9; Def.'s Excerpts: Melissa Hansen Dep. at 77:12-16.) The exhibit includes correspondence between Plaintiffs and Wells Fargo, as well as documents concerning the contracting agreement between Plaintiffs and DNR which was financed by Wells Fargo. Defendant refers to this document to support its claims concerning the amount Plaintiffs paid DNR to repair their roof, and the amount for which DNR had contracted to repair the Plaintiffs' home before they withdrew. (See Opp'n to Am. Class Certification at 19 nn.40-46.) Second, Exhibit 12 to Defendant's response memorandum contains hundreds of pages of excerpts from the 313-claim-file sample set produced in this case, portions of which Defendant uses to dispute Plaintiffs' conclusions as to whether CMIC underpaid certain claims as compared with contractors' bids. (See id. at 28 nn. 95-96; 34-42.) The court discusses each in turn.

1. Exhibit 10: Wells Fargo Documents

Plaintiffs' sole objection to Exhibit 10 is that “CMIC did not file any declaration asserting that the documents contained in Exhibit[] 10 . . . are authentic.” (Sealed Pl.'s Mot. to Exclude CMIC's Exhibits 10 & 12 & the Summs. Based on CMIC's Exhibit 12 (hereinafter “Motion to Exclude CMIC's Exhibits”) [341] at 11.) The court disagrees. Deposition testimony paired with the documents' appearance and contents establish Exhibit 10's authenticity to the court's satisfaction.

A party seeking to introduce evidence must “produce evidence sufficient to support a finding that the item is what the proponent claims it is.” FED. R. EVID. 901(a). This requirement is one of “relevancy”-that is, the evidence is only relevant if it is authentically what its proponent claims it is. Id. advisory committee's note; see also Rule 104(a). No formula is required to authenticate evidence, but Rule 901 offers multiple acceptable means to do so, including the “[t]estimony of a [w]itness with [k]nowledge” and “[t]he appearance, contents, substance, internal patterns, or other distinctive characteristics of the item, taken together with all the circumstances.” Id. Rule 901(b)(1), -(4).

Exhibit 10 begins with a letter from Chad and Melissa Hansen requesting that Wells Fargo, the escrow agent for their insurance payments, disburse “$12,600 to start work” repairing the “outside of [their] home.” (Wells Fargo Docs at 3.) Subsequent pages of the exhibit are a “Conditional Waiver of Lien” signed by both parties (id. at 5); a “Request for Taxpayer Identification Number and Certification” signed by DNR's owner (see id. at 6-7); a signed “Proposal” for “remov[al] and dispos[al] of existing roofing material . . . install[ation of] all roofing material to previous condition . . . remov[al] of siding . . . dispos[al] of all soffit/fascia/gutter,” and more, for $25,200 with “50% due upon starting” (id. at 8-9); and an additional conditional waiver of lien and signed proposal to do further work on the home for $164,091.98. (Id. at 10-13.)

Other parts of the record corroborate this exhibit's contents such that the court can fairly say that “[t]aken together . . . all the circumstances” prove its authenticity. FED. R. EVID. 901(b)(4). First, in her deposition, Melissa Hansen states:

So in order to get the roof fixed - you know, so DNR performed that work. We, then, submitted documents to our escrow with the contract information, and the next process was to get a lump sum for them to get started.

(Def.'s Excerpts: Melissa Hansen Dep. at 77:12-16 (emphasis added).) Sworn testimony thus supports the fact that the Hansens “submitted documents . . . with contract information” related to DNR's repairing the roof in order to “get a lump sum for them to get started” (id.)-details consistent with Exhibit 10. (See Wells Fargo Docs at 3.) Moreover, the amount in question- “25,200 . . . [with] 50% due upon starting”-is also consistent with] Plaintiffs' interrogatory response, in which they declare that they paid “$12,600” to “DNR” for roof repair. (Interrogatory Responses at 13.) Finally, the documents' appearance-the Wells Fargo letterhead and formatting-militates in favor of its authentication. See U.S. v. Hoag, 823 F.2d 1123, 1127 (7th Cir. 1987) (noting the use of “company letterhead” and details in the letters “pertain[ing] to specific real estate transactions in question” amounted to sufficient authentication under Rule 901), abrogated on other grounds by U.S. v. Staniforth, 971 F.2d 1355 (7th Cir. 1992).

Exhibit 10 thus contains correspondence with Wells Fargo whose existence Melissa Hansen's deposition testimony references and whose details both her testimony and the Hansens' interrogatory responses corroborate. Nor do the Hansens ever deny having made these contacts with Wells Fargo. Accordingly, Rule 901 provides no basis for its exclusion.

2. Exhibit 12: Excerpted Claims Files from the Sample Set

Plaintiffs make three primary objections to Exhibit 12 of Defendant's response, which includes documents from the 313-claim sample set CMIC provided during discovery: first, that the documents contain inadmissible hearsay; second, that Defendant's discussion of certain claims in its response briefing violates Federal Rule of Evidence 1006 and constitutes inappropriate expert opinions from CMIC's lawyers; and third, that the documents in the exhibit are not properly authenticated under Federal Rule of Evidence 901.

a. Hearsay Objections

Hearsay is an out-of-court statement introduced “to prove the truth of the matter asserted in the statement.” FED. R. EVID. 801(c). It is inadmissible absent an applicable exception. Id. Rule 802.¹⁸Defendant quotes language from the claim files in Exhibit 12 to argue that “repairs were, in fact, being made for the amounts paid by Country Mutual.” (See Sealed Opp'n. to Am. Class Certification at 28, 28 n.96.)

18 Courts in this district have differed as to whether to apply the hearsay rule to evidence submitted in class-certification motions. Compare Gomez v. PNC Bank, Nat'l Assoc., 306 F.R.D. 156, 159 (N.D. Ill. July 24, 2014) (applying hearsay rules to evidence at classcertification stage); with Smith v. City of Chicago, 340 F.R.D. 262, 273 (N.D. Ill. Aug. 31, 2021) (opting to “follow the approach” of other circuits “allowing hearsay and other inadmissible evidence to be considered for class certification purposes”). As explained earlier, this court adopts the former approach, but because it overrules Plaintiffs' objections, any difference in approach is immaterial.

The files contain written notes about the claims, recorded by CMIC's claims adjusters, and in addition, notes about statements reportedly made to claims adjusters on the phone. Both are hearsay. As one example, an adjuster reports in a claim file that “Victoria [the policyholder] called in [and] stated that repairs are complete and needs balance paid out on claim, wants to know what needs to be sent in.” (Id. at 28 n.96.) If introduced to prove that Victoria's repairs were completed, this statement is double hearsay, as both statements are being introduced for their truth: that is, that Victoria did in fact say that repairs had been done (as the claims adjuster records in the file); and that the repairs were in fact done (the truth of Victoria's statement to the adjuster). When evidence contains double hearsay, both layers must fall within applicable hearsay exceptions for the statement(s) to come in. See, e.g., Flanagan v. Office of Chief Judge of Circuit Court of Cook County, Illinois, 893 F.3d 372, 374-75 (7th Cir. 2018). Here, it is debatable whether either layer falls comfortably within an exception.¹⁹

19 The first layer-each adjuster's written memorialization contained in the claim file-could fall within the business records exception, but it appears as though CMIC has not offered the affidavit of any custodian testifying to the process for creating and maintaining these records, as required by FED. R. EVID. 803(6)(D). The second layer-Victoria's statement to the adjuster-might be construed as, in part, a request for payment (that is, a “verbal act,”) rather than a declaration of fact. But if offered to establish that repairs were in fact made, that statement appears to fall within no exception.

The court nevertheless concludes that Exhibit 12 is admissible under the rule of completeness. Federal Rule of Evidence 106 provides that:

If a party introduces all or part of a writing or recorded statement, an adverse party may require the introduction, at that time, of any other part - or any other writing or recorded statement - that in fairness ought to be considered at the same time.

FED. R. EVID. 106. The Rule's purpose is to enable a party to “qualif[y] or explain[] the evidence offered by the opponent.” United States v. Velasco, 953 F.2d 1467, 1475 (7th Cir. 1992). In making their own case, Plaintiffs rely on the same claim files whose admissibility they now challenge. (See, e.g., Supp. Docs. to Ex. BA to Sealed Second Class Certification Motion [32635].)²⁰ Indeed, Defendant explains Exhibit 12's purpose as contextualizing the Plaintiffs' evidence: “tak[ing] from Plaintiffs' Exhibit BA and using Plaintiffs' numbers . . . a review of these same underlying claim files shows evidence in each that repairs were, in fact, being made for the amounts paid by Country Mutual.” (Sealed Opp'n. to Am. Class Certification at 27-28.) In other words, CMIC highlights certain hearsay statements that Plaintiffs have already submitted in order to “qualif[y] or explain[]” the different conclusion Defendant draws from those statements. Velasco, 953 F.2d at 1475. The court concludes that if Plaintiffs may use the sample claim files to establish that CMIC consistently underpays insureds, then CMIC is entitled to use the same sample to rebut that showing.

20 The court recognizes that if offered by Plaintiffs, many-but still not all-of the claims files are likely admissible as statements of a party opponent. FED. R. EVID. 801(D)(2). But the inner layers of hearsay contained in those files-Veronica's statement, for example-remain inadmissible because they are statements of the insureds, not admissions of the Defendant or its representatives. Regardless, the admissibility of each statement in these hundreds of pages of claim files proves not to be decisive of the class-certification question.

b. Rule 1006 Objections

Plaintiffs next object to numerous portions of Defendant's opposition brief, claiming that they “easily fall within Federal Rule of Evidence 1006” and contain new “expert analysis” after discovery has closed. (Motion to Exclude CMIC's Exhibits at 9.)²¹

21 Specifically, Plaintiffs seek to exclude: footnotes “95-96, 112-113, 115-123, 125126, 138, 140, [and] 142-146” of Defendant's opposition brief, as well as “the text in the body associated with those footnotes” and “the summaries of Ex. 12 that occur on pages 25-28 & 3132” of the same. (Id.)

Federal Rule of Evidence 1006 permits a party to “use a summary, chart, or calculation to prove the content of voluminous writings . . . that cannot be conveniently examined in court.” FED. R. EVID. 1006. Here, what Plaintiffs challenge is the argument appearing in Defendant's opposition brief that there are “many . . . examples” of policyholders getting repairs done on claims in the sample set. (Sealed Opp'n. to Am. Class Certification at 27-28.) In support of this contention, CMIC has cited individual pages in the claim files, along with quotes, to provide specific corroborative examples. (See, e.g., id. n.96.) Later, Defendant draws conclusions about individual claims in the claim-file sample set, in each instance citing to the underlying data. (See id. at 34-38.) This is unlike Plaintiffs' summary exhibit that this court excluded in its earlier class certification ruling. There, Plaintiffs cited no specific data set, let alone individual pages of the record, as comprising the underlying data behind their summary exhibit. (First Class Certification Denial at 22-23.) Here, contrarily, Defendant cites to specific pages in the record to support each conclusion it draws. Because the Defendant cites the underlying data, Plaintiffs' objection under Rule 1006 fails. See United States v. White, 737 F.3d 1121, 1135 (7th Cir. 2013) (noting that summaries must point to the data on which they are based).

That said, Defendant's conclusions drawn from the claim files are not without problems. Defendant cites to multiple specific pages of the claim file to support its conclusions comparing contractors' bids to eventual payment, but those data points are often difficult to decipher without relying on Defendant's own (potentially biased) characterization of them. For instance, Defendant asserts that for “Claim No. 163-0035064, the difference [between the contractor's bid and CMIC's payment] was $324.60, which was an owner upgrade not covered under the policy.” (Sealed Opp'n. to Am. Class Certification at 36.) To support this, Defendant cites numerous pages from the claim file purporting to record the amount paid, the deduction of the policyholder's deductible, and the two “invoice/bid amounts.” (Id. at 36 n.119.) The cited pages appear to be at least somewhat consistent with what the Defendant describes: the two contract invoices, for example, jump from $8,874 to $9,180, accounting for a change from “3-tab [s]hingles” to “30-year [a]rchitectural [s]hingles.” (See Exhibit 12 Part I at 146, 150 (showing jump to account for this shingle change).) But Defendant fails to point to any explicit evidence that the change in shingle type to “30-year [a]rchitectural [s]hingles” was an upgrade under CMIC's policy. (See id. at 150.)²²Moreover, that jump is only $306 by the court's math, not $324.60, and a way of getting to Defendant's slightly larger figure is not immediately apparent. The court could therefore consider certain of Defendant's conclusions only if supported by more explicit citation or explicit expert analysis. See FED. R. EVID. 701 (limiting lay testimony to that “rationally based on the witness's perception”). Accordingly, the court limits its use of Exhibit 12-and the Defendant's briefing around it-to conclusions that can be drawn plainly from viewing the pages Defendant cites, without reliance on Defendant's characterization or summary.²³

22 In fact, as the court reads the documents, the contractor appears to write: “FOUND THE ROOF TO BE COVERED BY ARCHITECT STYLE SHINGLES . . . WE HAVE ATTACHED OUR ESTIMATE TO DO THE REPAIR.” (See id. at 147). It is unclear whether the architectstyle shingles the contractor claims were already covering the insured's roof were the same as the 30-year architectural shingles with which the contractor replaced them. If so, then presumably the shingles would have been covered under CMIC's RC policy. (See Hansen Insurance Policy at 2.)

23 In a separate motion, Plaintiffs suggest that Defendant's expert John LaPrade was involved in curating and summarizing the Exhibit 12 data as presented in Defendant's opposition brief. (See Pl.'s Mot. to Reinstate Pl.'s Mot. to Exclude the Op. Test. of John D. LaPrade &

c. Authentication Objections

The Plaintiffs' final objection to Exhibit 12 is that it is not authenticated. The court overrules this objection for reasons similar to its ruling on Exhibit 10. The sample set from which Exhibit 12 draws mirrors documents Plaintiffs append to their own motion. (See Supp. Docs. to Ex. BA to Sealed Second Class Certification Motion [326-35].) The court is satisfied as to both parties' use of the sample set that exhibited excerpts meet the authentication requirements of Federal Rule of Evidence 901.

C. The LaPrade Report

Finally, Plaintiffs have renewed their motion to exclude expert reports by Defendant's two experts, John LaPrade and Stephen Prowse. (See Renewed LaPrade Motion.) As in its prior order, the court cites to the LaPrade report for various factual details it otherwise could not find in the record; to rule on this motion, the court does not rely on LaPrade's broader opinions concerning CMIC's claims handling or payment, and thus denies Plaintiffs' renewed motion as moot. (See First Class Certification Denial at 2.)

II. Plaintiffs' Amended Class Certification Motion

Federal Rule of Civil Procedure 23(a) provides four requirements for a proposed class to be certified: “(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.” FED. R. CIV. P. 23(a). If a proposed class meets these requirements, it must also meet requirements imposed by Rule 23(b). Plaintiff here ask the court to certify a damages class under Rule 23(b)(3), and therefore must prove “that the

Stephen Prowse (hereinafter “Renewed LaPrade Motion”) [342] at 2; see also Screenshot of Ex. 12 with Annotations Pane Displayed, Ex. A to Renewed LaPrade Motion (seeming to show LaPrade as having “[h]ighlighted” text in Exhibit 12).) The court limits its use of Exhibit 12 to conclusions it can draw without relying on Defendant's characterization of them, and therefore need not address this objection. questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” FED. R. CIV. P. 23(b)(3).

“The party seeking certification bears the burden of demonstrating that certification is proper by a preponderance of the evidence.” Bell v. PNC Bank, Nat'l Ass'n, 800 F.3d 360, 373 (7th Cir. 2015) (citing Messner v. Northshore Univ. Healthsystem, 669 F.3d 802, 811 (7th Cir. 2012)). Plaintiffs thus must introduce “evidentiary proof” to meet the requirements of Rule 23. Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). Because “[t]he decision to certify a class or not can cause a considerable tilt in the playing fields of litigation . . . [it] is not one to take lightly.” Chi. Teachers Union, Local No. 1 v. Bd. of Educ. of City of Chi., 797 F.3d 426, 433 (7th Cir. 2015). The court must therefore conduct a “rigorous analysis” before certifying a class. Howard v. Cook Cnty. Sheriff's Off., 989 F.3d 587, 598 (7th Cir. 2021).²⁴ Such an analysis will “[f]requently . . . entail some overlap with the merits of the plaintiff's underlying claim.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 351 (2011).

24 Plaintiffs urge the court to “resolve any doubt regarding the propriety of certification in favor of allowing a class action so that class actions will remain an effective vehicle for deterring corporate wrongdoing.” (Redacted Second Class Certification Mem. at 31.) In support of this proposition, Plaintiffs quote older cases stating that in “doubtful case[s] . . . any error, if there is to be one, should be committed in favor of allowing the class action.” (Id. at 31 n.119 (quoting Esplin v. Hirschi, 402 F.2d 94, 101 (10th Cir. 1968), cert. denied, 394 U.S. 928 (1969)); see also Redacted Second Class Certification Mem. at 31 n.120 (for the same proposition, citing King v. Kansas City Southern Industries, 519 F.2d 20, 25-26 (7th Cir. 1975).) Whether these cases are reconcilable with more recent jurisprudence, including Wal-Mart Stores, Inc. v. Dukes, is debatable. It is nonetheless immaterial here because the court has no doubt as to the impropriety of certifying the class Plaintiffs propose.

A. Numerosity

A proposed class must be “so numerous that joinder of all members is impracticable.” FED. R. CIV. P. 23(a)(1). Plaintiffs' new class definition is, fundamentally, the same as their last: nearly every person insured by CMIC, from January 2008 onward, who “submitted a claim” and for whom CMIC “prepared an estimate using Xactimate and/or XactContents.” (Redacted Second Class Certification Mem. at 32.) This class definition would include “over 400,000 putative class members” (id. at 34), and easily meets the numerosity threshold. See Orr v. Shicker, 953 F.3d 490, 498 (7th Cir. 2021) (a class of 40 members is “often regarded as sufficient[ly]” numerous).

B. Commonality

Rule 23(a)(2) demands that Plaintiffs show there are “questions of law or fact common to the class.” FED. R. CIV. P. 23(a)(2). The commonality requirement “is easy to misinterpret, as ‘[a]ny competently crafted class complaint literally raises common ‘questions.'” Howard, 989 F.3d at 598 (quoting Wal-Mart Stores, 564 U.S. at 349). Although a “single common question of law or fact” can satisfy the commonality requirement, “the mere occurrence of all plaintiffs suffering as a result of a violation of the same provision of law is not enough.” Bell, 800 F.3d at 374. Rather, Plaintiffs “must show that the same conduct or practice by the same defendant gives rise to the same kind of claims from all class members.” McCaster v. Daren Rests., Inc., 845 F.3d 794, 800 (7th Cir. 2017). The common question thus must be such 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.” Ross v. Gossett, 33 F.4th 433, 437 (7th Cir. 2022) (quoting Wal-Mart Stores, 564 U.S. at 350).

In their renewed motion for class certification, Plaintiffs again fail to unite the class with a common question. Though Plaintiffs claim to have shown the existence of over twenty “common questions,” each of varying degrees of generality or applicability to the 400,000-plus person class, their allegations and proffered proof fail Rule 23(a)(2) at each level.

At the broadest level, Plaintiffs allege that CMIC, through “the use of common tactics that are applied to almost every claim in the proposed class . . . underpays the average claim by thousands of dollars.” (Redacted Second Class Certification Mem. at 11.) This general allegation alone, of course, fails to prove commonality. As this court noted in its first denial of class certification, a showing that CMIC routinely underpays insurance claims would still fail to satisfy Rule 23(a)(2)'s commonality requirement unless those underpayments could be tied to a specific class-wide practice. In other words, the violation of thousands of contracts-even if proven- cannot justify class treatment unless the cause for each breach is universal to the class and can be determined in one fell swoop. See Wal-Mart Stores, 564 U.S. at 352 (stressing that commonality requires “some glue holding the alleged reasons” for a widely occurring legal wrong “together . . . [such] that examination of all the class members' claims for relief will produce a common answer”).

Perhaps attempting to identify that glue, Plaintiffs contend that CMIC's alleged broad underpayment of insurance contracts is the product of a series of specific practices, mostly involving CMIC's methods for estimating claimants' losses. Plaintiffs claim to have “based the class definition on common questions ‘a'-'c' and ‘o'” of their amended class-certification brief. (Redacted Second Class Certification Mem. at 46.) The court therefore understands Plaintiffs to argue that those four questions establish conduct applicable to the entire class they propose. For the reasons explained below, however, Plaintiffs have not shown that any of the alleged conduct applies class-wide, let alone how proof of any class-wide conduct would resolve any element of a claim against CMIC.

Before addressing each proffered question, the court pauses to consider the legal claims underlying the parties' class-certification dispute. When determining whether a class should be certified, the court “should begin by identifying the elements of the plaintiff's various claims: ‘only by properly circumscribing the claims and breaking them down into their constituent elements can a district court decide which issues are common, individual, and predominant.'” Simpson v. Dart, 23 F.4th 706, 713-14 (7th Cir. 2022) (quoting Santiago v. City of Chicago, 19 F.4th 1010, 1018 (7th Cir. 2021)). Plaintiffs' claims encompass both breach of contract and unreasonable and vexatious practices in violation of the Illinois Insurance Code-specifically, 215 ILCS 5/155. (Redacted Second Class Certification Mem. at 18.) As to the breach-of-contract claim, in Illinois, “general rules of contract interpretation apply to the interpretation of insurance policies.” Patrick Schaumburg Autos., Inc. v. Hanover Ins. Co., 452 F.Supp.2d 857, 867 (N.D. Ill. Sept. 28, 2006).²⁵Insurance claims typically encompass “six broad elements”: that the policy was “in effect as to the time of the loss”; that the “loss falls within the coverage terms of the policy”; that the “amount of covered loss is the amount claimed”; that the claimant is the correct recipient of any proceeds from the policy; that the “insurer's obligation to pay has matured”; and that the insurer has “breached that obligation by denying the claim.” Id. at 867-68. Additionally, as Plaintiffs point out, the Illinois Administrative Code requires that an insurer's estimate of loss must

25 Plaintiffs' class includes people in nineteen different states. (Redacted Second Class Certification Mem. at 11.) Because even an Illinois-only class fails to pass muster under Rules 23(a)(2) and 23(b)(3), the court does not venture into discussing other states' insurance laws.

[b]e reasonable . . . and of an amount which will allow for repairs to be made in a workmanlike manner. If the insured subsequently claims, based upon a written estimate which he obtains, that necessary repairs will exceed the written estimate prepared by or for the company, the company shall review and respond promptly ....

(Id. at 47-48 (quoting ILL. ADMIN. CODE tit. 50, § 919.80(d)(7)(c)).)

CMIC's policy language entitles its insureds to recover the cost “actually and necessarily incurred to repair or replace” their property, with or without depreciation depending on the level of coverage; the policy language does not entitle them to any specific means of estimating the loss. (See Hansen Insurance Policy at 6; see also Kartman v. State Farm Mut. Auto. Ins. Co., 634 F.3d 883, 890 (7th Cir. 2011) (“[t]he essence of an insurance policy is a promise by the insurer to compensate the insured for the loss . . . not a covenant to use a particular standard for evaluating property damage”).) Plaintiffs, for their part, agree that “[p]recedent provides that claimants are entitled to the correct amount and not a specific method of estimating.” (Redacted Second Class Certification Mem. at 46.)

As Defendant sees things, this principle means that no estimation practice can form the basis of an insurer's liability. (See Redacted Opp'n to Am. Class Certification at 24 (“Importantly, on the issue of breach the question is whether the claim was underpaid, not whether the claim was underestimated.”).) This is an overstatement; though policyholders are not entitled to a specific estimation method, an insurer's estimation practice could breach policyholders' contracts if the practice itself is unreasonable or otherwise illegal. For example, in Stuart v. State Farm Fire and Casualty Co., the Eighth Circuit upheld the certification of a class based solely on State Farm's method of “depreciat[ing] the cost of labor” when calculating Arkansas policyholders' ACVs, because the Arkansas Supreme Court had explicitly outlawed the practice. 910 F.3d 371, 375 (8th Cir. 2018) (citing Adams v. Cameron Mut. Ins. Co., 430 S.W.3d 675, 679 (Ark. 2013)).²⁶

26 Accordingly, the court found that resolution of the question of whether State Farm deducted labor from ACV would be essential to proving liability for each class-member's claim and thus “[i]t was not an abuse of discretion for the district court to conclude that plaintiffs' claims share[d] a common, predominating question of law.” Stuart, 901 F.3d at 375. Arkansas eventually amended the statute to allow insurance companies to deduct for depreciation of labor. See Ark. Code Ann. § 23-88-106(a)(2) (2017).

Because Plaintiffs challenge so many of CMIC's estimating methods, the court understands Plaintiffs to be arguing that all of these methods violate the “reasonable expectations of CMIC's claimants,” or are otherwise somehow independently illegal. (See Pl.'s Consolidated Reply at 16 n.26.) This “reasonableness” inquiry is of course more nebulous than the clear statutory violation at issue in Stuart; in other words, an estimation practice could be reasonable in certain policyholders' cases but not others, while a practice that clearly violates a statute is necessarily illegal for every insured on whose loss assessment it is used. In assessing Plaintiffs' proposed common questions concerning estimation tactics, then, the court asks whether Plaintiffs have alleged-with evidentiary support-that CMIC adheres to an unfair estimation policy or practice class-wide; or, whether the class-wide practice Plaintiffs allege could be so patently unreasonable as to constitute breach of all class-members' contracts. Only then would answering Plaintiffs questions promise to resolve an element of each class-member's claim against CMIC in “one stroke.”²⁷ Ross, 33 F.4^th at 437.

27 Plaintiffs also allege systemic unreasonable and vexatious conduct in violation of 215 ILCS 5/155. Defendant points out that this provision rests on an underlying contract breach, which Plaintiffs appear not to dispute. (See Redacted Opp'n to Am. Class Certification at 73 (citing Cramer v. Ins. Exch. Agency, 675 N.E.2d 897, 904 (Ill. 1996).) Because the court finds that Plaintiffs have failed to prove commonality on the contract-breach claim, the court does not reach Plaintiffs' ILCS claim.

1. Proposed question “a”: Superficial inspections

In proposed common question “a,” Plaintiffs allege that CMIC systemically conducts “superficial inspections,” resulting in low estimates that dissuade contractors from making repairs on insureds' damaged property. (Redacted Second Class Certification Mem. at 14, 14 n.15.) In support of this allegation, Plaintiffs point to the Hansen Declaration, in which Chad Hansen describes a CMIC adjuster's failure to properly investigate the windows of the Hansens' damaged home. (Id. at 14 n.16 (citing Hansen Declaration ¶ 5).) Second, Plaintiffs point to deposition testimony of Robert Bieber, who heads CMIC's “Large Property Loss Unit.” (Sealed Second Class Certification Mem. at 14 n.15). As described above, Bieber believes it is “very rare . . . that a large claim is never going to change” due to “unforeseen” circumstances and the possibility that “there may be of additional damage.” (Pl.'s Excerpts: Bieber Dep. II at 83:23-84:19; see also Pl.'s Excerpts: Bieber Dep. I at 133:11-19 (“we write what we can physically see is damaged and, from there, . . . typically we're working with a general contractor.... And then once they get into the particular job, they open it up.”).)

This evidence proves neither that CMIC systematically conducts superficial inspections nor that their inspections dissuade contractors from repairing claimants' property. Bieber did not testify that adjusters are not thorough in their inspections; he makes only the unsurprising point that in large property-loss claims, adjusters may miss non-obvious damage on a first go. Nor do Plaintiffs provide any evidence whatsoever that contractors are scared off from working with CMIC policyholders due to the way CMIC's adjusters investigate losses, let alone that adjusters' inspections result in underpayment of claims. See Eddlemon v. Bradley Univ., 65 F.4th 335, 339 (7th Cir. 2023) (in both commonality and predominance inquiry, finding district court's “mere[] accept[ance of plaintiff's] proffered common questions without referring to the common evidence presented to answer those questions” an abuse of discretion). Plaintiffs are left extrapolating from the Hansen Declaration-their own, individual experience-which remains “grossly insufficient” on its own to support class treatment. (See First Class Certification Denial at 23.)

Beyond presenting insufficient evidence of class-wide conduct, Plaintiffs also fail to show how such conduct, even if proven, would “resolve [any] . . . element of the class['s] claims.” Howard, 989 F.3d at 603. Plaintiffs point to their expert's deposition, in which he claims that “very few” claims assessing structural damage would be accurate based solely on a visual inspection. (Sealed Second Class Certification Mem. at 14 n.15 (citing Thomas Dep. Tr. at 286:16-20).) But this does nothing to resolve any element of a class member's claim; Plaintiffs have not cited to any case finding visual inspections to be unreasonable or illegal. And Circuit precedent strongly implies the opposite. See Kartman, 634 F.3d at 891 (noting that “[P]laintiffs have cited no authority for the proposition that an insurer's use of an ad hoc loss-assessment standard, standing alone, qualifies as an independent basis for bad-faith liability” and thus liability “cannot be established on a class-wide basis”). Plaintiffs have thus failed to prove either that their first “common question” would be answered in the affirmative or that an affirmative answer would support class treatment.

2. Proposed question “b”: Improper instructions to adjusters

In Plaintiffs' second question, “b,” they appear to assert that CMIC instructs its adjusters that only CMIC's in-house estimates reflect the “true” ACV and RC for policyholders' claims and that adjusters therefore must “reject [presumably higher] contractor proposals that do not match CMIC's [own] estimate.” (Sealed Second Class Certification Mem. at 36.) Relevant to this claim, Plaintiffs speculate elsewhere in their brief that CMIC “attempts to conceal the fact that it produced and produces lowball estimates by instructing its adjusters that ‘[w]e should also see our own estimate in the file and not that of a contractor, as our estimates will reflect the proper Actual Cash Value and replacement cost.'” (Id. at 13 (quoting Agenda at 9).)

Again, Plaintiffs provide insufficient evidence that much of this conduct occurs at all (let alone how it is relevant to class-members' claims). Plaintiffs cite only to the Hansen Declaration- which describes nothing more than the Hansens' own experience-for the proposition that CMIC tells its adjusters to reject contractor proposals. (See id. at 36 n.145.) The court, again, refuses to extrapolate from that declaration to the experiences of a 400,000-person class. Nor do Plaintiffs show the existence of a common policy of excluding estimates from claim files; according to Plaintiffs, at least 63 of the 313 sample claim files CMIC produced in this litigation (about 20% of the total) do contain contractor estimates. (See Ex. BA at 8.) Plaintiffs explain this away by asserting that the only estimates CMIC included in the files were those in which “the estimate/invoice . . . was [actually] paid.” (Sealed Second Class Certification Mem. at 17.) But again, they cite only their own experience, together with testimony from their expert, Russell Thomas, that he would “expect to see a lot of documentation from contractors in typical insurance companies' claim files.” (Id. at 17 nn. 31-32 (citing Hansen Declaration ¶ 19; Thomas Dep. Tr. At 215:13-17).) Plaintiffs offer no evidence about how often insureds get contractor estimates, how many bids insureds typically get, or whether that number differs depending on the type of repairs needed. Nor have they taken the time to cite to the claim files themselves to support their claim that only paid estimates appear in the files. In other words, Plaintiffs have not shown that CMIC systemically instructs its adjusters to reject contractor bids or always omit every contractor estimate from insureds' files.

Nor is it even clear that such practices, if proven, would support class treatment. Charitably read, Plaintiffs may be claiming that CMIC violates specific provisions of state insurance law in its recordkeeping practices and alleged instruction to adjusters to omit contractors' estimates from claim files. Thus, Plaintiffs have submitted an exhibit in which they compile various state laws relevant to insurer recordkeeping. (See Redacted Second Class Certification Mem. at 17, 17 n. 30 (citing Insurance Law Table).) The exhibit cites, for example, two provisions of the Illinois Administrative Code. The first states that, for claim files subject to market conduct exams, “[d]etailed documentation shall be contained in each claim file in order to permit reconstruction of the company's activities relative to each claim file.” (Id. at 6 (quoting ILL. ADMIN. CODE tit. 50, § 919.30(c)).) The second states the following:

The company is authorized to dispose of or destroy records in its custody that do not have sufficient administrative, legal or fiscal value to warrant their further preservation and are not needed:

a) in the transaction of current business;

b) for the final settlement or disposition of any claim arising out of a policy of insurance issued by the company, except that these records must be maintained for the current year plus 5 years.

(Insurance Law Table at 6 (quoting ILL. ADMIN. CODE tit. 50, § 901.20).) Plaintiffs themselves note that the 5-year requirement was only “added in 2016.” (Insurance Law Table at 6.) Even if this provision applied to contractor estimates, then-and Plaintiffs have not shown whether it would²⁸-the law would apply differently to pre-2016 and post-2016 claim files for class-members, impeding the generation of common answers.

28 The Illinois law defines “records” as:

all books, papers and documentary materials regardless of physical form or characteristics, made, produced, executed or received by any domestic insurance company pursuant to law or in connection with the transaction of its business and preserved or appropriate for preservation by such company or its successors as evidence of the organization, function, policies, decisions, procedures, obligations and business activities of the company or because of the informational data contained therein. If doubt arises as to whether certain papers are ‘non-record' materials, it should be assumed that the documents are ‘records.'

ILL. AMIN. CODE tit. 50 § 901.10. The court is uncertain whether contractors' estimates of repair costs fall into this definition, and Plaintiffs have not clarified this.

As for the other states' recordkeeping laws, Plaintiffs paint with a broad brush, claiming that “the laws of the states where CMIC does business [including Illinois] . . . provide that all documents received by CMIC regarding a claim should be placed in the claim file.” (Redacted Second Class Certification Mem. at 17.) But, as discussed, this overstates at least Illinois law, which claims only that “detailed documentation” should appear in claim files, and for varying amounts of time. ILL. ADMIN. CODE tit. 50, § 919.30(c); id. § 901.20. Plaintiffs do not analyze in any meaningful way how CMIC's alleged recordkeeping practices could or would violate this or any other state's relevant laws, let alone whether the application of different state laws would necessitate subclasses for class-members residing in each implicated state. (See Redacted Second Class Certification Mem. at 17 n.30.)

Plaintiffs have not shown that the conduct alleged in question “b” is either common to the class or, even if it were, would bear on each class-member's claim.

3. Proposed question “c”: Deviating unit prices downward from Xactimate

Question “c,” boiled down, asks “how much CMIC underpays claims through . . . use of unit prices that are lower than those provided by Xactimate.” (Redacted Second Class Certification Mem. at 37.) Yet Plaintiffs' own analysis of the data shows that CMIC adjusters appear to have done so in only 47% of the claims in the sample set.²⁹ (Sealed Second Class Certification Mem. at 56.) In other words, by their own numbers, Plaintiffs' question “c” is far from common to the class: it would apply to less than half of the class's members. Without even reaching the question of whether such a policy could resolve any class-member's claim, then, Plaintiffs do not even allege class-wide conduct.

29 Plaintiffs do not cite to how they arrived at this figure; the court takes them at their word.

4. Proposed question “o”: Underpayment of contents claims

In question “o,” Plaintiffs allege a number of tactics by which CMIC allegedly overdepreciates personal property values and thus, according to Plaintiffs, “underpays contents claims.” First, Plaintiffs argue that “CMIC underpays contents claims by only considering the age of an item” as opposed to employing each of the five factors-“wear and tear, deterioration, obsolescence, physical condition and reduced market value”-listed in the policy as relevant to depreciation. (Redacted Second Class Certification Mem. at 41 (citing Hansen Insurance Policy at 6).) Second, Plaintiffs claim that CMIC makes inappropriate “assumption[s] regarding average use for an entire category of items”-for example, depreciating a “t-shirt . . . [and] dress shirt” by the same “life expectancy.” (Redacted Second Class Certification Mem. at 41.) Finally, Plaintiffs again stress that CMIC classifies all of claimants' property as being in “average” condition when estimating their contents' ACV, as opposed to relying on insureds' self-reporting or adjusters' individualized inspections of each item's condition. (Id. at 65.)

Significantly, however, this proposed common question is not in fact common to the class as a whole. It focuses on “underpay[ment] of contents claims.” (Id. at 41 (emphasis added).)³⁰The proposed class is far broader, as it consists of any “individuals or entities who submitted a claim under a homeowners, condominium, or renters insurance policy issued by CMIC”-in other words, policyholders with either contents or structure coverage. (Id. at 32.)³¹Plaintiffs admit elsewhere in their brief that not every policyholder's claim will implicate both types of coverage. (See id. at 41 n.164, 166). Plaintiffs' question “o” plainly fails to establish commonality.

30 Plaintiffs make a similar allegation for structure claims in their proposed question “n”-that is, that “CMIC underpays structure claims by refusing to consider anything other than age and average life expectancy of a building component.” (Id. at 40-41.) But in their footnote supporting this allegation, Plaintiffs admit that “[t]his common question applies to [only] 42-48.6% of structure claims depending on whether the sample set or Xactimate data dump is used.” (Sealed Second Class Certification Mem. at 41 n.163.) That is, the question is not common to all class members. Plaintiffs also claim in this footnote that further support exists for their structure-related allegations in their brief's Section I(B)(3)(n), but that section only addresses contents claims. (See Redacted Second Class Certification Mem. at 65-66.)

31 Plaintiffs' class definition includes a carve-out for claims where “only personal property was involved and no depreciation was withheld from the contents,” but this does not resolve the problem, as proposed question “o” concerns the depreciation withheld from contents claims regardless of whether structural coverage was also implicated. (See id.)

Nor have Plaintiffs proven that the conduct they allege in question “o” would constitute breach or result in underpayment for any affected class-member. For the proposition that CMIC's assuming “average condition” for insureds' property “harms claimants” (id. at 74), Plaintiffs cite their expert, Donald Stafford, who opined, “I would expect, given an opportunity, that insureds would express that maybe 25 percent of the items are in better than average condition.... For less than average condition, if you got 5 percent, that would be high.” (Sealed Second Class Certification Mem. at 74 n.309 (citing Stafford Dep. at 143:9-15).) Pointing as well to their own analysis of Xactimate data, Plaintiffs claim that “in the few instances where adjusters deviated from average condition,” upward deviations to “above average” were about five times more likely than downgrades to “below average.” (Sealed Second Class Certification Mem. at 74 (citing Contents Summary Table at 2).) Put differently, in the few times insureds asked adjusters to deviate from the default, they more frequently reported their own items as being in “above average” condition than “below average.”

This is the same claim that Plaintiffs made-and the court rejected-in their first classcertification motion. (See First Class Certification Denial at 7 n.2, 21.) Both Plaintiffs' Xactimate analysis and Stafford's deposition testimony simply show that insureds are more likely to selfreport that their property is in good condition than bad when asking CMIC to change its estimates of that property's value. Plaintiffs have not explained why any of CMIC's policyholders, other than the most scrupulously honest, would ever ask adjusters to revise their estimates downward and pay less money on their claims for property that was of less-than-average quality.

Nor have Plaintiffs' additional submissions “resolve[d] this Court's concern” as to whether the uniform use of “average” condition when estimating contents' depreciation results in breach of contract. (Id. at 75 (citing First Class Certification Denial at 14-15).) Plaintiffs cite to a settlement agreement in a California case, based on California class-action and insurance law, and against a different insurer, to argue that an insurance company's use of “average” condition in assessing property loss can constitute breach of contract and form the basis of a class action. (Redacted Second Class Certification Mem. at 75 (citing Settlement Agreement in The Doan v. State Farm (Case No.: 2018-1-CV-129264 Sup. Ct. of Santa Clara), Ex. CZ to Redacted Second Class Certification Mem. [325-43]).) Plaintiffs have not explained how assuming average condition violated California's Insurance Code; why California's Insurance Code should be construed as imposing the same requirements as Illinois insurance law or the terms of CMIC's contracts with insureds; how the California court reached the conclusion that class treatment was proper; or why that conclusion informs this court's determinations under FED. R. CIV. P. 23.

5. Conclusion: Plaintiffs have not identified common questions.

The court simply cannot “answer[]” any of Plaintiffs' proposed questions in a way that is “apt to drive the resolution of the litigation” for the claims of all, or even some, class-members. See Wal-Mart Stores, 564 U.S. at 350 (quoting Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L. Rev. 97, 132 (2009)). Plaintiffs have thus failed to show that the questions they claim define the whole class-“a”-“c” and “o”-unify all class-members in any way cognizable under Rule 23(a)(2). Instead, Plaintiffs' arguments and evidence reveal myriad differences, as to whom the alleged practices affect, whether the practices even occur, and whether, if proven, any practice would establish elements of the class-members' claims. The manifest disjointedness destroys commonality. See Wal-Mart Stores, 564 U.S. at 350 (“Dissimilarities within the proposed class are what have the potential to impede the generation of common answers.”).

Plaintiffs' twenty other “common questions”-“d”-“n” and “p”-“x”-also fail to add up in some way to unify the class. (Redacted Second Class Certification Mem. at 37-44.) Take questions “d”-“n” and “p-q,” which concern estimation practices unique to only certain types of property damage or specific kinds of claims. Questions “d”-“f” concern roof repairs; question “g” involves CMIC's estimation of “waste” on certain items, including shingles and carpeting; question “h” concerns whether CMIC only sometimes pays for industry-standard repair techniques; questions “i”-“k” concern whether CMIC properly protects, estimates, and pays for undamaged property that ought to be removed while repairs are taking place; question “l” appears to concern CMIC underpaying claims by “refusing to pay general contractor overhead and profit” when the insured acts as their own contractor; and questions “m”-“q” concern estimation practices on structure claims. (See id.) By their terms, not one of these questions will apply to every class member: not every class member needed their roof replaced; not every class member needed property removed during repairs; not every class member acted as their own general contractor; not every class member made structure claims (as opposed to contents claims), and so on. Thus, even if Plaintiffs had shown that CMIC has a policy of engaging in any number of these practices, it would do little to support Rule 23(a)(2)'s commonality inquiry for their massive proposed class. See Howard, 989 F.3d at 604 (rejecting commonality because “class members' experiences can vary significantly” depending on their circumstances). Then, question “r” asks “the appropriate types and measures of damages”-a question grossly premature and irrelevant to unifying the class Plaintiffs propose. (See Redacted Second Class Certification Mem. at 42.) And questions “s”-“x” all concern the application of Illinois' vexatious delay statute. (See id. at 42-44.) As discussed above, the court need not reach Plaintiffs' claims under 215 ILCS 5/155 because they require as a foundation a showing of contract breach for which Plaintiffs have not shown class treatment is proper.

Indeed, the very fact that Plaintiffs have ventured so many “common questions” defeats the conclusion that class treatment is appropriate. Plaintiffs allege a panoply of ways in which CMIC allegedly underpays its clients but have not shown that any one of them apply to all classmembers, or that class-wide evidence would help resolve individual class-members' claims. See McCaster, 845 F.3d at 801 (in affirming denial of class certification, stressing that plaintiffs “do not contend . . . [that Defendant's] policies facially violate the IWPCA. Nor have they alleged that [Defendant] had a statewide practice of withholding payment of accrued earned vacation-much less supplied evidence that could support such a finding.”) The court is unable to find in Plaintiffs' copious briefing any common question promising to resolve CMIC's potential underpayment of insurance claims more efficiently in the (enormous) aggregate, let alone how any nub of commonality hidden in this thicket would predominate over the manifest individualized issues that would plague each class-member's case.

C. Potential Subclasses

Nor will the court certify a potential subclass. For one, as noted, Plaintiffs do not want the court to do this; they state firmly that they “are not ‘hoping the court will certify a narrower class based on one or more of the 19 ‘common questions listed in their motion.'” (Redacted Second Class Certification Mem. at 46 n.187 (quoting First Class Certification Denial at 28).) Plaintiffs continue to criticize even the idea of subclasses, asserting that “[c]ertifying a class based on one or more of these cheats would: result in too narrow of a class; reward CMIC's grossly deficient production [in discovery], result in an underpayment of damages to the class; and be inconsistent with precedent such that CMIC might actually prevail on appeal.”³²(Redacted Second Class Certification Mem. at 46 n.187.) A “district judge [i]s not obliged to allow [a] suit to be maintained as a class action in the face of the refusal of the class representative's lawyer to cooperate in dividing the class into subclasses.” Culver v. City of Milwaukee, 277 F.3d 908, 913 (7th Cir. 2002). Given that it is against Plaintiffs' wishes, the court harbors serious doubt that certification of a subclass would leave class-members in the hands of adequate class representatives. Id. (noting that class counsel's aversion to subclasses was “ground[s] for decertification, as it was a further indication that the plaintiff was an inadequate representative of the class(es)”).

32 Instead, Plaintiffs claim the “better course is to certify the class proposed by Plaintiffs and deal with any overbreadth problem if a problem materializes.” (Id.) In support of this, Plaintiffs quote language from a Seventh Circuit case noting that when “minor overbreadth problems . . . do not call into question the validity of the class as a whole, the better course is not to deny class certification entirely but to amend the class definition as needed to correct for the overbreadth.” (Id. (quoting Suchanek v. Sturm Foods, Inc., 764 F.3d 750, 757 (7th Cir. 2014).) This argument, viewed generously, is one that asks the court not to certify multiple subclasses, but instead to certify just one, narrower class in the face of overbreadth problems. The court declines the invitation. In Suchanek, the common question was whether a company's “packaging was likely to mislead a reasonable consumer” as to the quality of the coffee product it sold. 764 F.3d at 755. The “minor overbreadth problem[]” was that “a later version” of the company's packaging included the word “instant” in its description of the coffee, changing slightly the conduct as applied to some subset of the class-members. Id. at 757. As discussed exhaustively above, Plaintiffs' problems go far beyond the “minor overbreadth” discussed in Suchanek.

And if the court were inclined to go against Plaintiffs' wishes and venture ahead without their assistance, Plaintiffs have not shown the existence of a subclass appropriate to certify.

Primarily, Plaintiffs have not even provided evidence tending to show that the conduct they allege in common questions “d”-“n” and “q” actually applies broadly to those within the potential subclasses hinted at in those questions. See McCaster, 845 F.3d at 801 (noting absence of evidence of Defendant's class-wide practice in finding no commonality). For example, question “e” alleges that CMIC “fail[s] to include steep and/or height surcharges on roof claims involving steep or higher roofs, and even where CMIC includes these surcharges CMIC uses the wrong labor rate for the removal steep and height surcharges.” (Redacted Second Class Certification Mem. at 37.) But the very phrasing of this question shows that Plaintiffs' allegations do not unify an implied subclass of policyholders who had roofs repaired: CMIC apparently sometimes “fail[s] to include steep and/or height surcharges on roof claims” and, in others, does, but uses “the wrong labor rate.” (Id.) Plaintiffs then note that CMIC failed to include such surcharges on only “15% of roof claims in the sample set.” (Id. at 37 n.152.) That, then, is not a practice even close to universal as to the subclass of those who had roofs repaired. As Defendant points out, a similar defect-Plaintiffs' proffered proof showing the opposite of even subclass-wide policy or conduct- infects most of Plaintiffs' claims. (See Sealed Opp'n to Am. Class Certification Mot. at 33-37; see also Sealed Second Class Certification Mem. at 37-42 nn. 152, 155-61, 163, 167.)

As the court sees it, only two of Plaintiffs' ventured “common questions” appear to reflect practices affecting all members of a potential subclass. The first is Plaintiffs' question “g,” which alleges in part that CMIC does “not allow[] sufficient scrap/waste material based on industry standards” when calculating “shingle waste.” (Sealed Second Class Certification Mem. at 38.) According to plaintiffs, CMIC uses “5%” when the industry standard is “10-15%,” and by their analysis, this alleged defect occurs in 90-95% of roof claims-enough to imply to the court the existence of a company policy. (Id. at 38 n.154.) Similarly, Plaintiffs' question “d” asks “whether CMIC underpays claims by using the general demolition labor rate instead of the higher roofer labor rate for roof removals” when estimating roof repair costs. (Id. at 37.) And, according to Plaintiffs' analysis of Xactimate data, this defect occurs in approximately 97% of the claims in which CMIC estimated the cost of an insured's roof repair. (Id. at 48.) Accordingly, regardless of whether either estimation practice is likely so unreasonable as to be illegal-and illegal as applied to each subclass-member's roof, be it flat, steep, low, high, shingled, and so on-at least Plaintiffs' evidence points toward the existence of some unifying, class-wide policy. As the court noted previously, there are thus “hint[s]” of common questions in these two allegations sufficient to meet Rule 23(a)(2)'s bar. (See First Class Certification Denial at 28.)

Yet these potential subclasses run into other difficulties. Rule 23(a)(3) requires that “the claims or defenses of the representative parties are typical of the claims or defenses of the class,” and Rule 23(a)(4) requires that “the representative parties will fairly and adequately protect the interests of the class.” FED. R. CIV. P. 23(a)(3),(4). Broadly, typicality asks whether the named plaintiffs' claims encompass the “same essential characteristics as the claims of the class at large.” Oshana v. Coca-Cola Co., 472 F.3d 506, 514 (7th Cir. 2006). And class representatives can be inadequate for multiple reasons, including “[t]he presence of even an arguable defense peculiar to the named plaintiff.” CE Design Ltd. v. Kind Architectural Metals, Inc., 637 F.3d 721, 724 (7th Cir. 2011).³³Unique elements of named plaintiffs' claims-including defenses-can make their claims atypical and thus render them inadequate class representatives. Understandably, then, “[i]n many cases, the requirement of typicality merges with the further requirement that the class representative will fairly and adequately protect the interests of the class.” Id.

33 Inadequacy of class counsel can defeat certification. Culver, 277 F.3d at 913 (noting that “the performance of the class lawyer is inseparable from that of the class representative” in assessing adequacy). A class lawyer's “pattern of ineffective representation” thus bears direct relevance to the propriety of certifying a class. Davidson v. Citizens Gas & Coke Util., 238 F.R.D. 225, 230-31 (S.D. Ind. 2006) (noting that counsel had “marred [the case's] progress in finding inadequacy, pointing in part to numerous “instances when Plaintiffs' counsel . . . improperly submitted briefs and supporting materials”); see also Culver, 277 F.3d at 913 (highlighting counsel's “flurry of frivolous motions” in adequacy analysis); Creative Montessori Learning Centers v. Ashford Gear LLC, 662 F.3d 913, 917 (7th Cir. 2011) (noting that “class counsel have demonstrated a lack of integrity that casts serious doubt on their trustworthiness as representatives of the class”). With his repeated misrepresentations of the record, multiple unfounded attacks on the opposing party (noted by the magistrate judge handling discovery), and failure to follow court orders by repeatedly injecting briefs disguised as exhibits, Plaintiffs' counsel raises legitimate concern as to his ability to adequately represent a class in this matter.

The Hansens' claim would be atypical and their representation inadequate for a roofing subclass. Potential roofing-claim subclasses would argue that CMIC used “cheats” to underestimate the cost of repairing claimants' roofs. (See Redacted Second Class Certification Mem. at 37, 38.) But as Plaintiffs' interrogatory responses, deposition testimony, and Defendant's Wells Fargo exhibit show, CMIC over-estimated the cost to repair Plaintiffs' roof by about $3,000. The evidence thus suggests promising defenses for CMIC on both breach and damages. In other words, ironically, a roofing subclass's best hope may well be that Plaintiffs' claim is atypical.³⁴This clear defect in their claim prompts serious concerns about whether Plaintiffs could adequately represent a subclass of CMIC policyholders whose roofs needed repairs.

34 Defendant's expert opines that, beyond the roof-repair issue, the Hansens' claim is unusual as compared to other policyholders' files:

This claim is an outlier and is not similar to any of the claims in the sample set for many reasons - the adversarial position of the insured, the amount of correspondence sent by the insureds, which was at times abusive and threatening, the desire of the insured to be general contractors and possibly profit from the loss, the number of complaints filed with the DOI [Department of Insurance], COUNTRY's willingness to accept bids from sub-contractors without close review while adding overhead and profit, the amount of clothing dry cleaned and later replaced without question, the number of drafts issued (90) to the insureds over the life of the claim, the lack of abatement for normal living expenses, the partial payment of building code upgrades prior to completion, the use of outside counsel to assist with the handling of the claim. The issues and circumstances involved in this claim are far beyond anything in the sample set, and it would be rare to encounter this type of hostility from first party insureds ....

(LaPrade Report at 54-55.) The accuracy of LaPrade's opinion aside, the Hansens' long-running and complex history with CMIC-and potential idiosyncratic disputes-exacerbate the court's concern that they would not adequately represent a subclass of roof claimants.

Plaintiffs' response is to argue that they would be entitled to damages regardless of whether CMIC's estimate was high enough to cover their roof repair. In support of this, they point to the fact that insureds can keep their ACV if it exceeds the actual cost of repair. (Pl.'s Consolidated Reply at 16 n.7, 8.) Expectation damages might thus exist even if CMIC over- estimated the price of getting Plaintiffs' roof repaired, as long as both the ACV they received and the “correct” ACV were higher than the RC. (See id. at 16 n.7 (citing Stuart, 910 F.3d at 37677).) In that (presumably rare) case, the insured would be entitled to whatever deficit existed in the improperly calculated ACV notwithstanding their having actually paid less for repairs. Perhaps such damages could exist in Plaintiffs' own case. But their unusual posture does not justify certifying a subclass of every CMIC policyholder who had roof damage. For one, the fact that Plaintiffs' proffered proof shows substantial over-estimation of their own claim casts doubt on the likelihood that they can adequately argue, on a class-wide basis, that CMIC's estimation practice is so unreasonable as to render it a breach of all of CMIC's contracts with roof-damaged policyholders. Furthermore, whether damages existed-let alone the form they would take- would differ substantially for other class-members. Some small number of policyholders, like Plaintiffs, might have expectation damages despite their estimates being higher than the cost of repair; some-necessarily a much greater number if CMIC's tactics were geared at low-balling costs-would have ACVs that underestimated the cost of repair. But then, of that latter category, some policyholders will have gotten repairs done within a year and thus been reimbursed some amount; and of those, that reimbursement might or might not account for the entire amount by which CMIC's tactic illegally underestimated the ACV, changing the availability of expectation damages. Plaintiffs do not address how, in their representation, they would adequately represent a class potentially encompassing all these different permutations (not to mention the obvious predominance concerns). For all these reasons, the court harbors no doubt that the best interests of any hypothetical subclass (or subclasses) lie in different named representatives.

D. 23(b)(3) Requirements

Because Plaintiffs' proposed class does not pass muster under Rule 23(a), the court need not go any further in addressing class certification. It is worth noting, however, that Rule 23(b)(3) requires that the court “find[] that questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” FED. R. CIV. P. 23(b)(3). The predominance inquiry “requires a court to ‘understand what the plaintiffs will need to prove and . . . evaluate the extent to which they can prove their case with common evidence.” Eddlemon, 65 F.4th at 339 (quoting In re Allstate Corp. Sec. Litig., 966 F.3d 595, 603 (7th Cir. 2020)). The predominance inquiry thus “calls upon courts to give careful scrutiny to the relation between common and individual questions in a case.” Tyson Foods, 577 U.S. at 453.

Having already addressed Plaintiffs' failure to show the existence of common questions unifying the class-which necessarily means none predominate-the court pauses to consider a final obstacle: Plaintiffs have not presented evidence of class-wide injury. Though such evidence is not strictly necessary at the class-certification stage, a high number of uninjured class-members prompts serious predominance concerns. See, e.g., In re Asacol Antitrust Lit., 907 F.3d 42, 5354 (1st Cir. 2018) (noting the patent predominance failure where “there are apparently thousands [of class-members] who in fact suffered no injury . . . [and] [t]he need to identify those individuals will . . . render an adjudication unmanageable”).³⁵The majority of Plaintiffs' allegations- proposed questions “a”-“o,” and “q”-focus explicitly on whether CMIC “underpays claims.” (See Redacted Second Class Certification Mem. at 26-33.)³⁶ And Plaintiffs themselves assert that a comparison of contractor bids with CMIC payments would show whether at least structure claims were properly paid. (Id. at 10 n.1.) Yet, by Plaintiffs' own analysis of the sample set, CMIC paid claimants 100% or more of contractors' bid amounts in “44%” of the claims in which such bids appeared.³⁷(Ex. BA at 8.) Moreover, to the extent CMIC's estimation practices standing alone could constitute common questions, Plaintiffs' own evidence undermines this: Plaintiffs' same table appears to show that CMIC's final estimates of repair costs also often matched or exceeded contractors' bids; by the court's count, in 21 of the 63 claims, or 33%. (See id. 1-8 (comparing “Country Final Estimate Amount” with “Lowest Bid/Invoice in File”); see also Redacted Opp'n to Am. Class Certification at 27, 27 n.94.) Extrapolating from the sample, then, and according to Plaintiffs' evidence, certifying this class would require locating and excluding potentially 176,000 putative class members (44% of the 400,000 total) whose insurance claims were properly paid and who thus suffered no injury.³⁸ See TransUnion LLC v. Ramirez, 594 U.S. __, 141 S.Ct. 2190, 2208 (2021) (“Every class member must have Article III standing [and thus concrete injury] in order to recover individual damages.”) Nowhere do Plaintiffs explain how the court could do so with common evidence.

35 Similarly, the fact that “few, if any, of the putative class members share the named representative's grievance[s] against the defendant” can cast doubt on the typicality of the named representatives and thus the propriety of class treatment. Suchanek, 764 F.3d at 757.

36 Plaintiffs' other “common questions” mostly revolve around the unreasonable and vexatious delay claim. (See id. at 33-35.) As the court noted earlier, these questions are irrelevant because they rely on class-wide breach whose suitability to class treatment this court has rejected.

37 For its part, Defendant argues that closer inspection of the relevant files in which Plaintiffs claim were underpaid-i.e., the other 56%-shows that, there too, CMIC paid for the requisite repairs in accordance with its policy. (See Sealed Opp'n to Am. Class Certification at 27-28, 34-42.) And Plaintiffs venture their own unique, claim-specific explanations for each estimate or payment that met contractors' bars, seeking to problematize the seemingly appropriate action on CMIC's part. (Ex. BA at 1-8 (commenting on claim files going against them with notes like “It does not say what was done or that Country's scope was followed”).) These explanations themselves confirm that questions of underestimation and payment of complicated insurance claims generate individualized inquiries unsuited to class treatment.

38 As discussed regarding the roofing subclass, the fact that some small portion of class-members may still be entitled to expectation damages on the theory that their ACV overestimated the RC but nonetheless was illegally low only proves the point: finding those classmembers would force the court to train even more heavily on individuals as opposed to the class as a whole.

Indeed, this Circuit has observed that in the insurance context, whether “a compensable loss occurred and was underpaid . . . [is] a claim-specific inquiry that turns on the nature of the damage to each plaintiff's [property] and the amount [the insurer] paid to repair it [T]he classaction device is not appropriate for resolving such highly individualized questions of fact.” Kartman, 634 F.3d at 891. The named Plaintiffs' own claim is a case in point. The parties vehemently dispute almost everything about the Hansens' claim: from coverage to estimation to payment, spanning from the highest level (the contents and structure payouts) down to the lowest (grouped and individual line items).

Then, zooming out, a comparison of Kartman and this case casts into even further relief the profound predominance problems here. The failed class in Kartman was comprised of 7,000 people whose roofs were damaged in a single hailstorm in 2007. Id. at 887. Plaintiffs seek to certify a class of over 400,000 people, in nineteen different states, whose property of any kind- external, internal, structural, personal-was damaged in any compensable way over the course of over a decade. Plaintiffs have utterly failed to show how their bewilderingly expansive class would somehow prove the exception to the Seventh Circuit's observation in Kartman.

CONCLUSION

In denying Plaintiffs' second attempt to certify this class, the court does not pass on the validity of any one policyholder's potential contract claims against CMIC. Nor does the court suggest that CMIC or any other insurer is somehow immune from being held liable on a classwide basis. In their two attempts at certification, however, Plaintiffs plainly failed to distill an actionable common question. The voluminous record and myriad allegations they have assembled reveal complications rather than commonalities, and any common question that could emerge from this record is dwarfed by the individualized inquiries necessary to resolve potential class-members' claims. Plaintiffs' motion [320] is denied. Defendants' motion to exclude portions of Chad Hansens' declaration [334] is granted. Plaintiffs' motion to exclude two of Defendant's exhibits [341, 343] is denied. The parties' remaining motions to exclude [333, 342] are denied as moot.