Opinion
No. 16 C 4321
06-04-2020
CRAFTWOOD LUMBER COMPANY, ET AL., Plaintiffs, v. ESSENDANT, INC., ET AL., Defendants.
MEMORANDUM OPINION & ORDER
Plaintiff hardware stores Craftwood Lumber Company and Craftwood II, Inc. (together, "Craftwood") allege that defendants Essendant, Inc. and Essendant Co. (together, "Essendant"), distributers of office products, janitorial supplies and more, sent Craftwood unsolicited fax advertisements in violation of the Telephone Consumer Protection Act of 1991 ("TCPA"), as amended by the Junk Fax Protection Act of 2005. R. 1. Craftwood also makes class allegations on behalf of others similarly situated. Before the Court is Craftwood's amended motion for class certification. R. 79. For the reasons explained below, that motion is denied.
Standard
To be certified, a putative class must satisfy the four prerequisites of Federal Rule of Civil Procedure 23(a): numerosity, commonality, typicality, and adequacy of representation. Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 811 (7th Cir. 2012). The action also must satisfy at least one of the three subsections of Rule 23(b). Id. Here, Craftwood seeks certification under Rule 23(b)(3), which requires a finding 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 the fair and efficient adjudication of the controversy." "Plaintiffs bear the burden of showing that a proposed class satisfies the Rule 23 requirements," Messner, 669 F.3d at 811, and must do so "through evidentiary proof." Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). Class certification "analysis will frequently entail 'overlap with the merits of the plaintiff's underlying claim,'" id. at 33-34 (quoting Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 351 (2011)), but "[m]erits questions may be considered . . . only to the extent . . . that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied." Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 568 U.S. 455, 466 (2013).
District courts have "broad discretion" when determining whether a proposed class satisfies Rule 23. Howland v. First Am. Title Ins. Co., 672 F.3d 525, 528 (7th Cir. 2012); see also Dukes, 564 U.S. at 369 ("[M]ost issues arising under Rule 23 . . . [are] committed in the first instance to the discretion of the district court.").
Background
Additional background information is set forth in the Court's November 3, 2017 Memorandum Opinion and Order granting Essendant's motion to deny certification of Craftwood's initial putative class discussed infra. 16 C 513, R. 117.
In December 2016, the Court consolidated this TCPA case with another, Alpha Tech Pet, Inc. et al. v. Essendant, Inc., No. 16 C 513 ("Alpha Tech"), for pre-trial proceedings. R. 45. The Alpha Tech plaintiffs and Craftwood (plaintiffs from both lawsuits together, "Plaintiffs") asserted the same claims against Essendant and effectively sought to represent the same class (the "initial class"), which included all persons and entities to whom Essendant sent fax transmissions from May 2011 to May 2015. The claims implicated approximately 1.5 million faxes in 725 separate transmissions to nearly 24,000 unique fax numbers. Plaintiffs allege that the opt-out notices included in the faxes were insufficient.
Essendant preemptively filed a motion to deny class certification following the D.C. Circuit decision Bais Yaakov of Spring Valley v. FCC, 852 F.3d 1078 (D.C. Cir. 2017), arguing that Bais Yaakov found the so-called FCC "Solicited Fax Rule" unlawful to the extent it required opt-out notices on solicited faxes. This Court agreed, and in granting Essendant's motion to deny certification of the initial class in November 2017, concluded based on Bais Yaakov that the TCPA requires opt-out language on only unsolicited faxes, and that individualized questions regarding consent "would require a series of mini-trials, thus defeating predominance and superiority" in Plaintiffs' cases. 16 C 513, R. 117 at 8. In so holding, this Court noted that Essendant had shown that consent for these purposes comes in many forms, including: orally; via consent forms; entries in its Trend database reflecting that consent forms had been collected from a particular customer; and through its practice of advising customers at the outset of the relationship of their option to receive Essendant faxes and requesting fax numbers for that purpose. See 16 C 513, R. 117 at 11-13 (citing the Junk Fax Prevention Act of 2005, 71 FR 25967-01, 2006 WL 1151584, and discussing the ways in which Essendant demonstrated that it had received and/or recorded consent). Essendant also demonstrated that some consent records were lost or destroyed in the aftermath of Hurricane Katrina, and that its written records underrepresented the consent actually obtained, not only because some consent was oral and not formally recorded, but also because its Trend database allows for just one fax number entry per recipient, and in many cases recipients have more than one and/or subsequently changed their fax number(s). Id. at 13-14 (citing 16 C 513, R. 71-2 ¶¶ 29-33 & Ex. 9 (former Essendant Director of Sales Jon Phillips's initial declaration)). To that end, Essendant also provided declarations from fax recipients who attested to having consented to receive faxes, but who were not reflected in the Trend database as having consented and for whom no consent forms have been located. Id. at 13. This Court held that all of this constituted "concrete evidence of consent" precluding certification under Rule 23(b)(3). Id. at 12 (collecting cases).
Plaintiffs appealed, and in December 2018, the Seventh Circuit affirmed, stopping short of holding that Bais Yaakov overruled the Solicited Fax Rule, but concluding that it "drained [it] of a great deal of force." Brodsky v. Humana Dental Ins. Co., 910 F.3d 285, 290 (7th Cir. 2019). The court held that this Court was "within [its] rights" to find that there were "enough problems with class treatment" that a class action was "not a superior mechanism for adjudicating" this case, because it was "necessary to distinguish between faxes sent with permission . . . and those that are truly unsolicited." Id. at 291. The court continued, "the question of what suffices for consent is central, and it is likely to vary from recipient to recipient (or so the district court reasonably could have concluded)," and as such "issues concerning solicitation, permission, pre-existing relationships, and the like . . . remain[ed] as obstacles to class treatment." Id. at 291-92.
While Brodsky did not officially sound the death knell for the Solicited Fax Rule in this Circuit, the FCC eliminated it in November 2018 and dismissed as moot all pending applications for retroactive waivers. See FCC Order, DA 18-1159 (available at https://docs.fcc.gov/public/attachments/DA-18-1159A1.pdf).
Plaintiffs subsequently informed the Court and Essendant of their intent to seek certification of a new class and amend their respective complaints. They described the class as concerning 9,848 "fresh numbers" of non-customers who received faxes on nine dates over a two-month period in 2015, and asserted that "the most likely explanation why Essendant's fax target list spiked . . . practically overnight" was that "Essendant received the fresh numbers from someone else—all at once." 16 C 513, R. 145 at 5. Plaintiffs claimed that additional discovery would help prove this theory, and defined the proposed class as:
All non-customers of Essendant that were subscribers of telephone numbers to which Essendant (including its subsidiaries) sent facsimile transmissions on [certain dates in March and April 2015] through ABC Fax.Id. at 3-4. According to Plaintiffs, this narrowed class addressed the Court's earlier concerns about individualized consent inquiries. But in declining to allow the amendment and reopen discovery, the Court pointed out that Plaintiffs' shift in focus to strictly non-customers changed the theory of the litigation in cases that were more than three-years-old and in which discovery had been closed for over two years. And the Court noted that it had already entertained and denied an earlier request by Plaintiffs to reopen discovery on the heels of Essendant's initial motion to deny class certification in 2017, at which point Plaintiffs inexplicably sought to take their very first deposition, and that their position was even less tenable now. Moreover, the Court concluded that the amendment would likely be futile in any case, because even for the proposed narrowed class, the parties disagreed over whether and how many of the fax recipients in question were customers or otherwise consented to receive faxes.
The Alpha Tech case settled thereafter. But Craftwood again sought leave to file an amended motion for class certification of "fresh numbers" recipients under the same third party theory, this time representing that no further discovery and no amended pleading would be necessary. 16 C 513, R. 149 at 2. Craftwood reasoned that Essendant had never claimed that non-customers had consented to receive faxes, so consent was a non-issue, and under the Seventh Circuit's decision in Mullins v. Direct Digital, LLC, 795 F.3d 654 (7th Cir. 2015), the inquiry into a putative class member's customer status would go to the manageability of the class under Rule 23(b)(3)'s superiority prong, and would not defeat certification. Id. at 5-6. In response, Essendant contended that certification was plainly improper for the same reason it was with regard to both the initial class, and the narrowed class for which Essendant subsequently sought certification; that is, because individualized consent issues predominated. The Court warned that no further discovery would be allowed, but granted Craftwood leave to file its amended class certification motion. That motion proposes a class definition that matches exactly that which the Court previously rejected when Craftwood initially sought to amend, implicating the same 9,848 "fresh numbers" fax recipients. See R. 79 at 2; R. 79-2 at 3.
Analysis
Generally, a court enjoys broad discretion to alter or amend a class certification order prior to final judgment. Fed. R. Civ. P. 23(c)(1)(C). And that includes the discretion to certify a narrowed class on a renewed certification motion. See Otero v. Dart, 306 F.R.D. 197, 208 (N.D. Ill. 2014); Mednick v. Precor, Inc., 320 F.R.D. 140 (N.D. Ill. 2017). But the putative, narrowed class still must satisfy Rule 23's requirements. Essendant attacks Craftwood's amended motion for class certification on every aspect of Rule 23(a) and (b)(3) and then some. But the Court need look no further than the parties' arguments regarding consent and Rule 23(b)(3) to resolve Craftwood's motion, so its analysis starts and ends there.
"[T]o determine whether any putative member of the proposed class ha[s] a TCPA claim," the Court first must "determine whether that proposed class member 'solicited,'" or consented to, "the faxes it received." Brodsky, 2017 WL 3704824, at *10. The consent question is context-dependent, and often precludes certification under Rule 23(b)(3). See e.g., id. at *5, *10 ("individual consent issues defeat predominance and superiority,"); Sandusky, 863 F.3d at 466 (individual consent issues "keep common questions from predominating"). Indeed, "[i]f individual issues predominate, then class certification is usually not a superior method for resolving the controversy, since management of such issues by a court will not be efficient." See Clark v. Experian Info., Inc., 233 F.R.D. 508, 511 (N.D. Ill. 2005), aff'd sub nom. Clark v. Experian Info. Sols., Inc., 256 F. App'x 818 (7th Cir. 2007).
In addition to potential individual questions regarding the manner in which consent is obtained or the extent of the consent provided, there can also be individual questions about whether consent was required at all. As the Seventh Circuit observed, "[t]here are exceptions to the prohibitions on sending unsolicited faxes, including situations where the sender has an established business relationship with the recipient [and] the sender obtains the recipient's fax number through some sort of voluntary communication." Brodsky, 910 F.3d at 290 (quoting 47 U.S.C. § 227(b)(1)(C)(i)-(iii)).
But there are no individualized consent concerns when recipient fax numbers are obtained from a single source, because the issue is then susceptible to class-wide resolution. See Wolfkiel v. Intersections Ins. Servs. Inc., 303 F.R.D. 287, 294 (N.D. Ill. 2014) (consent can be addressed on a class-wide basis where the source of the contact information for the fax recipients is a single "leads" list); see also Hinman v. M & M Rental Ctr., Inc., 545 F. Supp. 2d 802, 807 (N.D. Ill. 2008) (consent issue was common where "fax broadcasts at issue were sent en masse to recipients identified on a singular 'leads' list obtained from singular source"); Licari Family Chiropractic Inc. v. eClinical Works, LLC, 2019 WL 7423551, at *7 (M.D. Fla. Sept. 16, 2019) ("in cases in which a sender obtained all of the fax recipients' fax numbers from a single purveyor of such information there exists a class-wide means of establishing the lack of consent") (internal citations omitted). And that's Craftwood's theory here. As explained, Craftwood argues that the "fresh numbers" were likely obtained "at once" from a third party—suggesting one of Essendant's three buying group cooperative customers Do-It-Best, True*Serv or Ace Hardware, of which it contends many "fresh number" recipients, including Craftwood, were members—since Essendant had not faxed them before. But the best Craftwood can do to support its theory on the limited discovery before it is: (1) to point out that the "fresh numbers" associated with buying group members are listed on Essendant's Customer ID list by their respective buying group's Customer ID number, rather than a number that is unique to the recipient; and (2) reiterate that Essendant never claimed to have obtained consent from non-customers, which it contends includes buying group members. R. 103-1 ¶ 5. But former Essendant Director of Sales Jon Phillips attests that Essendant did not purchase the "fresh numbers" from a buying group or any other third party, and Craftwood has presented no other evidence that it did. See R. 91-1 ¶ 10 (Phillips' Second Declaration stating "Essendant did not purchase a third-party list of the 'fresh numbers'").
Further, the evidence shows that more than 2,750 of the 9,848 "fresh numbers" are accounted for on the Customer ID list by reference to their own specific Customer ID number (and not that of a buying group). See R. 103-1 ¶ 5. And thousands of the "fresh numbers" are accounted for in Essendant's records. See 91-2 ¶¶ 10, 12-16, 18-19 (reflecting that 2,160 of the "fresh numbers" have a fax number listed in Essendant's Trend database, 427 of the "fresh numbers" appear in Essendant's credit files, and that hundreds of additional "fresh numbers" were accounted for as matching with a Trend database entry or credit file using additional identifiers such as telephone numbers and customer names). Further, almost half of the fax recipients who confirmed via declaration in connection with Craftwood's initial class certification bid to have consented to receive faxes despite a lack of documentary evidence are among the "fresh numbers" population. R. 91 at 10. This evidence belies any inference that might otherwise be drawn regarding a buying group as the "singular source" of those numbers (and thus that consent is a question common to the putative class), and counsels against certification here. See Licari Family Chiropractic Inc., 2019 WL 7423551 at *8 ("Plaintiffs' inability to establish or even allege a generalized theory of prior express permission . . . demonstrates that this matter is unsuitable for class treatment."); see also Gene And Gene LLC v. Biopay LLC, 541 F.3d 318, 329 (5th Cir. 2008) (holding that "the determinative question of whether consent can be established via class-wide proof must . . . be answered in the negative" because the plaintiff "failed to advance a viable theory of generalized proof to identify those persons, if any, to whom BioPay may be liable under the TCPA.").
Further, even if Craftwood presented evidence of a "leads" list for those "fresh numbers" buying group members that are not accounted for in Essendant's files, the Court doubts that it would suffice to demonstrate a class-wide basis for determining consent to fax here. Indeed, in a best-case scenario, three buying groups necessarily implies three (or more) "leads" lists, and thus at least as many questions regarding whether a recipient's presence on any such list means that the recipient had consented to receive faxes. Cf. Hinman, 545 F. Supp. 2d 802, 807 (N.D. Ill. 2008) (predominance and superiority of class treatment established in part by fact that fax broadcasts at issue were sent "en masse" to recipients "identified on a singular 'leads' list obtained from a singular source.").
But Craftwood pivots once more in its reply brief to argue that it never contended that all of the 9,848 "fresh numbers" recipients were non-customers, and arguing for certification of some further narrowed class. According to Craftwood, those "fresh numbers" recipients which are accounted for in Essendant's records are customers, while those not accounted for are not, and again because Essendant never claimed to have obtained consent from non-customers, each such recipient must have a claim here. R. 103 at 13. Craftwood continues that any difficulty determining a particular recipient's customer status could not defeat class certification, because the Seventh Circuit directs that courts are not to deny certification "simply because it may be challenging to identify . . . class members." R. 103 at 6-9 (quoting Mullins, 795 F.3d at 664).
But these arguments also fail. First, in so arguing, Craftwood effectively admits the failure of its initial theory that the "fresh numbers" were obtained from a third party, while simultaneously (and confusingly) insisting that they were. Further, in arguing that Essendant never claimed to have obtained consent from non-customers, Craftwood fails to acknowledge that the parties' differing (and in Craftwood's case, shifting) definitions of "customer" explain the disconnect. Indeed, while Craftwood seems to argue that at the very least, buying group members such as themselves who purchased only indirectly from Essendant through their buying groups are not customers, it all but ignores that Essendant apparently regards all buying group members as customers—referring to those that purchased from it directly as "direct customers" in its brief, and those that purchased from it only through their respective buying groups as "indirect customers"—and treated them as such. Indeed, according to Mr. Phillips, Essendant communicated with all buying group members in the same way as any other customer, and obtained their consent to fax in the same ways, too. R. 91-1 ¶ 8. In other words, Essendant communicated in each of the manners of consent outlined and held to preclude certification in the Court's 2017 order (including oral consent), effectively eviscerating any argument that a lack of record means a lack of consent.
Indeed, Craftwood contends on reply that:
[i]n ten blasts in March and April 2015, Essendant faxed almost 10,000 new numbers. In the preceding four years Essendant had faxed only 14,000 numbers total. These "fresh" numbers appeared all-at-once—almost doubling the fax program. It defies logic to think these were customer numbers sitting in Essendant's files.R. 103 at 15. As discussed supra, the latter is exactly what the evidence reflects as to thousands of the 9,848 "fresh numbers."
Mr. Phillips further attested to his belief that "if an individual investigation were performed with respect to each of the fax recipients . . . in the purported 'fresh number' population (including by speaking to the appropriate representatives of each recipient), the results would show that Essendant had the prior express permission of most (if not all) of the recipients." Id. ¶ 9. Essendant provided actual examples of that consent, including among those meeting Craftwood's "non-customer" definition, along with evidence that consent for this population is underrepresented, just as it was for Craftwood's initial class. Compare R. 103 at 12-13 (Craftwood's reply brief arguing that any "fresh numbers" recipients not accounted for in the Trend database are "non-customers"), with R. 91-2 (Essendant attorney Lauri Mazzuchetti's declaration identifying "fresh numbers" recipients not in Essendant's Trend database or credit files but for which Essendant has a consent form; indicating that buying group members were among the hundreds of "fresh numbers" recipients for which a "yes" value for consent is reflected in Trend database entries; and explaining why Essendant's records likely hugely underestimate the consent actually obtained).
Aside from buying group members, Craftwood does not describe any other "type" of alleged non-customer among the "fresh numbers" population. And that makes sense, given Craftwood's theory that buying groups were the source of the numbers in the first place. Further, except for showing that 120 buying group member recipients' fax numbers are not on Essendant's customer list or reflected on Trend and to complain that Mr. Phillips' testimony is only "speculation," Craftwood offers little to counter Essendant's evidence and argument. But again, a lack of records does not mean a lack of consent, and as this Court previously told the parties, evidence need not be admissible for purposes of resolving class certification issues. See 16 C 513, R. 117 at 16 (citing Young v. Fortis Plastics, LLC, 294 F.R.D. 128, 135 (N.D. Ind. 2013)); see also In Re Front Loading Washing Mach. Class Action Litig., 2013 WL 3466821, at *10 (D.N.J. July 10, 2013) ("The Federal Rules of Evidence are not stringently applied during class certification and courts may consider evidence that might later be ruled inadmissible at trial"). The Court finds more than enough to preclude certification here.
In sum, because Essendant presents evidence that a significant percentage of the "fresh numbers" population was already accounted for in its records, without more, Craftwood's "leads" list theory fails, and so too any contention that consent is susceptible to class-wide proof. See Gorss Motels, Inc. v. Brigadoon Fitness, Inc., 331 F.R.D. 355, 360-61 (N.D. Ind. 2019), reconsideration denied, 2019 WL 5692168 (N.D. Ind. Nov. 4, 2019) (consent is "a question common to the class only if similar evidence and methodology will suffice to answer the question for each member or the issue is susceptible to generalized, class-wide proof"). Instead, the evidence demonstrates that Essendant sought and obtained consent from members of the "fresh numbers" population, even if further narrowed to include only buying group members or a subset thereof. As such, class treatment remains inappropriate because "we do not know what kind of pre-existing arrangement may have existed between [Essendant] and the other fax recipients that [Craftwood] wants to represent" without inquiring into whether those recipients provided a consent form, or were marked as having consented in the Trend database or otherwise in Essendant's files. See Brodsky, 910 F.3d at 291. And because Essendant obtained consent orally and certain of its files were destroyed or underrepresent the consent obtained, even then the Court's review would not be complete. Assessing consent even for this narrowed putative class—no matter how defined—is thus "sufficiently individualized to preclude class certification." Sandusky, 863 F.3d at 468-69; see also G.M. Sign, Inc. v. Brink's Mfg. Co., 2011 WL 248511, at *9 (N.D. Ill. Jan. 25, 2011) (denying class certification where it "seems unavoidable that the Court would have to conduct a series of mini-trials to determine . . . consent").
Accordingly, the Court need not address the parties' arguments under Rule 23(a) and otherwise. See Priddy v. Health Care Serv. Corp., 870 F.3d 657, 660 (7th Cir. 2017) ("The plaintiff bears the burden of proving by a preponderance of the evidence all necessary prerequisites to the class action."). Class certification is improper.
Conclusion
In light of the minimal discovery Craftwood took in this case, it should come as no surprise that it once more fails to meet its burden to demonstrate that common questions predominate and class treatment is superior to individual litigation. The amended motion for class certification is denied. R. 79.
ENTERED:
/s/_________
Honorable Thomas M. Durkin
United States District Judge Dated: June 4, 2020