[PUBLISH]
IN THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
________________________
No. 19-13242
________________________
D.C. Docket No. 1:16-cv-22482-RNS
TIMOTHY CHERRY, JILL GARRETT, et al.,
Plaintiffs-Appellants-Cross Appellees, versus
DOMETIC CORPORATION, Defendant-Appellee-Cross Appellant.
________________________
Appeals from the United States District Court for the Southern District of Florida
_______________________
(February 2, 2021)
Before WILLIAM PRYOR, Chief Judge, JORDAN and MARCUS, Circuit Judges. WILLIAM PRYOR, Chief Judge:
The main issue on appeal is whether putative class representatives must prove the existence of an administratively feasible method to identify absent class
2
members as a precondition for certification of a class action under Federal Rule of Civil Procedure 23. Owners of allegedly defective refrigerators manufactured by Dometic Corporation, as putative class representatives, moved to certify a class of similarly situated owners, but the district court denied certification based on their failure to prove administrative feasibility. The district court then dismissed the action because, in its view, the denial of class certification divested it of subject- matter jurisdiction. The district court erred. Because jurisdiction does not turn on the denial of class certification and Rule 23 provides no basis to require administrative feasibility, we vacate the order denying class certification and dismissing this action and remand for further proceedings.
I. BACKGROUND
Dometic Corporation manufactures and sells gas-absorption refrigerators that are used in recreational vehicles. Unlike regular refrigerators, Dometic refrigerators are designed to remain operable even when disconnected from electricity. They rely on a chemical solution that can be dangerous if it leaks.
Some Dometic refrigerators have a defect that exacerbates the risk of leakage and creates a risk of fire. In 2006 and 2008, Dometic initiated limited recalls to address this defect. It estimated that the defect affected one hundredth of one percent of the refrigerators it recalled.
3
The putative class representatives—18 owners of Dometic refrigerators—
argue that the defect is far more widespread. In their view, almost every refrigerator that Dometic sold between 1997 and 2016 has a design defect that corrodes the refrigerator’s boiler tubes. They allege that this defect has caused thousands of fires or leaks and that it gradually ruins the functionality of the refrigerators. They also allege that Dometic knew of but concealed these facts.
Based on these allegations, the putative class representatives sued Dometic for violations of the Magnuson-Moss Warranty Act and various state laws. They moved for class certification under Rule 23(b)(3). They proposed a class consisting of all persons who purchased in selected states certain models of Dometic refrigerators that were built since 1997.
The main issue at the class-certification stage was whether the proposed class satisfied the ascertainability requirement of Rule 23. The putative class representatives framed ascertainability as an issue of class definition and argued that “[t]he proposed class is ascertainable because the class definition relies exclusively on objective criteria.” They also argued that class-member identification would be administratively feasible, in any event. They supported their analysis with a citation to Briseno v. ConAgra Foods, Inc., a decision that rejects administrative feasibility as a prerequisite to certification. 844 F.3d 1121, 1132–33 (9th Cir. 2017). Dometic argued that ascertainability requires proof of
4
administrative feasibility. It contended that the class representatives failed to satisfy that element because they provided no evidence that their proposed method of identification would be workable.
The district court denied class certification because it agreed that the class representatives failed to prove administrative feasibility. It based this decision on one of our unpublished opinions, which stated that administrative feasibility is an element of the ascertainability requirement. See Karhu v. Vital Pharms., Inc., 621 F. App’x 945, 947–48 (11th Cir. 2015). The district court then determined that the denial of certification divested it of subject-matter jurisdiction under the Class Action Fairness Act, the only asserted basis for jurisdiction. So it dismissed the action without prejudice.
The class representatives appeal the denial and dismissal. They ask us to clarify that Rule 23 does not require proof of administrative feasibility. Dometic cross-appeals the dismissal. Several groups submitted briefs as amicus curiae regarding the validity of an administrative-feasibility requirement.
II. STANDARD OF REVIEW
We review jurisdictional issues de novo. AT&T Mobility, LLC v. NASCAR, Inc., 494 F.3d 1356, 1360 (11th Cir. 2007). We review the denial of class certification under Rule 23 for abuse of discretion. Little v. T-Mobile USA, Inc., 691 F.3d 1302, 1305 (11th Cir. 2012). Within that framework, we review factual
5
determinations for clear error and legal determinations de novo. Vega v. T-Mobile USA, Inc., 564 F.3d 1256, 1265 (11th Cir. 2009). A district court abuses its discretion “if it applies an incorrect legal standard” in its analysis of class certification. Little, 691 F.3d at 1305 (internal quotation marks omitted).
III. DISCUSSION
We divide our discussion in three parts. We first explain that the doctrines of invited error and forfeiture do not bar our review of the administrative-feasibility issue. We next review the role of administrative feasibility in class-certification analysis: it is not a requirement for certification, either as an element of ascertainability or otherwise. But a district court may consider administrative feasibility as one factor among several under Rule 23(b)(3). We last reiterate that jurisdiction under the Class Action Fairness Act does not turn on the availability of class certification.
A. The Doctrines of Invited Error and Forfeiture Do Not Bar Our Consideration of the Issue of Administrative Feasibility.
Dometic argues that we should not reach the merits because the putative class representatives either invited error as to the role of administrative feasibility or forfeited their challenge to that alleged requirement. We may not review an error if the “party induce[d] or invite[d] the district court into making [that] error.” United States v. Brannan, 562 F.3d 1300, 1306 (11th Cir. 2009) (internal quotation marks omitted). And we ordinarily do not consider an issue that could have been
6
raised before the district court but is not raised until appeal. Blue Martini Kendall, LLC v. Miami Dade County, 816 F.3d 1343, 1349 (11th Cir. 2016). But neither circumstance applies here.
There is no invited error because that doctrine is triggered only by unambiguous statements or representations. United States v. Hayes, 762 F.3d 1300, 1310 n.6 (11th Cir. 2014). Nowhere in its district-court filings did the putative class representatives concede that administrative feasibility is a requirement for class certification. Although their filings could have been clearer, we read them to argue that the proposed class is ascertainable because it is defined with objective criteria and then to address administrative feasibility only in the alternative. A fallback position does not invite error. See Yellow Pages Photos, Inc. v. Ziplocal, LP, 795 F.3d 1255, 1278 n.7 (11th Cir. 2015).
We also conclude that the putative class representatives preserved this issue for appeal. A party preserves an issue if it raises the issue “in such a way as to afford the district court an opportunity to recognize and rule on it.” Juris v. Inamed Corp., 685 F.3d 1294, 1325 (11th Cir. 2012) (internal quotation marks omitted). The class representatives cleared this hurdle because, in their motion for class certification, they cited Briseno v. ConAgra Foods, Inc., 844 F.3d at 1132, to support the proposition that their class is ascertainable. Briseno rejects administrative feasibility as a prerequisite to certification, id. at 1133, and the
7
putative class representatives cited a passage questioning why such a requirement would be necessary to protect the rights of a defendant, id. at 1132. That citation, together with their argument that objective criteria made the class ascertainable, should have put the district court on notice that the necessity of proving administrative feasibility was in dispute. Cf. Clark v. Wainwright, 701 F.2d 895, 897 (11th Cir. 1983).
B. Administrative Feasibility Is Relevant under Rule 23(b)(3) But It Is Not a Prerequisite for Certification.
Administrative feasibility is “[o]ne of the most hotly contested issues in class action practice today.” Rhonda Wasserman, Ascertainability: Prose, Policy, and Process, 50 Conn. L. Rev. 695, 697–99 (2018). Our sister circuits have debated whether administrative feasibility is a requirement for class certification. We conclude it is not a requirement. But it remains relevant to whether a proposed class may proceed under Rule 23(b)(3).
Ascertainability is an implied prerequisite of Rule 23. Little, 691 F.3d at 1304. Class representatives bear the burden to establish that their proposed class is “adequately defined and clearly ascertainable,” and they must satisfy this requirement before the district court can consider whether the class satisfies the enumerated prerequisites of Rule 23(a). Id. (internal quotation marks omitted).
Traditionally, we have collapsed class definition and ascertainability into one inquiry. See DeBremaecker v. Short, 433 F.2d 733, 734 (5th Cir. 1970). A
8
class is inadequately defined if it is defined through vague or subjective criteria. See id. And without an adequate definition for a proposed class, a district court will be unable to ascertain who belongs in it. Id.
The Third Circuit has applied a heightened standard for ascertainability. See Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015). Under this heightened standard, proof of ascertainability encompasses both the definition of a class and its administrative feasibility. Id. The latter requires putative class representatives to prove that the identification of class members will be “a manageable process that does not require much, if any, individual factual inquiry.” Carrera v. Bayer Corp., 727 F.3d 300, 307–08 (3d Cir. 2013) (internal quotation marks omitted).
Other circuits have split over the proper role of administrative feasibility.
Most have considered this issue in terms of ascertainability, but even circuits that do not require proof of ascertainability as a prerequisite to certification have considered whether to require administrative feasibility as a prerequisite. Briseno, 844 F.3d at 1124 n.4; Sandusky Wellness Ctr., LLC v. Medtox Sci., Inc., 821 F.3d 992, 995–96 (8th Cir. 2016). Like the Third Circuit, the First and Fourth Circuits require proof of administrative feasibility as a prerequisite for certification. In re Nexium Antitrust Litig., 777 F.3d 9, 19 (1st Cir. 2015); EQT Prod. Co. v. Adair, 764 F.3d 347, 358–59 (4th Cir. 2014). But the Second, Sixth, Seventh, Eighth, and Ninth Circuits reject that approach. In re Petrobas Sec., 862 F.3d 250, 267 (2d Cir.
9
2017); Rikos v. Procter & Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015); Mullins v. Direct Digit., LLC, 795 F.3d 654, 662 (7th Cir. 2015); Sandusky Wellness Ctr., 821 F.3d at 995–96; Briseno, 844 F.3d at 1123; see also Seeligson v. Devon Energy Prod. Co., 761 F. App’x 329, 334 (5th Cir. 2019). We have addressed the issue only in unpublished decisions that applied the heightened standard of the Third Circuit, Karhu, 621 F. App’x at 946; Bussey v. Macon Cnty. Greyhound Park, Inc., 562 F. App’x 782, 787 (11th Cir. 2014), but those decisions do not bind us as precedent, 11th Cir. R. 36-2.
To resolve the issue, we ask whether either circuit precedent or the text of Rule 23 establishes administrative feasibility as a requirement for class certification. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620 (1997); In re Lambrix, 776 F.3d 789, 794 (11th Cir. 2015). We conclude that neither does. Proof of administrative feasibility cannot be a precondition for certification.
Circuit precedent establishes that a district court must determine that a proposed class is “adequately defined and clearly ascertainable” before it may consider whether the requirements of Rule 23(a) are satisfied. Little, 691 F.3d at 1304 (internal quotation marks omitted). This rule follows from the text of Rule 23(a). Although no form of the word “ascertainability” appears in the rule, the text includes “what is implicit.” Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts § 8, at 96 (2012). And ascertainability—at least as
10
traditionally understood—is an implied prerequisite to the requirements of Rule 23(a). See William B. Rubenstein, 1 Newberg on Class Actions § 3:2, at 155–56 (5th ed. 2011).
Without an adequate class definition, a district court would be unable to evaluate whether a proposed class satisfies Rule 23(a). For example, if a class is defined in terms so vague as to be indeterminate, then a district court lacks a way to assess whether “there are questions of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2); cf. DeBremaecker, 433 F.2d at 734. A clear definition too is necessary for a district court to determine whether “the class is so numerous that joinder of all members is impracticable,” Fed. R. Civ. P. 23(a)(1), and whether “the claims or defenses of the representative parties are typical of the claims or defenses of the class,” id. R. 23(a)(3). So a district court must decide that a class is ascertainable before it may turn to the requirements of Rule 23(a). Little, 691 F.3d at 1304.
Our ascertainability precedents though do not mandate proof of administrative feasibility. A class is “clearly ascertainable” if we are certain that its membership is “capable of being” determined. Ascertain, Webster’s New International Dictionary (3d ed. 1993); Ascertainable, Webster’s New International Dictionary (3d ed. 1993). But membership can be capable of
11
determination without being capable of convenient determination. Administrative feasibility is not an inherent aspect of ascertainability.
Because no precedent governs, this appeal turns on whether the text of Rule 23(a) or (b) necessarily requires proof of administrative feasibility. See Sargeant v. Hall, 951 F.3d 1280, 1283 (11th Cir. 2020). The Supreme Court has made clear that district courts must grant class certification in “each and every case” where the conditions of Rule 23(a) and (b) are met. Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co., 559 U.S. 393, 398–400 (2010) (internal quotation marks omitted). We lack discretion to add requirements to the Rule.
Administrative feasibility does not follow from the text of Rule 23(a).
Unlike traditional ascertainability, administrative feasibility does not bear on the ability of a district court to consider the enumerated elements of that subsection. A plaintiff proves administrative feasibility by explaining how the district court can locate the remainder of the class after certification. See, e.g., In re Cmty. Bank of N. Va. Mortg. Lending Pracs. Litig., 795 F.3d 380, 396–97 (3d Cir. 2015). The plaintiff satisfies this requirement if the district court concludes that the proposed process will be manageable and successful. Byrd, 784 F.3d at 163–64. But neither foreknowledge of a method of identification nor confirmation of its manageability says anything about the qualifications of the putative class representatives, the practicability of joinder of all members, or the existence of common questions of
12
law or fact. Fed. R. Civ. P. 23(a). Because administrative feasibility has no connection to Rule 23(a), it is not part of the ascertainability inquiry.
Nor does a requirement of administrative feasibility follow from Rule 23(b).
To be sure, administrative feasibility has relevance for Rule 23(b)(3) classes, in the light of the manageability criterion of Rule 23(b)(3)(D). See Rubenstein, 2 Newberg on Class Actions § 4:76, at 301 (5th ed. 2012). Rule 23(b)(3)(D) instructs the district court, in deciding whether “a class action [would be] superior to other available methods for fairly and efficiently adjudicating the controversy,” to consider “the likely difficulties in managing a class action.” Fed. R. Civ. P. 23(b)(3). A difficulty in identifying class members is a difficulty in managing a class action. See Briseno, 844 F.3d at 1126. But because Rule 23(b)(3) requires a balancing test, it does not permit district courts to make administrative feasibility a requirement. The manageability inquiry focuses on whether a class action “will create relatively more management problems than any of the alternatives,” not whether it will create manageability problems in an absolute sense. Klay v. Humana, Inc., 382 F.3d 1241, 1273 (11th Cir. 2004), abrogated in part on other grounds by Bridge v. Phoenix Bond & Indem. Co., 553 U.S. 639 (2008). And the district court must balance its manageability finding against other considerations. Fed. R. Civ. P. 23(b)(3). So administrative difficulties—whether in class-member identification or otherwise—do not alone doom a motion for certification. Indeed,
13
we have made clear that manageability problems will “rarely, if ever, be in [themselves] sufficient to prevent certification.” Klay, 382 F.3d at 1272.
We are unpersuaded by the other arguments raised by Dometic and by the United States Chamber of Commerce as amicus curiae. Dometic distinguishes Klay by arguing that the manageability criterion of Rule 23(b)(3)(D) concerns only problems that arise after class members have been identified, while administrative feasibility addresses problems in the identification process. But Rule 23(b)(3)(D) contains no such limitation, Briseno, 844 F.3d at 1126, and Dometic’s argument only underscores the extratextual nature of the administrative-feasibility requirement: aside from its limited relevance to Rule 23(b)(3)(D), administrative feasibility is entirely unrelated to either Rule 23(a) or (b). The Chamber argues that the requirement flows from Rule 23(c), but Shady Grove establishes that the only requirements for certification are those in Rule 23(a) and (b). 559 U.S. at 398–99. Rule 23(c) concerns only the content of a certification order and the tasks that arise after certification; it imposes no prerequisite for certification. See Fed. R. Civ. P. 23(c). And because we are concerned with the text of Rule 23, we do not consider arguments about the policy merits of administrative feasibility.
We hold that administrative feasibility is not a requirement for certification under Rule 23. In doing so, we limit ascertainability to its traditional scope: a proposed class is ascertainable if it is adequately defined such that its membership
14
is capable of determination. Cf. DeBremaecker, 433 F.2d at 734. Our decision might render redundant the phrase in our precedent that a proposed class must be “adequately defined and clearly ascertainable.” Id. But “[d]oublets . . . abound in legalese,” Scalia & Garner, Reading Law § 26, at 177, and this one is required by Rule 23.
If a district court reaches Rule 23(b), and the action involves a proposed Rule 23(b)(3) class, it may consider administrative feasibility as part of the manageability criterion of Rule 23(b)(3)(D). See, e.g., Briseno, 844 F.3d at 1127– 28. If there “appear to be unusually difficult manageability problems” at this step, a “district court[] ha[s] discretion to insist on details of the plaintiff’s plan for notifying the class and managing the action.” Mullins, 795 F.3d at 664. And it has discretion to decertify a certified class that turns out to be unmanageable. Id.
A district court must evaluate this issue in comparative terms; that is, because the superiority requirement of Rule 23(b)(3) turns on whether a class action is better than other available methods of adjudication, Fed. R. Civ. P. 23(b)(3), it involves two forms of comparison. First, would a class action create more manageability problems than its alternatives? Klay, 382 F.3d at 1273. And second, how do the manageability concerns compare with the other advantages or disadvantages of a class action? See id. Administrative feasibility alone will rarely,
15
if ever, be dispositive, id. at 1272–73, but its significance will depend on the facts of each case.
C. Jurisdiction Under the Class Action Fairness Act Does Not Depend on Class Certification.
Even if the decision of the district court to deny class certification were
correct, its dismissal of this action still would be incorrect. “[F]ederal jurisdiction under the Class Action Fairness Act does not depend on certification,” so a district court retains jurisdiction even after it denies certification. Wright Transp., Inc. v. Pilot Corp., 841 F.3d 1266, 1271 (11th Cir. 2016) (internal quotation marks omitted). Whether as a class or otherwise, the parties may proceed in federal court.
IV. CONCLUSION
We VACATE the dismissal of this action and denial of class certification and REMAND for further proceedings consistent with this opinion.