District courts typically grant or deny class certification on a claim-by-claim basis.  If a plaintiff can’t meet the Rule 23 requirements for a particular claim, then class certification is denied. See Fed. R. Civ. P. 23(a)–(b). Once in a while, however, district courts will certify a class on an element of a claim, which, when viewed in isolation, satisfies Rule 23’s requirements. This leaves the remaining elements, which do not meet Rule 23’s requirements, to be resolved in individualized litigation.  

The authority to partially certify a claim for class treatment comes from Rule 23(c)(4), which, “[w]hen appropriate,” permits “an action [to] be brought or maintained as a class action with respect to particular issues.” For example, courts will sometimes certify a class on liability, and leave damages for individualized determination. There would then be a class-wide trial on liability, followed by individualized trials––perhaps hundreds or thousands of them––on damages. The majority of circuits have also interpreted Rule 23(c)(4) to permit class certification on fewer than all liability elements of a cause of action.  At least one circuit, however, essentially prohibits certifying a class on an issue that will not resolve liability.

Defense-oriented commentary has generally characterized issue-class certification as a snare for class-action defendants—one that should be avoided without fail, and ideally written out of Rule 23. See, e.g., Terence Hawley, “Third Circuit takes issue with certification of Rule 23(c)(4) issue class,” Viewpoints (Sept. 30, 2021); Michael R. Pennington & Scott Burnett Smith, “The Growing Split Over Issue Class Certification as an End Run Around Predominance of Common Issues,” Declassified (July 26, 2018). But could the availability of issue-class certification, in some cases, actually help class-action defendants, by preventing full class certification and showcasing to class counsel the arduous road ahead for obtaining relief for the class?

We explore that possibility—with a look, first, at the circuit split on issue-class certification, then at the Third Circuit’s recent decision in Russell v. Educational Commission for Foreign Medical Graduates, No. 20-2128 (3rd Cir. Sept. 24, 2021), and finally at a 2012 issue-class certification in an employment-discrimination class action out of the Seventh Circuit.

The Circuit Split

The circuits have largely split around whether Rule 23(b)(3)’s superiority and predominance requirements apply before or after dividing a claim into issues. 

The majority of circuits to decide the issue, including the Second, Sixth, Seventh, and Ninth, permit class certification on fewer than all elements of a cause of action—provided the issues to be certified satisfy both the Rule 23(a) prerequisites and Rule 23(b)(3)’s predominance and superiority requirements. For example, in Martin v. Behr Dayton Thermal Products, LLC, the Sixth Circuit reasoned that “Rule 23(c)(4) contemplates using issue certification to retain a case’s class character where common questions predominate within certain issues and where class treatment of those issues is the superior method of resolution.”  896 F.3d 405, 413 (6th Cir. 2018). 

The Fifth Circuit is the outlier. It has suggested that issue-class certification is never “appropriate” under Rule 23(c)(4) on fewer than all elements of a cause of action, and, in an oft-quoted footnote, cautioned against “manufactur[ing] predominance through the nimble use of subdivision (c)(4).” Castano v. Am. Tobacco Co., 84 F.3d 734, 745 n.21 (5th Cir. 1996). According to the Fifth Circuit, such slicing and dicing of issues would “eviscerate the predominance requirement,” leading to wanton “certification in every case.” Id.    

The Third Circuit, generally aligned with the majority, has taken a Goldilocks-type approach. In Gates v. Rohm & Haas Co., the Court identified a number of factors for determining when issue-class certification is appropriate.  655 F.3d 255, 273 (3rd Cir. 2011). The Gates factors—which help courts to “treat common things in common and to distinguish the distinguishable”—include, among others, the type of claim involved, efficiencies to be gained by granting partial certification, and the potential preclusive effect (or lack thereof) of the proposed issue class.  Id. 

Russell Vacates, but Tacitly Encourages, Issue Classes

The Third Circuit reinforced the Gates approach in Russell. Russell involved claims by a class of former patients of a foreign-born doctor who fraudulently obtained his license to practice medicine in the United States. Russell, Slip Op. at 9.  The putative class members asserted claims of negligent infliction of emotional distress, not against the doctor himself, but against the Educational Commission for Foreign Medical Graduates, a non-profit that certified him to practice medicine in the United States. Id. The named plaintiffs, who were allegedly injured when the doctor performed unplanned emergency cesarean-section surgery to deliver their children, sought to hold the Commission responsible for injuries caused to themselves and a class of “similarly situated” individuals who received treatment from the doctor. Id.

Unsurprisingly, the district court concluded that class certification was inappropriate to determine whether and to what extent emotional distress was inflicted on each putative class member. Id. at 10. However, it certified a class on the issues of whether the Commission owed a duty to the class members to better investigate the “doctor,” and whether those duties were breached when the Commission failed to prevent his fraud—leaving for individualized resolution the questions of whether any class member suffered emotional distress, the Commission’s role in causing the alleged distress, and damages.  Id.

The Third Circuit vacated the issue-class certifications, clarifying the Circuit’s process-driven approach for Rule 23(c)(4) certification, and remanded for application of that approach. Id. at 32. 

The Third Circuit clarified that “Rule 23(a) and Rule 23(b) decide if the proposed issues can be brought or maintained as class action, while the Gates factors determine whether they should.” Id. at 22. Thus, district courts considering issue-class certification in the Third Circuit must ask three questions:

(1)      Does the proposed issue class satisfy the four prerequisites to class certification in

  Rule 23(a)?

(2)      Does the proposed issue class fit within a Rule 23(b) category?

(3)      If so, is certifying the issue class appropriate under the Gates factors?

Id. at 20–21. Only if the answer to all three questions is yes may an issue class be certified. 

According to the Third Circuit, the district court in Russell erred by not performing any Rule 23(b) analysis whatsoever. Id. at 23. The district court also failed to “rigorously consider” several of the Gates factors. Id. at 24. For example, while the district court “briefly discussed” the benefits of a single trial, it did not consider whether any real efficiencies would be created by certifying issue classes on duty and breach, where proximate cause, damages, affirmative defenses, and emotional distress would still have to be resolved in individual proceedings. Id. at 25–26. 

Nonetheless, while the Russell Court vacated the district court’s issue-class certification on the duty and breach elements, it expressly instructed that issues certified for class treatment “need not be limited to those that decide a party’s liability,” so long as the Rule 23(a) and (b) requirements and Gates factors are met. Id. at 21. Russell thus provides some encouragement to Third Circuit plaintiffs seeking to pursue class actions on issues that will not resolve a defendant’s liability.

A Tool for Class-Action Defense?

Returning to where we started, do class-action defendants have any use for issue class certification? The Russell court noted that, in most cases, the practical effect of issue-class certification is not to leave individualized issues for trial, but instead to guide settlement discussions. Russell, Slip Op. at 20. The same could likely be said for certification rulings in general. Could this mean that, in some situations, certification of an issue class can help class action defendants?

Maybe—considering that courts may be inclined to grant class certification, in some fashion or another, if they sense some merit to a putative class action. Courts may be inclined to grant certification in such a case, even if there are strong arguments against the putative class representative’s ability to fully satisfy Rule 23’s requirements on all of the liability elements. If the only type of certification available is certification on all liability elements, courts may find a way to gloss over the Rule 23 shortcomings and grant such certification anyway. 

In the majority of circuits, however, district courts have the option to certify a class on only those liability issues for which Rule 23’s requirements are fully met, and to leave the remaining issues for individualized resolution. In these circuits, defendants face three possibilities: 

(A)     Class certification on liability as a whole, meaning the defendant faces liability for the entire claim on a class-wide basis;

(B)     Class certification on some, but not all liability elements, meaning the defendant faces a preclusive judgment against the class on the issues for which certification is granted, and defense of whatever individualized proceedings may be brought for the remaining liability issues; or

(C)     No class certification at all, and thus no risk of class-wide liability of any kind.

Certainly, option (C) is preferable for any defendant. And that is what the defendant in Russell may ultimately obtain.  But Russell, with its claim of negligent infliction of emotional distress by a third-party licensing organization, is not exactly a prime case for class treatment in any context. In closer cases, if the prospect of certification is likely looming—meaning (A) and (B) are the only options—could it help defendants to hint at issue-class certification only on some elements of liability, as an alternative to class certification on liability as a whole? 

Class certification is typically a major victory for the class and its counsel. It means they will have the chance to get a class-wide judgment and class-wide damages, a significant portion of which counsel will likely receive as attorneys’ fees. A traditional class-certification ruling thus puts the class in a favorable settlement position. 

On the other hand, an order prohibiting a class-wide trial of a significant part of the claim is a much less decisive victory for the class. It means that, to recover, each plaintiff must still prevail in her own individualized trial on the remaining issues. Getting a class certified on only one of several necessary elements of the claim may make the class and counsel willing to settle for far less than if they had gotten full class certification. 

Even if the class had gotten full certification, and an objective observer would realize that it will still be exceedingly difficult to prove some elements of the claim with common evidence (such as causation and damages), class counsel will likely feel more confident in bargaining position. Unlike full certification, issue certification arguably provides little more than an open acknowledgement of the challenges ahead for judgment in each class-member’s favor. Defendants, then are in a better position in settlement negotiations with only an issue class certified.

Additionally, issue-class certification may allow a defendant to efficiently defeat claims of the entire class. As the Seventh Circuit has observed, “Defendants spend much time and money fighting Rule 23 certifications to the hilt. Yet certification is largely independent of the merits…and a certified class can go down in flames on the merits. We say this…to remind defendants that the class-action glass is sometimes half-full: dismissed claims of a certified class end litigation once and for all.” Beaton v. SpeedyPC Software, 907 F.3d 1018, 1031 (7th Cir. 2018) (quotation omitted). In Russell, for example, if the Commission owed no duty to the plaintiffs, then certifying the duty issue alone could allow defeating all claims by the entire class, and only require litigating one narrow element. 

Of course, to obtain issue-class certification, a plaintiff must show that permitting a class action on less than the full claim creates efficiencies for resolution of the entire dispute. So issue-class certification may be available only where there is a manageable way to individually resolve the un-certified issues. Nonetheless, defendants may, in some cases, be well served by not contesting this requirement, as a way of suggesting limited certification of an issue class only on some elements of liability, rather than certification of all.

Take, for example, McReynolds v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 672 F.3d 482 (7th Cir. 2012), which permitted class certification on whether Merrill Lynch had engaged in workplace practices with a disparate impact on black traders, but not on other aspects of the racial discrimination claims at issue—including causation, damages, and intentional discrimination. Not long after the issue-class certification ruling, the case settled. 

The settlement agreement highlighted that, even if plaintiffs prevailed at trial on the certified issue, “the process for resolving each Settlement Class Member’s claim for individual damages would then be separately determined,” “[e]ach Settlement Class Member might have been required to prove his or her right to recovery on an individual basis,” and “the Parties would also have to litigate hundreds of separate individual trials regarding the intentional discrimination claims, which were not certified for class treatment.” No. 1:05-cv-06859, DE 585-1 at 18 (N.D. Ill. Aug. 28, 2013). 

McReynolds settled for $160 million, with a potential six-figure recovery for each of the 700 class members. See id. at 27–41. Not shabby. But would a settlement have required more if the plaintiffs had gotten certification of the full claim? Would they then have had more confidence (even if false confidence) in the prospect of a class-wide money judgment, instead of an issue class that required them to individually prove causation, damages, and intentionality?

The parties to that case may be the only ones who know. And certification of the full claim may not have been an option in McReynolds, anyway, where the Seventh Circuit cautioned that class certification would have been inappropriate on causation and damages. 672 F.3d at 490–91.  But, even so, the possibility of an issue class as an alternative to full class certification, and that an issue-class alternative may place downward pressure on settlement, suggests that, for class-action defendants, the idea may be worth some thought.

Click here to subscribe to our monthly “Best In Class” blog. 

**Ellis & Winters’s Best in Class Blog covers developments in class actions in courts across the United States, focusing on insights for class-action defense.**