In the sprawling National Prescription Opiate Litigation (MDL 2804), proponents of a “negotiation class” recently asked the Sixth Circuit for en banc review of a September 2020 decision that struck down the novel class mechanism. Arguing that the 2-1 panel decision imposed an impermissibly rigid “top-down” interpretation of the Federal Rules, in conflict with Sixth Circuit and Supreme Court precedent, class proponents urged the full court to review and allow the negotiation class to proceed.
The opioid MDL is possibly one of the most complex cases in U.S. history. Pending in the Northern District of Ohio before Judge Dan A. Polster, the MDL encompasses over 2,000 cases filed by 33,000 city and county governments against drug manufacturers, distributors, and pharmacies to recoup costs of addressing the opioid crisis.
From the MDL’s outset, Judge Polster encouraged settlement. To that end, the court appointed the late Duke Law Professor Francis McGovern as special master to oversee settlement negotiations. Professor McGovern quickly found that settlement of such sprawling litigation presented intractable problems: How could defendants achieve “global peace”? And in a case with “heterogeneous” class members—some with big claims and many with small claims—how could the court encourage cooperation among plaintiffs and avoid hold-outs? Out of this came a novel idea: the negotiation class.
The Novel Negotiation Class
Working with Harvard Law Professor William Rubenstein, a court-appointed expert on class-certification issues, Professor McGovern developed and explained his idea for negotiation-class certification in a law review article and pitched the approach to a group of opioid MDL plaintiffs. Under the new negotiation class mechanism, plaintiffs would develop: (1) a plan to allocate any lump-sum settlement among class members; and (2) a plan to vote on settlement offers they might receive. They then would ask the court to certify a “negotiation class” according to the normal Rule 23 factors and to issue a class notice to explain the distribution and voting plans, giving class members the opportunity to opt out. After certification, plaintiffs would negotiate with defendants, and any settlement would be put to a binding vote by class members who had not opted out.
As Professor Rubenstein explains it, the proposition to plaintiffs is this: “We can’t tell you exactly how much you’re going to get. But we can tell you what percentage of the settlement you’ll get. And we can safeguard your rights by making sure when a settlement is finally achieved every class member will get to vote on the settlement.” Meanwhile, for defendants, the negotiation class sets the scope of the settling class prior to discussing settlement amounts, letting the defendants know how much peace a settlement actually buys them.
A group of opioid MDL plaintiffs pursued the idea and, in the summer of 2019, asked Judge Polster to approve a negotiation class, which he did in a September 2019 order. Judge Polster recognized that the negotiation class was novel. But he reasoned that Rule 23 does not explicitly limit the uses of class actions and “neither specifies that the class to be certified [must be] for ‘trial’ or ‘settlement’ purposes.” Rule 23’s history also supported new uses of the class mechanism, because, as Judge Polster explained, certification of settlement classes was itself a novel practice shortly after Rule 23 was adopted in 1966. In fact, Rule 23 contained no explicit reference to settlement classes until 2018, when amendments fleshed out settlement class procedures to better capture existing practices. And the Supreme Court had long accepted settlement classes as a “stock device” well before the 2018 amendments.
Judge Polster noted that, as a practical matter, the settlement class would only provide an option to facilitate settlement discussions, and would not stop any parties from continuing to litigate or pursuing their own settlement discussions, if they preferred. He found that “there is nothing coercive about this process” because “no Defendant has to employ it.” To be fair, however, defendants might have reasonably feared that even this novel form of certification could strengthen plaintiffs’ hand by laying groundwork for any eventual motion to certify classes for trial, should negotiations fall through.
Panel Review at the Sixth Circuit
The Sixth Circuit panel reviewing Judge Polster’s order had a less sanguine view of Rule 23’s flexibility. In September 2020, a 2-1 panel majority applied a textualist approach to reverse the certification order. The majority treated Rule 23 effectively as a statute (because the Rules are promulgated pursuant to the Rules Enabling Act) and purported to enforce it according to its plain terms. This, the court held, required striking down the negotiation class because the new form of class action was “untethered” to any express authorizing language in Rule 23. As for the argument that settlement class certification arose—and was judicially blessed—before Rule 23 ever said anything to authorize settlement classes, the majority disagreed. Prior to the 2018 amendments, Rule 23(e) had read:
A class action shall not be dismissed or compromised without the approval of the court, and notice of the proposed dismissal or compromise shall be given to all members of the class. . . .
Therefore, the Rule contemplated courts overseeing “compromised” class actions. And, for the majority, authority over “compromised” class actions implicitly authorized courts to certify classes for the purpose of settling them.
The rest of the majority opinion focused on the structure of Rule 23 and ways that the negotiation class did not fit the traditional analyses lower courts must apply to certify trial, settlement, or issue classes. In effect, Judge Polster had tried to shoehorn something new into existing federal common law rules that simply were not made for it.
The Panel Dissent
Sixth Circuit Judge Karen Nelson Moore dissented, presenting a dramatically different view of how the Federal Rules work by arguing that judges are not and have never been bound to strict textualist interpretations of the Rules, in large part because the Rules are not statutes and are not issued by Congress.
Walking through the “origin story [and] life cycle” of the Rules, the dissent highlighted instances where courts “drew on traditions of equity practice” and flexibly applied the Rules by “placing substance over form.” For instance, the Rules don’t expressly authorize courts to recall a jury for further deliberations, to hear a motion in limine, or to dismiss cases for forum non conveniens. But the Supreme Court blessed each of these practices as they arose through “district court creativity.” Other Rules were interpreted contrary to plain text and altered through court practice. For example, the Court fashioned the attorney work-product exception to Rule 26 discovery in opposition to the Rule’s plain text, until the Rule was amended in 1970. The Court modified Rule 56’s summary judgment standards in Celotex Corp. v. Catrett, 477 U.S. 317 (1986), to avoid a plain-text interpretation that guaranteed summary judgment rarely would be granted. (The modified summary-judgment standard would not be incorporated into Rule 56’s text until amendments in 2009).
Judge Moore argued that the same was true of Rule 23. She recited a host of articles and cases acknowledging that Rule 23, when adopted, “lacked any mention of ‘settlement,’ delineated no procedures to be followed for different kinds of class actions, and ‘contain[ed] no standards at all governing judicial approval of class action settlements.’” But, Judge Moore explained, “[a]s settlement class actions flooded the dockets” the courts of appeal quickly “recognize[d] that settlement classes had ‘utility,’ . . . ‘afford[ed] considerable economies to both the litigants and the judiciary[,] and [are] also fully consistent with the flexibility integral to Rule 23.’”
According to Judge Moore, the creation and certification of a novel negotiation class, particularly in a complex MDL, “simply follows the incremental development of settlement class actions” and is no less authorized under Rule 23 than settlement classes ever were.
En Banc Application
In October 2020, a group of plaintiffs filed for en banc review. Relying heavily on Judge Moore’s dissent, they argued that the panel majority had misconstrued both the history of Rule 23 and the flexibility district courts arguably enjoy in applying the Federal Rules. Plaintiffs questioned how the majority could use a strict textualist interpretation to find that a single use of the word “compromised” in the old version of Rule 23—a thin textual reed—had authorized settlement class certification. This was particularly notable when, as Judge Moore had explained in dissent, every prior court to have addressed the question found no textual authorization for settlement classes. Moreover, even if the majority was right, and settlement class certification could be derived from the term “compromise,” surely negotiation class certification could be derived similarly from the term “settlement,” because negotiation is always a critical part of settlement.
Whether the Sixth Circuit will take up the en banc petition remains to be seen. Given the size and complexity of the opioid MDL, and the novelty of the negotiation class issue, the petition likely meets the Sixth Circuit’s standard for rehearing matters of exceptional importance. En banc rehearing, however, is rare. Negotiation class certification could prove to be an inventive and useful tool in complex litigation, though it may have to wait for other litigants to try in another circuit.
 The Negotiation Class: A Cooperative Approach to Class Actions Involving Large Stakeholders, 99 TEX. L. REV. (forthcoming 2020), https://ssrn.com/abstract=3403834.
 Francis McGovern, Elizabeth Burch, and William Rubenstein, The Negotiation Class, Judicature, vol. 104 no. 1 (Spring 2020), https://judicature.duke.edu/articles/the-negotiation-class/.
 In re Nat’l Prescription Opiate Litig., 332 F.R.D. 532 (N.D. Ohio 2019).
 Id. at 539.
 Id. at 539–540.
 Amchem Prod., Inc. v. Windsor, 521 U.S. 591, 618 (1997).
 332 F.R.D. at 537.
 No. 19-4097, 2020 WL 5701916 (6th Cir. Sept. 24, 2020).
 Id. at *5 (citing Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 363 (2011) (mere negative inference in Rule 23’s text did not allow certification of (b)(2) class for money damages in addition to injunctive relief) and Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620 (1997) (“Courts are not free to amend a rule outside the process Congress ordered.”)).
 Id. at *6.
 No. 19-4097, 2020 WL 5701916, at *9 (6th Cir. Sept. 24, 2020) (Moore, J. dissenting).
 Id. at *10.
 Id. (citing Hickman v. Taylor, 329 U.S. 495, 505 (1947)).
 Id. at *13 (quoting William W Schwarzer, Settlement of Mass Tort Class Actions: Order Out of Chaos, 80 Cornell L. Rev. 837, 841 (1995)).
 Id. (quoting Amchem, 521 U.S. at 618 and In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 794 (3d Cir. 1995)).
 Petition for Rehearing En Banc at 10–11, Nos. 19-4097/19-4099 (6th Cir.), ECF No. 97.