In re Nat'l Prescription Opiate Litig.

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 20a0311p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

IN RE: NATIONAL PRESCRIPTION OPIATE LITIGATION. ┐ ___________________________________________ │ │ ALBANY COUNTY, NEW YORK, Negotiation Class’s │ Class Representatives; CO-LEAD NEGOTIATION CLASS │ > Nos. 19-4097/4099 COUNSEL; CO-NEGOTIATION CLASS COUNSEL, │ Plaintiffs-Appellees, │ │ CITY OF NORTH ROYALTON, OHIO; CITY OF EAST │ CLEVELAND, OHIO; CITY OF MAYFIELD HEIGHTS, │ OHIO; CITY OF LYNDHURST, OHIO; CITY OF HURON, │ OHIO; CITY OF WICKLIFFE, OHIO, │ │ Plaintiffs-Appellants (19-4099), │ │ v. │ │ MCKESSON CORPORATION; CARDINAL HEALTH, INC.; │ AMERISOURCEBERGEN DRUG CORPORATION; │ PRESCRIPTION SUPPLY, INC.; DISCOUNT DRUG MART, │ INC.; WALMART, INC.; WALGREEN COMPANY; │ WALGREEN EASTERN CO., INC.; CVS PHARMACY, INC.; │ CVS INDIANA, LLC; CVS Rx SERVICES, INC.; RITE │ AID OF MARYLAND, INC., dba Rite Aid of Mid-Atlantic │ Customer Support Center, │ │ Defendants-Appellants (19-4097). │ ┘

Appeal from the United States District Court for the Northern District of Ohio at Cleveland. Nos. 1:17-md-02804; 1:18-op-45090—Dan A. Polster, District Judge.

Argued: July 28, 2020

Decided and Filed: September 24, 2020

Before: MOORE, CLAY, and McKEAGUE, Circuit Judges. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 2

_________________

COUNSEL

ARGUED: Sonya D. Winner, COVINGTON & BURLING LLP, San Francisco, California, for Appellants in 19-4097. Kevin K. Russell, GOLDSTEIN & RUSSELL, P.C., Bethesda, Maryland, for Appellants in 19-4099. Samuel Issacharoff, New York, New York, for Appellees. ON BRIEF: Sonya D. Winner, COVINGTON & BURLING LLP, San Francisco, California, Beth S. Brinkman, Mark H. Lynch, COVINGTON & BURLING LLP, Washington, D.C., Ashley W. Hardin, WILLIAMS & CONNOLLY LLP, Washington, D.C., Kim M. Watterson, REED SMITH LLP, Los Angeles, California, Tina M. Tabacchi, Tara A. Fumerton, JONES DAY, Chicago, Illinois, Benjamin C. Mizer, JONES DAY, Washington, D.C., Kaspar J. Stoffelmayr, BARTLIT BECK LLP, Chicago, Illinois, Kelly A. Moore, MORGAN, LEWIS & BOCKIUS LLP, New York, New York, Elisa P. McEnroe, MORGAN, LEWIS & BOCKIUS LLP, Philadelphia, Pennsylvania, Timothy D. Johnson, CAVITCH, FAMILO & DURKIN CO., LPA, Cleveland, Ohio, Alexandra W. Miller, ZUCKERMAN SPAEDER LLP, Washington, D.C., John J. Haggerty, FOX ROTHSCHILD LLP, Warrington, Pennsylvania, for Appellants in 19-4097. Kevin K. Russell, Thomas C. Goldstein, Eric F. Citron, GOLDSTEIN & RUSSELL, P.C., Bethesda, Maryland, for Appellants in 19-4099. Samuel Issacharoff, New York, New York, Christopher A. Seeger, Diogenes P. Kekatos, Jennifer R. Scullion, SEEGER WEISS LLP, Ridgefield Park, New Jersey, Robert Klonoff, Portland, Oregon, for Appellees. Mary Massaron, PLUNKETT COONEY, Bloomfield Hills, Michigan, Oramel H. (O.H.) Skinner, OFFICE OF THE ARIZONA ATTORNEY GENERAL, Phoenix, Arizona, Mikal C. Watts, Shelly A. Sanford, Meredith Drukker Stratigopoulos, WATTS GUERRA LLP, Austin, Texas, Dennis C. Reich, Shreedhar R. Patel, REICH & BINSTOCK, LLP, Houston, Texas, Jeff Reeves, Cheryl Priest Ainsworth, Kevin N. Royer, THEODORA ORINGHER PC, Costa Mesa, California, Marc J. Bern, MARC J. BERN & PARTNERS LLP, New York, New York, Robert T. Eglet, Robert M. Adams, Erica D. Entsminger, EGLET ADAMS, Las Vegas, Nevada, Jennifer Scifo, KRALOVEC, JAMBOIS & SCHWARTZ, Chicago, Illinois, John B. White, Jr., Marghretta H. Shisko, HARRISON WHITE, P.C., Spartanburg, South Carolina, Andrew J. Pincus, MAYER BROWN LLP, Washington, D.C., for Amici Curiae.

CLAY, J., delivered the opinion of the court in which McKEAGUE, J., joined. MOORE, J. (pp. 18–59), delivered a separate dissenting opinion.

OPINION _________________

CLAY, Circuit Judge. The latest appeal in this multi-district litigation (“MDL”) relating to the opioid crisis centers on the district court’s order certifying a “negotiation class” under Federal Rule of Civil Procedure 23. The court has certified a class of all cities and counties throughout the United States for purposes of negotiating a settlement between class members and Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 3

opioid manufacturers, distributors, and pharmacies. Appellants, objecting opioid distributors and retail pharmacies (“Defendants”), as well as six objecting Ohio cities, appeal the district court’s order certifying this negotiation class. Appellees, putative representatives of the negotiation class (“Plaintiffs”), request us to approve this novel form of class action. For the reasons provided below, we decline to do so, and therefore REVERSE the district court’s order.

BACKGROUND

The national prescription opioid MDL, consolidated in the Northern District of Ohio, consists of over 1,300 public-entity-led lawsuits, primarily filed by cities and counties. In re Nat’l Prescription Opiate Litig., 927 F.3d 919, 923 (6th Cir. 2019). These cities and counties allege that opioid manufacturers, opioid distributors, and opioid-selling pharmacies and retailers acted in concert to mislead medical professionals into prescribing, and millions of Americans into taking and often becoming addicted to, opiates. In re Nat’l Prescription Opiate Litig., 2018 WL 4895856, at *2–5, *36–37 (N.D. Ohio Oct. 5, 2018), report and recommendation adopted in part and rejected in part, 2018 WL 6628898 (N.D. Ohio Dec. 19, 2018). Plaintiffs allege that approximately 350,000 individuals in the United States died from an opioid overdose between 1999 and 2016. Id. at *2. The opioid industry’s practices allegedly harmed Plaintiffs by forcing them to divert significant funding to emergency public health and public safety responses to the crisis. Id. at *37. Plaintiffs across the MDL allege numerous causes of action, including claims based upon the federal Racketeer Influenced and Corrupt Organizations Act (“RICO”), 18 U.S.C. § 1961 et seq., its state analogues, state statutory public nuisance law, and several other state common law claims. Id. at *5, *31, *35, *41, *44, *46.

Throughout its time overseeing the MDL, the district court has repeatedly sought to facilitate settlements between the parties. See, e.g., In re Nat’l Prescription Opiate Litig., 332 F.R.D. 532, 536 (N.D. Ohio 2019) (“From the outset of this MDL, the Court has encouraged the parties to settle the case.”); R. 800, Op. & Order, PageID # 18971 (noting that the court’s order will assist “in litigating (and hopefully settling) these cases”).1 Plaintiffs aligned themselves with that goal when, on June 14, 2019, attorneys representing fifty-one cities and

1 Unless otherwise stated, all citations to the record refer to Case No. 1:17-md-02804-DAP. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 4

counties moved to certify a negotiation class under Federal Rule of Civil Procedure 23(b)(3). In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 537. Plaintiffs sought to include within the class every city and county within the United States, some 34,458 identified municipal entities, id. at 543, unless a prospective class member exercised its right to opt-out within sixty days of class certification. Id. at 539.

The idea for a negotiation class was developed by Professor Francis E. McGovern, a special master appointed to aid the district court in the MDL, in collaboration with Professor William B. Rubenstein. See Francis E. McGovern & William B. Rubenstein, The Negotiation Class: A Cooperative Approach to Large Claim Class Actions (Duke Law Sch. Pub. Law & Legal Theory Series, Paper No. 2019-41, 2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3403834. The primary difference between a negotiation class and a litigation class is that rather than forming a class to aggregate and try common issues, as with a litigation class, the negotiation class is designed to attempt to reach a settlement while the individual MDL cases continue along their respective litigation paths. In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 539.

The difference between a negotiation class and a settlement class is that in the case of a negotiation class, the class certification and opt-out process occur prior to a settlement being reached. Id. at 537. This is to “fix a class size and provide the Defendants a sense of the precise scope of the group with whom they are negotiating,” before settlement talks begin. Id. The court indicated that class members would likely not have a second opportunity to opt-out; they must decide at the negotiation class certification stage—without knowing the settlement figure— whether they wish to “bind themselves to a negotiation process, from which they will receive a known portion of the outcome and in which they will get a right to vote on the settlement.” Id. at 540.2

Despite not knowing the terms of the proposed settlement before deciding to opt-out of the class, putative class members’ rights are allegedly protected in two ways. For one, Plaintiffs developed a county-level formula allocating any eventual settlement with Defendants.

2 The district court did note that it “always retains the option of enabling a second opt-out opportunity if circumstances require.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 540; see also Fed. R. Civ. 23(e)(4). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 5

Prospective class members could use the formula (available online) to ascertain what percentage of a settlement their county could expect to receive.3 Id. at 537. This is meant to give prospective class members sufficient information to determine if they stand to gain access to a large enough percentage of any possible settlement to justify remaining in the negotiation class. Second, a proposed agreement could only be accepted by the class if a supermajority (i.e., 75%) of six categories of voting class members assent to it. Id. If a settlement receives supermajority support, then the parties may move for judicial approval of the agreement. The district court would then need to determine that the proposal is “fair, reasonable, and adequate” in accordance with Rule 23(e). This process is supposed to eliminate one common obstacle to settlement in cases involving classes certified under 23(b)(3): the uncertainty defendants have with respect to whether enough putative class members will not opt-out of a settlement class after a monetary figure is reached.4 From a defendant’s perspective, if too many class members opt-out, then the size of the settlement class may no longer justify the cost of the agreement because the defendant will remain exposed to litigation by the opt-out members. However, in the case of this negotiation class, as long as the settlement is approved by a supermajority of the class, every member—including those who oppose the settlement—will be bound by it.

Several distributor and pharmacy defendants objected to Plaintiffs’ motion, as did six putative city class members, thirty-seven state Attorneys General, and the Attorneys General of the District of Columbia and Guam. In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 537– 38. On September 11, 2019, the district court disregarded these complaints and certified the class. The court found that:

[T]here is nothing coercive about this process: no Defendant has to employ it. There is nothing exclusive about this process: it does not interfere with the States settling their own cases any way they want, and it does not stop parties in the MDL from settling in other ways. And there is nothing intrusive about this process: it does not stop any litigation from continuing and in no way interferes

3 The formula does not provide for a city-level allocation. Instead, the cities would need to negotiate with their respective counties for a portion of the county’s allocation. See In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 538. 4 Pursuant to Rule 23, only members of classes certified under Rule 23(b)(3), rather than (b)(1) or (b)(2), are entitled to notice of class certification and an opportunity to opt-out of the class. Fed. R. Civ. P. 23(c)(2)(A), (B). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 6

with the upcoming bellwether trials in this MDL. This process simply provides an option – and in the Court’s opinion, it is a powerful, creative, and helpful one.

Id. at 537. The district court also found that Plaintiffs’ federal RICO and Controlled Substances Act (“CSA”) claims, 21 U.S.C. § 801 et seq., which were brought by many Plaintiffs in the MDL, satisfied the threshold requirements of Rule 23(a) as well as the predominance and superiority requirements of Rule 23(b)(3). Id. at 548–51. On January 10, 2020, Plaintiffs informed the district court that only 556 of the 34,458 putative class members opted-out of the negotiation class.

The aggrieved Defendants and putative classes members then filed motions in this Court for permission to file an appeal of the class certification order under Federal Rule of Civil Procedure 23(f). See No. 19-0305, Doc. # 1 (motion by the objecting distributor and pharmacy defendants); No. 19-0306, Doc. # 1 (motion by the six objecting putative class member cities). Plaintiffs filed an opposition to these motions, but a motions panel of this Court rejected their arguments and granted the motions for permission to appeal. This consolidated appeal followed.

DISCUSSION

Plaintiffs first contend that this appeal was improvidently granted because the district court did not reach a final decision suitable for appellate review. However, we already rejected this argument when Plaintiffs raised it before the motions panel in opposition to Defendants’ petition under Rule 23(f) to appeal the district court’s certification order. The motions panel concluded that “the district court entered a final order to certify the class, with no indication that it will review its decision in the future.” See No. 19-0305, Doc. 24; No. 19-0306, Doc. 22 (Guy, Griffin, Kethledge). We are not inclined to disregard that decision. See Kraus v. Taylor, 715 F.3d 589, 594 (6th Cir. 2013) (stating that “this court’s interlocutory orders are rarely altered as a practical matter,” although they may be changed); Wallace v. FedEx Corp., 764 F.3d 571, 583 (6th Cir. 2014) (holding that “[l]ater panels cannot simply choose to disregard” decisions by the motions panel).

In any event, Plaintiffs’ argument on this point is without merit. Rule 23(f) provides that “[a] court of appeals may permit an appeal from an order granting . . . class-action certification Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 7

under this rule.” This is precisely what the district court intended to do through its order. It “certifie[d] a Negotiation Class on the claims and issues identified, against the Defendants identified, and appoint[ed] Class Counsel.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 556. It authorized the class to “negotiate settlements with any of the 13 sets of Defendants identified herein, on any of the claims or issues identified here, or those arising out of a common factual predicate.” Id. This situation is unlike the settlement class cases cited by Plaintiffs, in which courts rejected as premature appeals of a court’s “conditional” certification of a settlement class for case management purposes (such as to provide notice to absent class members), reserving actual certification for a later date. See, e.g., In re Nat’l Football League Players’ Concussion Injury Litig., 775 F.3d 570, 584–89 (3d Cir. 2014). In National Football League, the district court expressly stated in its order granting conditional certification of a settlement class that “the proposed settlement class and subclasses ‘preliminarily satisf[ied]’ the requirements of Rule 23(a) and (b).” Id. at 585 (quoting In re Nat’l Football League Players’ Concussion Injury Litig., 301 F.R.D. 191, 202 (E.D. Pa. 2014)). Conversely, the district court’s decision to certify a negotiation class in this case was final. As the district court itself noted, the putative class members were asked to decide whether to “bind themselves to a negotiation process,” with no guarantee of a second opportunity to opt-out after a settlement figure was reached. In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 540. Therefore, this appeal is not premature; Defendants and the opt-out cities have properly challenged the district court’s “order granting . . . class-action certification.” Fed. R. Civ. P. 23(f).

Plaintiffs also challenge whether Defendants have appellate standing to pursue this appeal, suggesting that because Defendants in this case may simply ignore the negotiation class “with no legal consequences . . . they are under no legal order and accordingly lack standing to appeal.” (Appellee’s Br. at 39.) A party has appellate standing—which is distinct from Article III standing—when that party is “aggrieved by a judgment or order of a district court.” Deposit Guaranty Nat’l Bank v. Roper, 445 U.S. 326, 333 (1980); see also City of Cleveland v. Ohio, 508 F.3d 827, 836 (6th Cir. 2007) (“To have appellate standing, ‘a party must be aggrieved by the judicial action from which it appeals.’” (quoting Vogel v. City of Cincinnati, 959 F.2d 594, 599 (6th Cir, 1992))). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 8

Defendants are plainly aggrieved by the certification of a negotiation class. Although the district court was clear that no Defendant must negotiate with the class, its presence obviously affects the state of play. The negotiation class is designed to fundamentally alter the nature of the MDL—to foster settlement through a novel means of class action that binds an unprecedented number of municipalities into a single bloc. The district court noted that the purpose of the certification order was to “promote global settlement.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 537. Defendants are pressured, or at least strongly incentivized, to negotiate with the class. The impact the negotiation class will have on the contours of future settlement discussions and the individual MDL cases suffices for appellate standing. See also Blair v. Equifax Check Servs., Inc., 181 F.3d 832, 834 (7th Cir. 1999) (“[A] grant of class status can put considerable pressure on the defendant to settle, even when the plaintiff’s probability of success on the merits is slight.”).

We review the district court’s decision to certify a class for an abuse of discretion. In re Whirlpool Corp. Front-Loading Washer Prods. Liab. Litig., 722 F.3d 838, 850 (6th Cir. 2013). It is an abuse of discretion for a district court to commit a legal error in certifying a class. See id. Further, the Supreme Court has warned that district courts do not have the liberty to invent a procedure with “no basis in the Rule’s text,” even absent language expressly prohibiting it. Wal- Mart Stores, Inc. v. Dukes, 564 U.S. 338, 363 (2011); see also Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620 (1997) (“And, of overriding importance, courts must be mindful that the Rule as now composed sets the requirements they are bound to enforce. . . . The text of a rule thus proposed and reviewed limits judicial inventiveness.”).

Class actions are “an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Dukes, 564 U.S. at 348 (quoting Califano v. Yamasaki, 442 U.S. 682, 700–701 (1979)). They can serve important judicial interests in efficiency and finality by aggregating common claims and resolving them via a single lawsuit rather than numerous piecemeal actions. In addition, where many individuals have suffered the same injury, but each concomitant claim would yield negligible relief to a single plaintiff, such claims would rarely be brought but for the availability of the class action device. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 161 (1974) (“A critical fact in this litigation is that Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 9

petitioner’s individual stake in the damages award he seeks is only $70. No competent attorney would undertake this complex antitrust action to recover so inconsequential an amount. Economic reality dictates that petitioner’s suit proceed as a class action or not at all.”). Class actions also allow private actors, who would otherwise be deterred from pursuing their insubstantial claims, to assist in the enforcement of the law and deter harmful behavior by defendants. See Deposit Guaranty, 445 U.S. at 339 (“The aggregation of individual claims in the context of a classwide suit is an evolutionary response to the existence of injuries unremedied by the regulatory action of government.”).

Rule 23 balances these laudable goals against the individual rights of litigants by imposing demanding standards on class certification. For example, a court must determine that the class is “so numerous that joinder of all members is impracticable,” that there are “questions of law or fact common to the class,” that the claims or defenses of the representative party are “typical of the claims or defenses of the class,” and that the representative party is able to “fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(1)–(4); see also Nessel ex rel. Michigan v. AmeriGas Partners, L.P., 954 F.3d 831, 835 (6th Cir. 2020) (observing that these four requirements “are the defining characteristics of Rule 23”). Moreover, when a court seeks to certify a class under Rule 23(b)(3), like the district court did in the present case, it must find that that “the 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.”

We have recently affirmed that the Federal Rules of Civil Procedure “are binding upon court and parties alike, with fully the force of law” because they are “[p]romulgated pursuant to the Rules Enabling Act, 28 U.S.C. § 2072.” In re Nat’l Prescription Opiate Litig., 956 F.3d 838, 844 (6th Cir. 2020). Therefore, in assessing the lawfulness of the district court’s order, “we begin with the text” of Rule 23. See Limtiaco v. Camacho, 549 U.S. 483, 488 (2007). “[W]hen [a] statute’s language is plain, the sole function of the courts—at least where the disposition required by the text is not absurd—is to enforce it according to its terms.” Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6 (2000) (internal quotations omitted). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 10

Rule 23 is replete with references to litigation and settlement classes. For example, the Rule begins by providing that class representatives “may sue or be sued . . . on behalf of all members.” Fed. R. Civ. P. 23(a). Then, Rule 23(b)(1)(A) notes that “class actions may be maintained if . . . prosecuting separate actions . . . would create a risk of inconsistent or varying adjudications.” And, as noted above, Rule 23(b)(3) provides that a class action certified under that provision must be “superior to other available methods for fairly and efficiently adjudicating the controversy.” Settlement classes are specifically identified in Rule 23(e): “The claims, issues, or defenses of a certified class—or a class proposed to be certified for purposes of settlement—may be settled, voluntarily dismissed, or compromised only with the court’s approval.” (emphasis added). Notably, the Rule does not mention certification for purposes of “negotiation” or anything along those lines. While negotiation may lead to settlement, there is no discussion in Rule 23 identifying negotiation as a separate category of certification distinct from settlement.

Plaintiffs recognize this but maintain that Rule 23 does not explicitly prohibit negotiation classes. In their view, “the open-ended wording of Rule 23(b)(3) . . . [a]nd the text of Rule 23(c)(4)” indicate that a class can be certified for purposes other than trial or settlement. (Appellees’ Br. at 40.) Moreover, Plaintiffs continue, Rule 23 did not explicitly sanction settlement classes until 2018. This, Plaintiffs contend, shows that the lack of express textual authorization is not necessarily fatal to a novel form of class action. Instead, in accordance with the district court’s “‘substantial discretion in determining whether to certify a class’ . . . [t]he Negotiation Class certified in this case falls precisely within the latitude of district courts recognized by this Court.” (Id. at 43 (quoting Sandusky Wellness Ctr, L.L.C. v. ASD Specialty Healthcare, Inc., 863 F.3d 460, 466 (6th Cir. 2017)).)

What Plaintiffs fail to appreciate is that a new form of class action, wholly untethered from Rule 23, may not be employed by a court. The Supreme Court has specifically cautioned that “a mere negative inference does not in our view suffice to establish a disposition that has no basis in the Rule’s text.” Dukes, 564 U.S. at 363. In Dukes, the plaintiffs claimed that a Rule 23(b)(2) class action authorized to seek injunctive relief could also seek money damages because the issues related to injunctive relief predominated and there was no explicit textual prohibition Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 11

on that practice. Id. The Court rejected that attempt as incompatible with the text of Rule 23. See id. at 362–63. The Supreme Court has also emphasized that we are to be “mindful that the Rule as now composed sets the requirements [courts] are bound to enforce,” that the Rule “limits judicial inventiveness,” and that “[c]ourts are not free to amend a rule outside the process Congress ordered.” Amchem, 521 U.S. at 620.

In their attempt to locate the negotiation class within the bounds of Rule 23, Plaintiffs compare negotiation classes to settlement classes and rely on the history of settlement classes to justify this expansion of Rule 23. However, the history of how settlement classes entered into common use only undermines Plaintiffs’ overly-broad reading of the Rule’s text. Although it was not until 2018 that settlement classes were expressly added to the text of Rule 23, the version of Rule 23(e) in effect before that time could be interpreted relatively easily to permit them. It 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 in such manner as the court directs.

Amchem, 521 U.S. at 620–21 (emphasis added) (quoting Fed. R. Civ. P. 23(e) (1997)). The Rule expressly contemplated courts overseeing “compromised” class actions, that is, class actions where a settlement has been privately negotiated and the parties wished to present it to the court. Implicit in that authority is, arguably, the authority to certify a class for that precise purpose. The Supreme Court was so untroubled by the regular certification of settlement classes that it referred to them as a “stock device.” Id. at 618.

Rule 23(e), which is titled “Settlement, Voluntary Dismissal, or Compromise,” now provides, “[t]he claims, issues, or defenses of a certified class—or a class proposed to be certified for purposes of settlement—may be settled, voluntarily dismissed, or compromised only with the court’s approval.” Unlike settlement classes under the pre-2018 Rule, there is no textual basis in this subsection—or any other—for the existence of a negotiation class. The class formed in the present case is not being formed for the purposes of litigation or “for purposes of settlement,” but rather for the purposes of negotiation. At most, the class is being formed, pursuant to a set of rules outside the parameters of Rule 23, to explore the possibility of Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 12

negotiating a settlement. But the Rule contemplates settlement classes that are formed after a deal has been reached and the parties wish to formalize their arrangement. The district court even emphasized that participation in the negotiation class is not compulsory—no MDL plaintiff must join it, nor do any of the Defendants have to negotiate with it. It made clear that its certification order “does not approve, or commence the approval process for, any specific settlement or proposed settlement.” (R. 2591, Order Certifying Negotiation Class, PageID # 413624.) This further distinguishes a negotiation class from a permissible settlement class because the latter presents the court with a proposal that would permit it to dispose of the case under appropriate circumstances contemplated by Rule 23. Ultimately, the speculative possibility that this negotiation class will settle a broad swath of the MDL does not bring it within the narrow textual confines of Rule 23 as a settlement class.

To hold otherwise would also ignore other elements of Rule 23(e) that clearly indicate certification of a class for settlement purposes may occur only after a settlement has been proposed. For example, before approving a proposed settlement, the court must determine that “class representative and class counsel have adequately represented the class,” that “the proposal was negotiated at arm’s length,” and “the relief provided for the class is adequate.” Fed. R. Civ. P. 23(e)(2) (emphasis added). None of these determinations can be made in the case of a negotiation class because there is no proposal to consider at the time the negotiation class is presented to the court for approval.

The language of Rule 23 that governs litigation classes also does not support Plaintiffs’ interpretation of the Rule. It is true that litigation classes can lead to settlement, and, pursuant to Rule 23(e)(4), a court is not required to offer a second opt-out period before approving the agreement. This has the effect of forcing putative litigation class members in classes certified under Rule 23(b)(3) to decide whether to stay in the litigation class when it is first proposed for certification, with the knowledge that they may ultimately be bound by an unknown future settlement that they would play a marginal, if any, role in negotiating. So too in the case of the negotiation class in the present case. But the parallel ends there. The negotiation class was expressly certified for the purposes of fostering global settlement, rather than litigating common issues. That is presumably why the court indicated that the individual cases in the MDL would Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 13

proceed alongside the negotiation class. Rule 23 permits litigation classes primarily for the purposes of aggregating and adjudicating common claims for trial, which can avoid conflicting judgments in individualized proceedings and can more efficiently resolve the claims of the class through a single lawsuit. That aggregation cannot occur where, as here, independent cases continue in parallel with a negotiation class that will not bring any common claims to trial. Moreover, as will be discussed in greater detail below, the district court did not fully engage in the requisite analysis for a litigation class because it understood the negotiation class as constituting its preferred method of proceeding for the purposes of settlement. This further separates the negotiation class, as certified by the district court, from a litigation class.

The structure of Rule 23 also does not support Plaintiffs’ novel attempt to certify a negotiation class. See Dukes, 564 U.S. at 363 (finding proposed interpretation of Rule 23(b)(2) unpersuasive, in part, because it did “obvious violence to the Rule’s structural features”). As the above analysis suggests, a negotiation class does not fit into either the litigation class or settlement class tracks. The district court’s order states that when the negotiation class begins its work it “will not displace or interfere with any of th[e] on-going litigation.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 551. But in conducting the requisite Rule 23(a) and 23(b) certification analyses, the court must frame its inquiry in different ways depending on the type of class being certified. For example, in determining whether to certify a litigation class, courts must determine “how a trial on the merits would be conducted if a class were certified.” Sandusky, 863 F.3d at 468 (quoting Gene & Gene L.L.C. v. BioPay L.L.C., 541 F.3d 318, 326 (5th Cir. 2008)). In the case of a settlement class the analysis is in certain respects simpler. The Supreme Court observed in Amchem that “[c]onfronted with a request for settlement-only class certification, a district court need not inquire whether the case, if tried, would present intractable management problems . . . for the proposal is that there be no trial.” 521 U.S. at 620 (citing Fed. R. Civ. P. 23(b)(3)(D)). As one leading treatise comments, courts “regularly certify settlement classes that might not have been certifiable for trial purposes because of manageability concerns.” William B. Rubenstein, Newberg on Class Actions § 4:63 (5th ed. 2020). In fact, the district court in the present case made such a determination in finding that the manageability prong “is inapplicable to the proposed negotiation class, as the proposal is not for litigation or trial, but simply for settlement negotiations.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 14

at 551–52. The negotiation class device thus frustrates a court’s analysis of whether a class action is the superior method of adjudication. It avoids some of the procedural requirements of litigation class certification without halting the underlying litigation and ideally settling the claims instead, as in the case of a settlement class.

This structural problem is compounded by the district court’s attempt to frame the negotiation class as an “issue class.” See id. at 556; see also Fed. R. Civ. P. 23(c)(4) (“[A]n action may be brought or maintained as a class action with respect to particular issues.”). The district court certified the negotiation class based upon federal RICO and Controlled Substances Act claims brought by many of the putative class members, finding that “common issues predominate over individualized issues with respect to both.” In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 551. Nevertheless, the court made clear that the class was permitted “to negotiate settlements . . . on any of the claims or issues identified here, or those arising out of a common factual predicate.” Id. at 556 (emphasis added). Many of the claims “arising out of a common factual predicate” are disparate state law claims brought by cities and counties throughout the country. Therefore, while the court’s predominance analysis might make some sense in the context of a litigation class or settlement class limited to resolving the narrow federal issues cited in the certification order, the court’s order minimized or marginalized the myriad state law claims that arguably divide the putative class members. The district court’s order creates confusion surrounding the scope of negotiations—a putative class member cannot be sure whether, and how, the negotiation class representatives, empowered by the court, will address their state law claims during settlement discussions.

Therefore, in certifying a negotiation class, the district court has papered over the predominance inquiry. See Martin v. Behr Dayton Thermal Prods. LLC, 896 F.3d 405, 413 (6th Cir. 2018) (holding that issue-class certification is appropriate “where common questions predominate within certain issues and where class treatment of those issues is the superior method of resolution” (emphasis added)). The issue class device permits a court to split common issues off for class treatment; it does not provide an end-run around the weighty requirements of Rule 23(b)(3). For example, the court needed to determine whether a class action is superior to other forms of adjudication by considering, among other criteria, “the class members’ interests in Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 15

individually controlling the prosecution or defense of separate actions.” Fed. R. Civ. P. 23(b)(3)(A). The district court found that because only “a small fraction” of the over 34,000 municipal entities are individually litigating their own claims, the negotiation class will increase “individual control and involvement” by engaging class members in the negotiation and voting process. In re Nat’l Prescription Opiate Litig., 332 F.R.D. at 551. Actually, the district court’s approach would do the opposite of increasing individual control and involvement by requiring class action participants to commit to the negotiation class without knowing the issue parameters or the amount or prospect of any potential recovery.

Therefore, even if the court were determined to be correct in its assessment of the federal law claims, there would remain the problem of potential and actual numerous state law claims. If the negotiation class can pursue settlement of all claims raised by the class members, it is possible that those members’ individual control over their diverse or uncommon state law claims may be compromised. This is especially true because the district court has the authority to decline to offer a second opt-out period. See Fed. R. Civ. P. 23(e)(4).

The primary problem here is that the negotiation class ordered by the district court simply is not authorized by the structure, framework, or language of Rule 23. But even if not unauthorized, it is unlikely that the problems presented by the negotiation class, as conceived by the district court, can be overcome. For instance, even without prejudging whether common issues do in fact predominate in this MDL, we think that the court’s analysis was made incomplete by its decision to restrict its reasoning to the common federal claims in order to certify a negotiation class. Rule 23’s structure does not permit the court to evade a proper (b)(3) analysis by certifying a negotiation class based upon a few common federal issues, while at the same time empowering the class to negotiate settlement on virtually all claims brought by the class members. The district court’s course of action neither fully satisfied the critical fairness concerns of Rule 23(b)(3), nor respected the important differences between litigation classes and settlement classes. We therefore believe that Plaintiffs’ interpretation of Rule 23 does “obvious violence to the Rule’s structural features.” Dukes, 564 U.S. at 363.

Plaintiffs have also not shown why the options of a litigation class or settlement class are infeasible in the present case. While there are drawbacks to both approaches, the opioid crisis Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 16

MDL might be amenable to either. If common issues actually predominate, then a litigation class might be certifiable. Alternatively, Plaintiffs may privately seek a settlement with Defendants and ultimately propose an agreement to the court for certification of a settlement class. There is no apparent reason why some of the procedural elements of the negotiation class, such as the supermajority voting scheme and county-level allocation formula, could not be used to facilitate the participation of more Plaintiffs in a lawful settlement class.

In the final analysis, we do not see how the negotiation class can be squared with Rule 23. However innovative and effective the addition of negotiation classes would be to the resolution of mass tort claims—particularly those of grave social consequence—we are to be “mindful that the Rule as now composed sets the requirements [courts] are bound to enforce,” and we “are not free to amend a rule outside the process Congress ordered.” Amchem, 521 U.S. at 620. That process involves careful review of a proposed amendment by the Rules Advisory Committee, the Judicial Conference, the Supreme Court, and Congress. See 28 U.S.C. §§ 2072– 74. After the Rules Advisory Committee has recommended a change to the Judicial Conference, that body may propose that the Supreme Court promulgate the amendment. Even if the Court does so, Congress may prevent the change through statutory enactment before the new rule goes into effect. See id. This multi-layered review process ensures that alterations to the Rules can only be made after thorough deliberations by multiple expert bodies, which can assess the virtues and drawbacks of a proposed change as well as evaluate the possible implications of the proposed rule across the entire judicial system, rather than by individual judges facing the pressures of litigation.

Additionally, this process is “properly tuned to the instruction that rules of procedure ‘shall not abridge . . . any substantive right.’” Amchem, 521 U.S. at 620 (quoting 28 U.S.C. § 2072(b)). This is critical in the class action context where a certification decision, which is procedural, can abrogate the substantive rights of class members to pursue their claims independently by binding them to a litigation class or settlement class. The fact that negotiation class members in the present case must choose whether to opt-out before knowing the settlement amount only heightens our concern. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 17

Perhaps in due course the Rules Advisory Committee and Judicial Conference will recommend revising Rule 23 to sanction the negotiation class, which the Supreme Court and Congress may then approve. Until that time, however, this novel mechanism is beyond the Rule’s scope.

CONCLUSION

For the reasons provided above, we REVERSE the district court’s judgment certifying a negotiation class and REMAND this case for further proceedings consistent with this opinion. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 18

DISSENT _________________

KAREN NELSON MOORE, Circuit Judge, dissenting. The Federal Rules of Civil Procedure were not written and have never been interpreted to manacle district courts that innovate within the Rules’ textual borders. The district court has breathed life into a novel concept—a class certified for negotiation purposes—to aid in its Promethean duty to secure the just, speedy, and inexpensive resolution of this byzantine multidistrict litigation. We should be in the business of encouraging, not exterminating, such resourcefulness. Certifying a negotiation class honors the Rules’ equitable heritage, complements the settlement class’s history, hews to Federal Rule of Civil Procedure 23’s textual requisites, and stirs no constitutional or policy qualms. So, with respect, I dissent.

I.

Text, origin story, and life cycle should propel our interpretation of the Federal Rules of Civil Procedure.

The Rules’ “design[],” Celotex Corp. v. Catrett, 477 U.S. 317, 327 (1986), and “paramount command,” Dietz v. Bouldin, — U.S. —, 136 S. Ct. 1885, 1891 (2016), are clear: courts should construe, administer, and employ the Rules to secure “the just, speedy, and inexpensive resolution of disputes,” id. (citing Fed. R. Civ. P. 1). Embodying this edict are the plethora of Rules that invite district courts to apply the Rules’ provisions with “local imagination.”1 Jack B. Weinstein, After Fifty Years of The Federal Rules of Civil Procedure: Are the Barriers to Justice Being Raised?, 137 U. Pa. L. Rev. 1901, 1911 (1989). Those who

1 See, e.g., FED. R. CIV. P. 8(e) (“Pleadings must be construed so as to do justice.”); FED. R. CIV. P. 15(a)(2) (“The court should freely give leave [to amend pleadings] when justice so requires.”); FED. R. CIV. P. 21 (“On motion or on its own, the court may at any time, on just terms, add or drop a party.”); FED. R. CIV. P. 40 advisory committee’s note to 2007 amendment (“The best methods for scheduling trials depend on local conditions. . . . It is not useful to limit or dictate the provisions of local rules.”); FED. R. CIV. P. 54(b) (“[T]he court may direct entry of a final judgment as to one or more, but fewer than all, claims or parties only if the court expressly determines that there is no just reason for delay.”); FED. R. CIV. P. 83 (“A judge may regulate practice in any manner consistent with federal law, rules adopted under 28 U.S.C. §§ 2072 and 2075, and the district’s local rules. No sanction or other disadvantage may be imposed for noncompliance with any requirement not in federal law, federal rules, or the local rules . . . .”). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 19

endowed the Rules with such flexibility drew on traditions of equity practice and were inspired by the contemporaneous shift toward placing substance over form.2 See id. at 1906. The Rules’ distinctive promulgation procedures similarly counsel against our suffocating the district court with textual piety. The Rules spring not from the head of Congress; the Supreme Court itself promulgates and implements the Rules. This, in large measure, eliminates judicial threats to separation-of-powers and emancipates courts from traditional concerns surrounding statutory interpretation, such as deference to Congress. See Joseph P. Bauer, Schiavone: An Un-Fortune- ate Illustration of the Supreme Court’s Role As Interpreter of the Federal Rules of Civil Procedure, 63 Notre Dame L. Rev. 720, 720 (1988); see also Lumen N. Mulligan & Glen Staszewski, Civil Rules Interpretive Theory, 101 Minn. L. Rev. 2167, 2168 (2017) (similar argument); Karen Nelson Moore, The Supreme Court’s Role in Interpreting the Federal Rules of Civil Procedure, 44 Hastings L.J. 1039, 1093 (1993) (same). Texts and canons of construction should not ensnare our interpreting and applying a Federal Rule of Civil Procedure. Rather, courts should contemplate a liberal reading that fulfills the Rules’ broader design and that does justice for the parties. See Bauer, supra, at 720.3

This understanding of the Federal Rules of Civil Procedure is neither a historical footnote nor a scholarly chimera. The Supreme Court’s refusal to strictly construe the Rules “inconsistent[ly] with the[ir] liberal atmosphere,” Hickman v. Taylor, 329 U.S. 495, 505 (1947), continues throughout decades of precedent. The Rules contain no text providing for the power to rescind a jury discharge order and recall a jury for further deliberations, to hear a motion in 2 “Few disagree” that the Federal Rules of Civil Procedure’s drafters intended to “sweep[] away the dirt and cobwebs” and for the Rules “to open wide the courthouse doors . . . and let the sunlight of substance shine into them.” Jack B. Weinstein, After Fifty Years of The Federal Rules of Civil Procedure: Are the Barriers to Justice Being Raised?, 137 U. PA. L. REV. 1901, 1906 (1989). The Rules were “written to further, not defeat[,] the ends of justice,” Surowitz v. Hilton Hotels Corp., 383 U.S. 363, 373 (1966), and were “designed to further the due process of law that the Constitution guarantees,” Nelson v. Adams USA, Inc., 529 U.S. 460, 465 (2000) (citing FED. R. CIV. P. 1). 3 See also Lumen N. Mulligan & Glen Staszewski, Civil Rules Interpretive Theory, 101 MINN. L. REV. 2167, 2168 (2017) (“[The] principle of legislative supremacy and the related notion that the federal courts should serve as a faithful agent of Congress, which undergird every traditional theory of statutory interpretation, do not apply in the Rules context.”); Karen Nelson Moore, The Supreme Court’s Role in Interpreting the Federal Rules of Civil Procedure, 44 HASTINGS L.J. 1039, 1093 (1993) (“Given these substantial, although largely unexercised, powers of the Court in the promulgation process, a more activist role in the interpretative stage, one that considers purpose and policy, is appropriate. Congress has explicitly delegated to the Court rulemaking power, and it is not inconsistent to imply the Court has greater power to interpret Rules than it does to interpret statutes.”) (footnote omitted). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 20

limine, or to dismiss cases for forum non conveniens. See Dietz, 136 S. Ct. at 1891. Yet the Court approved each of these—and many other—creatures of district court creativity. See id.; Link v. Wabash R. Co., 370 U.S. 626, 630–31 (1962); Gulf Oil Corp. v. Gilbert, 330 U.S. 501, 507 (1947). Courts are “necessarily vested” with the power “to manage their own affairs so as to achieve the orderly and expeditious disposition of cases”; such power is “governed not by rule or statute,” Dietz, 136 S. Ct. at 1891 (quoting Link, 370 U.S. at 630–631), but by “considerations of justice,” United States v. Hasting, 461 U.S. 499, 505 (1983). This principle has “long gone unquestioned” by the Supreme Court. Link, 370 U.S. at 631.4

Even when confronted with unfriendly text, the Court has capaciously interpreted theories underlying the Federal Rules of Civil Procedure to promote efficient litigation and to react to modern litigation trends. In Hickman, for instance, the Court conceded that the petitioner had failed to follow the black letter of Federal Rule of Civil Procedure 26, which espoused a “broad and liberal treatment” of discovery matters and seemingly required adverse attorneys to turn over documents formed in the course of their legal duties. 329 U.S. at 507. But, out of concern for the “public policy underlying the orderly prosecution and defense of legal claims,” id. at 510, the Court fashioned the work-product exception to the discovery rules, see id. at 509–12. The 1986 summary judgment trilogy—Celotex Corp. v. Catrett, 477 U.S. 317, 327 (1986), Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (1986), and Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574 (1986)—emerged from the same vein. Historically, district courts rarely granted summary judgment due to the “onerous” burden on moving parties articulated in Adickes v. S. H. Kress & Co., 398 U.S. 144 (1970). See Craig M. Reiser, The Unconstitutional Application of Summary Judgment in Factually Intensive Inquiries, 12 U. Pa. J. Const. L. 195, 200 (2009). In the lead-up to the trilogy, the Court comprehended the need for a “strengthened, modern view of the summary judgment standard,” id., that adequately reflected how the “Federal Rules as a whole . . . are designed to ‘secure the just, speedy, and inexpensive determination of every action.’” Celotex, 477 U.S. at 327 (quoting Fed. R. Civ. P. 1). But an

4 See also Goodyear Tire & Rubber Co. v. Haeger, — U.S. —, 137 S. Ct. 1178, 1186 (2017) (“Federal courts possess certain ‘inherent powers,’ not conferred by rule or statute, ‘to manage their own affairs so as to achieve the orderly and expeditious disposition of cases.’”) (internal citations omitted); Degen v. United States, 517 U.S. 820, 823 (1996) (“Courts invested with the judicial power of the United States have certain inherent authority to protect their proceedings and judgments in the course of discharging their traditional responsibilities.”). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 21

encumbrance remained: Federal Rule of Civil Procedure 56’s text. The Court admitted that two sentences in Rule 56(e) were added to disapprove of a line of cases that allowed a nonmoving party to resist a summary judgment motion by reference only to its pleadings. See Celotex, 477 U.S. at 325. But the Court found a path through the text’s hot gates: although the sentences “were not intended to reduce the burden of the moving party, [the sentences] were not adopted to add to that burden” either. Id. Purpose—not plain text—bolstered the Court’s creation of our modern summary judgment standards.5

In two cases of more recent vintage, Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009), the Court grafted a “plausibility standard” onto the Federal Rules of Civil Procedure’s pleading requirement, Twombly, 550 U.S. at 560–61. Although the standard finds no support in Federal Rule of Civil Procedure 8’s text, the Court determined that courts must conduct a “plausibility” analysis at the pleading stage “to avoid the potentially enormous expense of discovery in cases with no reasonably founded hope that the [discovery] process will reveal relevant evidence[.]” Id. at 559 (internal quotation marks and citation omitted); see also Iqbal, 556 U.S. at 685 (citing the costs that litigation imposes on government officials). Finally, text did not manacle the Court when it dealt with the meaning of “commonality” under Rule 23(a)(2) in Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011). Fretting that the proposed class lacked any “glue” to hold the alleged reasons for “literally millions of employment decisions” together, id. at 352, the Court defined “commonality” not by its plain meaning but by reference to academic literature6 and precedent that emphasized the need for district courts to “rigorous[ly] analy[ze]” the Rule 23 prerequisites, id. at 350–51.

Multidistrict litigation’s peculiar properties demand that we follow the Supreme Court’s tack in this case. The genesis of multidistrict litigation mirrors the nativity of the Federal Rules of Civil Procedure. Multidistrict litigation’s creators anticipated a “litigation explosion” coming to the federal courts and that permanent reform to federal civil procedure was necessary to

5 Just as decades-old procedures regarding proposed settlements in class actions were codified in Rule 23 in 2003, the work-product exception was codified in Rule 26 only in 1970. See FED. R. CIV. P. 26 advisory committee’s note to 1970 amendment. Likewise, Celotex’s standard for summary judgment was codified in Rule 56 only in 2009. See FED. R. CIV. P. 56 advisory committee’s note to 2009 amendment. 6 The Dukes Court heavily cited Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L. REV. 97 (2009). See, e.g., Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 349–50 (2011). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 22

handle it. See Andrew D. Bradt, Something Less and Something More: MDL’s Roots As A Class Action Alternative, 165 U. Pa. L. Rev. 1711, 1713 (2017) (citing Judicial Administration: Hearings on H.R. 3991, H.R. 6703, H.R. 8276, and H.R. 16575 Before Subcomm. No. 5 of the H. Comm. on the Judiciary, 89th Cong. 26 (1966)). Understanding that “pretrial proceedings were increasingly becoming the main event in large-scale litigation,” id. at 1713, scholars sculpted a system of “procedural exceptionalism,” Abbe R. Gluck, Unorthodox Civil Procedure: Modern Multidistrict Litigation’s Place in the Textbook Understandings of Procedure, 165 U. Pa. L. Rev. 1669, 1674 (2017), that allows judges to develop “flexible and creative” procedures in every multidistrict litigation case, id. at 1689. We should respect that every multidistrict litigation is unorthodox. See id. Courts overseeing multidistrict litigation are adept at replicating and refining procedures over time, in true common-law fashion. See id. at 1689–91 (noting the gradual development of “bellwether trials” and “fact sheets”). We should encourage liberal constructions of the Federal Rules of Civil Procedure that abet, rather than constrict, this process.

All of this is not to say that the Federal Rules of Civil Procedure’s textual hooks do not apply to multidistrict litigation or class action certifications.7 See, e.g., In re National Prescription Opiate Litig. (Opiate Litig. 6th Cir. Op. Apr. 15, 2020), 956 F.3d 838, 841 (6th Cir. 2020) (reversing MDL court for allowing belated amended complaint in violation of Federal Rule of Civil Procedure 16(b)); see also Gluck, Unorthodox Procedure, supra, at 1706 (noting that MDL litigation judges themselves express a “desire for some decisional law”). We need no reminder that we are all textualists now.8 But, even taken at face value, the text of Rule 23 does not prohibit the certification of a negotiation class.

7 In practice, it appears that no judge overseeing multidistrict litigation adheres closely to the text of Rule 16. See Abbe R. Gluck, Unorthodox Civil Procedure: Modern Multidistrict Litigation’s Place in the Textbook Understandings of Procedure, 165 U. PA. L. REV. 1669, 1688–89 (2017) (“[V]irtually all of the judges interviewed reported that typical MDL management goes far beyond the confines of Rule 16. As one judge put it: ‘It’s like Rule 16 on steroids. In the MDL, you need to strategize more. You have to look beyond immediate deadlines and see how all the pieces fit together.’ . . . All of the judges respected the FRCP, but every judge opposed the idea of a new FRCP for MDLs or even a uniform common law approach.”). 8 See Brett M. Kavanaugh, Fixing Statutory Interpretation, 129 HARV. L. REV. 2118 (2016) (book review) (“[W]e’re all textualists now.”) (quoting Justice Elena Kagan, The Scalia Lecture: A Dialogue with Justice Kagan on the Reading of Statutes at 8:09 (Nov. 17, 2015), http://today.law harvard.edu/in-scalia-lecture-kagan-discusses- statutory-interpretation). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 23

The majority observes that Rule 23 is “replete with references to litigation and settlement classes” and does not mention certification for the purposes of “negotiation.” Maj. Op. at 10. Further, the majority reasons, Rule 23(e)’s language—particularly the use of the past tense— indicates that the certification of a class for settlement may occur only after a settlement has been proposed. Maj. Op. at 11–12. The majority is mistaken. The world of class actions is neither constituted in entirety nor cleft in two by the rigid categories of litigation classes and settlement classes. I find not one textual reference to the phrases “litigation class” or “settlement class” in the Rules. Indeed, the very first words of Rule 23(e) recognize that classes that have already been certified may settle or compromise their cases in the future: “The claims, issues, or defenses of a certified class—or a class proposed to be certified for purposes of settlement—may be settled, voluntarily dismissed, or compromised only with the court’s approval.” Fed. R. Civ. P. 23(e) (emphasis added). Certainly, references to certification for “settlement and litigation purposes” crop up in the Advisory Committee Notes. Fed. R. Civ. P. 23 advisory committee’s note to 2018 amendment. But, as a matter of logic, negotiation is part and parcel of any class certified for settlement purposes. And the Rules’ language does not separate the concept of negotiation from “settlement” or “compromise,” Fed. R. Civ. P. 23(e); nor did the Rules Advisory Committee rip negotiation out of “issues related to settlement,” Fed. R. Civ. P. 23 advisory committee’s note to 2018 amendment. Thus, the district court’s finding that Rule 23 permits certification for negotiation purposes is no “mere negative inference,” Dukes, 564 U.S. at 363; such a reading is a permissible, and encouraged, contemplation of the Rule’s plain text.

II.

The remarkable history of the negotiation class’s ancestor, the settlement class, further buttresses the district court’s certifying a class for negotiation purposes.

As with the other Federal Rules of Civil Procedure, equity practice bore and nurtured Rule 23. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 613 (1997).9 The drafters of the

9 See also Francis E. McGovern & William B. Rubenstein, The Negotiation Class: A Cooperative Approach to Class Actions Involving Large Stakeholders, TEX. L. REV. 1, 56 (forthcoming 2020) (“Rule 23 emerged from equity practice and therefore the text of the Rule is, not surprisingly, and likely purposefully, open-ended.”) (footnote omitted); Suzette M. Malveaux, The Modern Class Action Rule: Its Civil Rights Roots and Relevance Today, 66 U. KAN. L. REV. 325, 328 (2017) (“The American class action rule was born in 1938 with the goal of Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 24

original Rule 23 drew language and standards from the former Federal Equity Rules. See Fed. R. Civ. P. 23 advisory committee’s note to 1937 amendment.10 The original Rule’s “murky, indefinite, unclear, [and] obscure” text, Samuel Issacharoff & Peter Zimroth, An Oral History of Rule 23: an Interview with Professor Arthur Miller, 74 N.Y.U. Ann. Surv. Am. L. 105, 107 (2018) (Arthur R. Miller comments),11 was quickly put to the test by an unanticipated species of suit—the civil rights class action, see Suzette M. Malveaux, The Modern Class Action Rule: Its Civil Rights Roots and Relevance Today, 66 U. Kan. L. Rev. 325, 328–29 (2017). Class actions that challenged racial segregation failed to align with the textual requirements of original Rule 23’s three categories; yet “[m]any, perhaps most,” judges “shoehorn[ed]” civil rights class actions into the Rule’s provisions “regardless.”12 Id. at 329–30. By the mid-1960s, practitioners, judges, and scholars were disappointed that Rule 23 was “underused” and were disgruntled with Rule 23’s textual anachronisms; they desired to amplify the equitable theories

promoting equity and access.”); Benjamin Kaplan, Continuing Work of the Civil Committee: 1966 Amendments of the Federal Rules of Civil Procedure (I), 81 HARV. L. REV. 356, 376–80 (1967) (elaborating how the original Rule 23 incorporated the Federal Equity Rules). 10 The Advisory Committee Notes document that “[Rule 23(a)] is a substantial restatement of former Equity Rule 38 (Representatives of Class) as that rule has been construed[;]” that “[t]he rule adopts the test of former Equity Rule 38[;]” and that “[t]he provision in [former] Equity Rule 94 was later embodied in [former] Equity Rule 27, of which the present Rule 23 is substantially a copy.” FED. R. CIV. P. 23 advisory committee’s note to 1937 amendment. 11 Professors Issacharoff and Zimroth recorded their interview with Professor Arthur R. Miller, who “was present at the creation and drafting of Rule 23 in 1966.” Samuel Issacharoff & Peter Zimroth, An Oral History of Rule 23: an Interview with Professor Arthur Miller, 74 N.Y.U. ANN. SURV. AM. L. 105, 105 (2018) (Arthur R. Miller comments). 12 See also Malveaux, supra note 9, at 329 (“The aggregation rule, however, required litigants to shoehorn their case into one of three jural relations: ‘true,’ ‘hybrid’ or ‘spurious.’ In a nutshell: ‘true’ classes involved joint or common rights; ‘hybrid’ classes involved several rights concerning a claim over a specific property; and ‘spurious’ classes involved several rights, with only a common question and common relief justifying aggregation. Numerosity and adequacy of representation—principles required from the Rule’s inception—were often satisfied for ‘race relations’ civil rights cases from the late 1930s to 1960s. Within this three-tiered structure, civil rights cases were housed in the ‘true’ and ‘spurious’ classes.”). We cannot ignore that the 1966 revisions to Rule 23(b) were “rooted in the turbulent history of the civil rights movement of the 1960s and [were] designed to enhance civil rights enforcement.” Id. at 327. “In response to fierce resistance to desegregation following the iconic 1954 Brown v. Board of Education decision, the rule drafters amended Rule 23 in 1966 to enable structural reform and broad remedial relief.” Id.; see also Issacharoff & Zimroth (Arthur R. Miller comments), supra note 11, at 110 (“[Y]ou could feel that what happened in Brown with regard to school desegregation had legs, and it was going to osmose into other contexts. And you had to give the civil rights group a mobilizing capability, and the vehicle for doing that was the class action.”). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 25

undergirding the Rule. See Issacharoff & Zimroth (Arthur R. Miller comments), supra, at 107;13 see also Benjamin Kaplan, Continuing Work of the Civil Committee: 1966 Amendments of the Federal Rules of Civil Procedure (I), 81 Harv. L. Rev. 356, 375–94 (1967). This winter of discontent spawned the “innovative” 1966 revisions, which gave Rule 23 its “current,” Amchem, 521 U.S. at 613, and “most adventuresome,” id. at 614, shape. The drafters “[f]ramed” the revised Rule 23 “for situations in which ‘class-action treatment is not as clearly called for’” but “may nevertheless be convenient and desirable.” Id. at 615 (internal citation omitted).14

True to form, history recurred. Soon after the 1966 revisions, yet another class action device arose going beyond Rule 23’s plain text: the settlement class action. As early as 1970, a deluge of district courts began certifying classes for settlement purposes, see William B. Rubenstein, 2 Newberg on Class Actions § 4:93 n.1 (5th ed. June 2020 Update) (collecting cases); see also McGovern & Rubenstein, supra, at 54 n.199 (collecting cases), even though “there [was] an absence of clear textual authorization for settlement classes” in Rule 23, In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 786 (3d Cir. 1995). The Rule 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.” William W Schwarzer, Settlement of Mass Tort Class Actions: Order Out of Chaos, 80 Cornell L. Rev. 837, 841 (1995). This textual desert was no mistake—the framers of the 1966 revisions “conceive[ed] the rule as a trial-ready rule” and “[settlement] was not part of the discussion.” Issacharoff & Zimroth (Arthur R. Miller comments), supra, at 126. Notwithstanding vague language in Rule 23(e) that required a court’s

13 See also Issacharoff & Zimroth (Arthur R. Miller comments), supra note 11, at 107 (“There was a sense that the text of the rules at that time was murky, indefinite, unclear, obscure, whatever you want to call it. And that judges and lawyers were having difficulty and creating inconsistencies in the application of some of those rules. Those rules were underused, and it was thought time to rationalize them, to tie them together better than they had been tied in the 30s, and to clarify the text to capture the 25 years of experience, and to insert that experience under the Federal Rules of Civil Procedure into the rules.”). 14 Notably, the 1966 revisions, and Rule 23(b)(3) in particular, coaxed district courts to certify classes in “those cases in which a class action would achieve economies of time, effort, and expense, and promote uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results.” FED. R. CIV. P. 23 advisory committee’s note to 1966 amendment. The 1966 revisions “enable[d] structural reform and broad remedial relief,” Malveaux, supra note 9, at 327. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 26

“approval” before a class action is “dismissed or compromised,” Fed. R. Civ. P. 23(e) (1966),15 the Rule plainly did not contemplate parties coming to court “with a done deal and fil[ing] the class action complaint together with the proposed class action settlement.” Note, Back to the Drawing Board: The Settlement Class Action and the Limits of Rule 23, 109 Harv. L. Rev. 828, 829 (1996) (emphasis added). Notably, the courts’ postponing formal certification—an inherent element of settlement class certification—disobeyed Rule 23(c)(1), which “command[ed] the court[s] to determine ‘[a]s soon as practicable after the commencement of [the] action’ whether an action brought as a class action is to be ‘so maintained.’” Id. at 829–30 (quoting Fed. R. Civ. P. 23(c)(1) (1996)).16

As settlement class actions flooded the dockets of the federal courts, appellate courts struggled to reconcile the existence of settlement classes with Rule 23’s plain text.17 The Courts of Appeals knew that settlement classes were “not specifically authorized by Rule 23,” General Motors, 55 F.3d at 792, and demanded a “practical construction of the class action rule,” id. at 794. But, because “settlement classes [were] not specifically precluded by [Rule 23] either,” id. at 792, the Rule’s text liberated the federal circuits to recognize that settlement classes had “utility,” Amchem, 521 U.S. at 618, “afford[ed] considerable economies to both the litigants and the judiciary[,] and [are] also fully consistent with the flexibility integral to Rule 23,” General Motors, 55 F.3d at 794. Settlement classes “jostl[ed] uneasily with the language of Rule 23(c)(1)[,]” Mars Steel Corp. v. Cont’l Ill. Nat’l Co. of Chi., 834 F.2d 677, 680 (7th Cir. 1987); yet every federal appellate court and the Supreme Court declined to hold that courts were violating Rule 23 per se by certifying classes for settlement purposes, Amchem, 521 U.S. at 618.

15 The full text of the 1966 Rule 23(e) was: “Dismissal or Compromise. 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 in such manner as the court directs.” FED. R. CIV. P. 23(e) (1966). 16 Rule 23(e) remained unchanged between 1966 and 1996. Compare FED. R. CIV. P. 23(e) (1966) with FED. R. CIV. P. 23(e) (1996). 17 Confident that Rule 23 permitted settlement class actions to exist, some appellate courts skipped the textual analysis of the Rule altogether. See In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 794 (3d Cir. 1995) (“[F]ew cases since the late 1970’s and early 1980’s even bother to squarely address the propriety of settlement classes. Moreover, no court of appeals that has had the opportunity to comment on the propriety of settlement classes has held that they constitute a per se violation of Rule 23.”). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 27

Here, the district court’s certifying a negotiation class has no direct antecedent. The bespoke five-stage certification process emerged from an academic chrysalis, see generally McGovern & Rubenstein, supra, and followed no specific procedure delineated in Rule 23. This emergence of a negotiation class simply follows the incremental development of settlement class actions. Rule 23 bore no references to “settlement” until 2003,18 when the rules were revised to refer to “Settlement, Voluntary Dismissal, or Compromise” and to codify some district courts’ procedures associated with proposed settlements in class actions. See Fed. R. Civ. P. 23(e) (2003). Thus, the Rules did not explicitly reference settlements in the class action context for thirty-seven years after the 1966 revisions and six years after the Supreme Court ratified settlement class certifications as a “stock device” in 1997, Amchem, 521 U.S. at 618. Yet during this decades-long textual vacuum, not only did courts of every tier repeatedly avow that Rule 23’s text gave life to settlement class actions, but district courts also wrought sundry procedures that would, within ten years, constitute the “most elaborate settlement mechanism developed for class actions,” see Developments in the Law: Class Actions, 89 Harv. L. Rev. 1318, 1555 (1976). Standing on not one textual pillar, district courts experimented with designating one attorney as the official class negotiator, forming subclasses to ensure that all viewpoints are represented in negotiations, playing with different types of notice, and conferencing informally with the parties. See id. at 1559–60. Subsequent revisions to Rule 23(e) in 2003 and 2018 merely “codified some existing practices,” McGovern & Rubenstein, supra, at 56, which the Advisory Committee Notes

18 There seems to be some confusion about when Rule 23 was amended to include references to settlement classes. See Maj. Op. at 11 (“[I]t was not until 2018 that settlement classes were expressly added to the text of Rule 23 . . . .”); Opiate Litig., 332 F.R.D. at 539 (“Defendants point to the fact that several passages in Rule 23 specifically reference settlement, as opposed to trial, classes. . . . The passages they reference were not added to Rule 23 until December 2018.”); Appellees’ Br. at 41 (“It was not until December 1, 2018 that Rule 23 was amended to explicitly recognize settlement classes.”) (citing FED. R. CIV. P. 23(e), advisory committee’s note to 2018 amendment). Prior to 2003, Rule 23 contained no references to settlement, only to “Dismissal or Compromise.” FED. R. CIV. P. 23(e) (2002) (“Dismissal or Compromise. 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 in such a manner as the court directs.”). In 2003, Rule 23(e) was revised to refer to “Settlement, Voluntary Dismissal, or Compromise” and to codify some procedures for class actions that produced settlements. FED. R. CIV. P. 23(e) (2003); see also FED. R. CIV. P. 23(e) advisory committee’s note to 2003 amendment (“Subdivision (e) is amended to strengthen the process of reviewing proposed class-action settlements.”). The 2018 revisions codified some additional procedures related to settlements in class actions. See FED. R. CIV. P. 23(e) advisory committee’s note to 2018 amendment (“Rule 23 is amended mainly to address issues related to settlement, and also to take account of issues that have emerged since the rule was last amended in 2003.”). Thus, settlement classes were not expressly added to the text of Rule 23 until 2003, not 2018. But this altered timeline in no way affects the underlying historical takeaway—district courts certified settlement classes en masse for decades before the Rules contained any references to settlements in class actions. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 28

thoroughly acknowledge, see Fed. R. Civ. P. 23 advisory committee’s notes to 2003 and 2018 amendments.19 The parallel between the settlement class and the negotiation class is unmistakable; by certifying a novel negotiation class via a series of new-fashioned procedures, the district court here embraces Rule 23’s equitable heritage and the developments of district courts past.20 We ought not disturb the relationship between the innovative experiments of district courts and the subsequent codification of those developments in the revisions of the class action rule.

The majority is not convinced that the negotiation class and the settlement class can be cut from the same cloth. The majority determines that the pre-200321 version of the Rule— specifically the presence of the words “compromise” and “compromised” in Rule 23(e)—can be “interpreted relatively easily to permit [settlement classes].” Maj. Op. at 11. Perhaps tellingly, the appellate courts of the 1970s and 80s that ratified the existence of settlement class actions did not rely on Rule 23(e)’s use of the term “compromise”; their decisions instead were based on textual analyses of Rule 23(d)’s “appropriate orders” clause and Rule 23(c)(1)’s “conditional orders” clause. See Drawing Board, supra, at 834; see also General Motors, supra, at 792–94 (explaining the appellate courts’ different approaches as of 1995 to analyzing Rule 23’s text). Clearly none of these turns of phrase provide the explicit textual authorization that the majority seeks to distinguish the apparent textual support prior to 2003 for “settlement classes” from the alleged lack of current textual support for “negotiation classes.” These strategies to parse the

19 The 2003 Advisory Committee Notes document that “[s]ubdivision (e)(1)(C) confirms and mandates the already common practice of holding hearings as part of the process of approving settlement, voluntary dismissal, or compromise that would bind members of a class[]” and that “[s]ubdivision (e)(4) confirms the right of class members to object to a proposed settlement, voluntary dismissal, or compromise.” FED. R. CIV. P. 23 advisory committee’s note to 2003 amendment. The 2018 Notes further document that “[t]he central concern in reviewing a proposed class-action settlement is that it be fair, reasonable, and adequate. Courts have generated lists of factors to shed light on this concern. . . . The goal of this amendment is not to displace any factor.” FED. R. CIV. P. 23 advisory committee’s note to 2018 amendment. 20 Talented lawyers and judges have also experimented with settlement class actions in the context of multidistrict litigation, which has “fundamentally alter[ed] class action practice, and in particular, the role of class members as participants.” Elizabeth J. Cabraser & Samuel Issacharoff, The Participatory Class Action, 92 N.Y.U. L. Rev. 846, 851 (2017). 21 The majority cites to the 1997 version of Rule 23(e), see Maj. Op. at 11 (citing FED. R. CIV. P. 23(e) (1997)), which is the year that the Court decided Amchem. Rule 23 was amended once between 1997 and 2003 (in 1998). But the 1998 amendments did not alter the text of Rule 23(e), compare FED. R. CIV. P. 23(e) (1998) with FED. R. CIV. P. 23(e) (1997), so it is fair to refer to the 1997 version of Rule 23(e) as the “pre-2003” version of the Rule. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 29

text of the pre-2003 version of Rule 23 are necessarily scattershot because the drafters of the rule never intended class certification for any purpose other than trial, see Issacharoff & Zimroth (Arthur R. Miller comments), supra, at 125. No judicial body, not even the Supreme Court, pretended otherwise when validating the existence of settlement class actions. Further, if the concept of “settlement” can be drawn from the word “compromise,” surely the district court’s derivation of “negotiation” from “settlement” and “compromise” is by no means far-fetched. The district court’s development of a negotiation class has as much of a basis in current Rule 23 as the creation of settlement classes had in past versions of Rule 23. See General Motors, 55 F.3d at 792–94. Current Rule 23(e)(2)(B) embeds “negotiation” in settlement certification procedures. See Fed. R. Civ. P. 23(e)(2)(B) (“the proposal was negotiated at arm’s length”). Indeed, the majority admits that “negotiation” is a critical ingredient of “settlement.” Maj. Op. at 11 (“The [1966] Rule expressly contemplated courts overseeing ‘compromised’ class actions, that is, class actions where a settlement has been privately negotiated and the parties wished to present it to the court.”) (emphasis added).

Certainly, as the Supreme Court has stated, “[t]he text of a rule thus proposed and reviewed limits judicial inventiveness,” Amchem, 521 U.S. at 620, and I am mindful that we have previously reversed the district court for allowing parties in this case to amend their complaints under a standard that contradicted Federal Rule of Civil Procedure 16’s plain text, Opiate Litig. 6th Cir. Op. Apr. 15, 2020, 956 F.3d at 844, and violated this court’s holding in Leary v. Daeschner, 349 F.3d 888, 907 (6th Cir. 2003) (determining that “modification [of a scheduling order] is permitted under Rule 16 if Plaintiffs can demonstrate ‘good cause’ for their failure to comply with the original schedule, by showing that despite their diligence they could not meet the original deadline.”) (citing Fed. R. Civ. P. 16 advisory committee’s note to 1983 amendment). Here, there is neither text nor precedent to contravene. Given the absence of language in Rule 23 that bars the district court from distilling “negotiation” from “settlement” or “compromise,” the district court’s certifying a negotiation class is a permissible exercise of “find[ing] efficiencies within the Civil Rules, rather than [violating] them.” Opiate Litig. 6th Cir. Op. Apr. 15, 2020, 956 F.3d at 845. And, by virtue of its innovative nature, the certification of this negotiation class violates not one iota of case law. Indeed, in the same decision where the Supreme Court reminded us that “[c]ourts are not free to amend a rule outside the process Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 30

Congress ordered,” Amchem, 521 U.S. at 620, the Supreme Court upheld the existence of settlement class actions with nary a textual grumble. We should follow the Court’s wake and preserve the class action ecosystem in which district court flexibility thrives.

III.

No doubt, a class action “may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23[] have been satisfied.” Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 161 (1982). The majority, Defendants, and the Six Cities maintain that the district court neglected Rule 23’s essentials when certifying Plaintiffs’ two federal RICO “claim[s]” “under Rule 23(b)(3)” and certifying “issues related to Defendants’ obligations under the Controlled Substances Act” (“CSA”) “under Rule 23(c)(4).”22 Opiate Litig., 332 F.R.D. at 542. Of course, courts certifying Rule 23(b)(3) class actions must comply with the “safeguards” of both Rule 23(a) and Rule 23(b)(3). Amchem, 521 U.S. at 621 (“The safeguards provided by the Rule 23(a) and (b) class-qualifying criteria, we emphasize, are not impractical impediments—checks shorn of utility—in the settlement-class context.”). But the district court’s findings regarding Rule 23(b)(3)’s predominance and superiority requirements and Rule 23(a)’s adequate representation prong conform to the rule’s text and the Supreme Court’s precedent.23

The majority and Defendants hurl an array of challenges at the district court’s finding “that common issues predominate over individualized issues with respect to both the RICO claims and the CSA issues[.]” Opiate Litig., 332 F.R.D. at 551. So, I begin with Rule 23(b)(3)’s first requirement—predominance. For class actions certified under Rule 23(b)(3), district courts

22 The CSA issues are not independent from the RICO claims. Plaintiffs allege that the “factual allegations [arising under the CSA] underlie the second RICO claim [] and are also pertinent to adjudication of myriad state- based legal claims, from public nuisance to negligence.” Opiate Litig., 332 F.R.D. at 542. 23 Rule 23(b)(3) requires that “the court finds that the 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. The matters pertinent to these findings include: (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.” FED. R. CIV. P. 23(b)(3). Rule 23(a)’s “Prerequisites” are that “[o]ne or more members of a class may sue or be sued as representative parties on behalf of all members only if: (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). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 31

must “find[] that the questions of law or fact common to class members predominate over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). Predominance “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation[,]” Amchem, 521 U.S. at 623, which requires “courts to give careful scrutiny to the relation between common and individual questions in a case[,]” Tyson Foods, Inc. v. Bouaphakeo, — U.S. —, 136 S. Ct. 1036, 1045 (2016). “The predominance inquiry ‘asks whether the common, aggregation- enabling, issues in the case are more prevalent or important than the non-common, aggregation- defeating, individual issues.’” Id. at 1045 (quoting Rubenstein, 2 Newberg on Class Actions, supra, § 4:49).

That the RICO claims and CSA issues are common, rather than individual, issues should be uncontroversial, as the district court found that the “allegations concerning the existence of two national enterprises that disseminated a set of standard falsehoods in marketing and distributing opioids[,]” Opiate Litig., 332 F.R.D. at 549, that underlie the RICO claims and CSA issues are “common across many, if not most, of the MDL litigants and putative Class Representatives[,]” id. at 542.24 But Defendants criticize the district court’s “declin[ing] to analyze predominance for the claims of the class as a whole” and “simply eliminat[ing] from consideration the numerous non-common issues and claims, including all the state-law claims asserted by class members.” Defs.’ Br. at 39–40; see also Maj. Op. at 14 (“[T]he court’s order minimized or marginalized the myriad state law claims that arguably divide the putative class

24 Defendants contend that “a large proportion of the class . . . do not even assert RICO claims.” Defs.’ Br. at 40. But there is no evidence that the district court abused its discretion in finding that the RICO claims and CSA issues are common claims. Summit County, Ohio submitted a complaint in this MDL that contained RICO claims and made factual allegations under the CSA. Opiate Litig., 332 F.R.D. at 542; see also Summit County Complaint, No. 1:17-md-02804, R. 513. Summit County’s complaint was the basis of a “short form complaint” process that “enabled all plaintiffs in this MDL to incorporate by reference certain of the legal and factual allegations therein.” Opiate Litig., 332 F.R.D. at 542; see also Op. and Ord. No. 1:17-md-02804, R. 1282. Accordingly, the “vast bulk of the [forty-nine] putative class representatives[,]” adopted the Summit County pleadings, id., and “so many other plaintiffs here have adopted those same claims and issues through the short-form process and/or have filed complaints that are substantially identical in relevant passages to the Summit County complaint[,]” id. n.3; see also Appellees’ Br. at 63 n.21 (“86.3% of class representatives assert RICO claims, 100% assert CSA-related public nuisance claims against Distributor Defendants, and 96.1% against Manufacturer Defendants[.]”). Indeed, Defendants themselves may have conceded that most of the putative class members allege RICO claims. See id. Regardless, the district court’s predominance analysis should not hang on statistics because the predominance inquiry is qualitative, not quantitative. See CHARLES ALAN WRIGHT, ARTHUR R. MILLER, & MARY KAY KANE, 7B FEDERAL PRACTICE AND PROCEDURE § 1778 (3d ed. Apr. 2020 Update)) (collecting cases holding that predominance inquiry is not a quantitative test). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 32

members.”).25 But predominance need not be “established for the entirety of the claims that a class is certified to address[,]” Defs.’ Br. at 44; instead, the Supreme Court has held that “[w]hen ‘one or more of the central issues in the action are common to the class and can be said to predominate, the action may be considered proper under Rule 23(b)(3) even though other important matters will have to be tried separately,’” Bouaphakeo, 136 S. Ct. at 1045 (quoting Charles Alan Wright, Arthur R. Miller, & Mary Kay Kane, 7AA Federal Practice and Procedure § 1778 (3d ed. 2005)). Such “important matters” include actual injury, see Martin v. Behr Dayton Thermal Prods. LLC, 896 F.3d 405, 414 (6th Cir. 2018), cert. denied, 139 S. Ct. 1319 (2019), damages, see Bouaphakeo, 136 S. Ct. at 1045; see also Martin, 896 F.3d at 414, and “some affirmative defenses peculiar to some individual class members[,]” Bouaphakeo, 136 S. Ct. at 1045. Here, any individual issues present in state law claims or RICO’s injury element do not drown the common RICO claims and CSA issues. Yes, “the predominance inquiry must focus on common questions that can be proved through evidence common to the class.” In re Whirlpool Corp. Front-Loading Washer Prods. Liab. Litig., 722 F.3d 838, 858 (6th Cir. 2013) (citing Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 466–468 (2013)). But “Rule 23(b)(3) . . . does not require a plaintiff seeking class certification to prove that each element of her claim is susceptible to classwide proof.” Amgen, 568 U.S. at 469 (2013) (emphasis in original, internal citations and alterations omitted); see also Charles Alan Wright, Arthur R. Miller, & Mary Kay Kane, 7B Federal Practice and Procedure § 1778 (3d ed. Apr. 2020 Update) (“The common questions need not be dispositive of the entire action. In other words, ‘predominate’ should not be automatically equated with ‘determinative.’”) (footnote omitted). Here, Plaintiffs did not need to prove that their state law claims were certifiable for the district court to find that Plaintiffs’ RICO claims and CSA issues are predominant, common claims. Aside from highlighting that the district court did not certify Plaintiffs’ state law claims, Maj. Op. at 14, Defs.’ Br. at 44–46, no one has sufficiently “explain[ed] how . . . [P]laintiff class’s failure to prove an essential element of its claim for relief will result in individual questions predominating over common ones[,]” Amgem, 568 U.S. at 469.

25 It is worth noting that we cannot evaluate whether RICO claims and CSA issues are present in the Six Cities’ claims, as there is no accessible record that the Six Cities filed any complaints in the MDL. See infra note 33. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 33

Contrary to the majority’s and the Defendants’ qualms, see Maj. Op. at 15; Defs.’ Br. at 43, the district court’s certifying factual allegations arising under the CSA per Rule 23(c)(4) is thoroughly legitimate. This court has already addressed the relationship between Rule 23(b)(3)’s predominance inquiry and Rule 23(c)(4), which provides that “an action may be brought or maintained as a class action with respect to particular issues.” Fed. R. Civ. P. 23(c)(4). In Martin v. Behr Dayton Thermal Products LLC, we adopted a “broad view” that “instructs courts to engage in the predominance inquiry after identifying issues suitable for class treatment[,]” which “permits utilizing Rule 23(c)(4) even where predominance has not been satisfied for the cause of action as a whole,” 896 F.3d at 411–13.26 Here, the district court correctly identified that “the nature of each Defendant’s obligations under the [CSA] and the question of whether each Defendant complied with those obligations” were common issues before proceeding with its predominance analysis. Opiate Litig., 332 F.R.D. at 550. Indeed, this court has already considered and rejected a “narrow view” that would “prohibit[] issue classing if predominance has not been satisfied for the cause of action as a whole,” Martin, 896 F.3d at 412; such an approach to predominance “would undercut the purpose of Rule 23(c)(4) and nullify its intended benefits[,]” id. at 413.27

Nor do the majority’s policy objections undermine the district court’s predominance analysis. Mysteriously, the majority concedes that the district court’s “predominance analysis

26 Other circuits have spoken to the interaction between Rules 23(b)(3) and 23(c)(4). The Second, Ninth, Fourth, and Seventh Circuits support the “broad” approach that this court embraced in Martin v. Behr Dayton Thermal Products LLC, 896 F.3d 405 (6th Cir. 2018), cert. denied, 139 S. Ct. 1319 (2019). See Martin, 896 F.3d at 411 (collecting cases). The Fifth Circuit has described the “narrow view” in a footnote. See id. at 412 (citing Castano v. Am. Tobacco, 84 F.3d 734, 745 n.21 (5th Cir. 1996)). Although the Eleventh Circuit “referenced” the narrow view “with tenuous support,” Martin, 896 F.3d at 412 (citing Sacred Heart Health Sys., Inc. v. Humana Military Healthcare Servs., Inc., 601 F.3d 1159, 1176 (11th Cir. 2010)), the Fifth Circuit’s footnote has not been adopted by any other circuit, see Martin, 896 F.3d at 412. “[S]ubsequent caselaw from within the Fifth Circuit itself indicates that any potency the narrow view once held there has dwindled.” Martin, 896 F.3d at 412 (citing Steering Comm. v. Exxon Mobil Corp., 461 F.3d 598, 603 (5th Cir. 2006)). 27 Defendants also argue that “[i]ssue certification under Rule 23(c)(4) is available only when a trial is contemplated.” Defs.’ Br. at 43. But, like the rest of Rule 23’s language, nothing in Rule 23(c)(4)’s text limits the use of issue certification to classes certified for the purposes of trial. See FED. R. CIV. P. 23(c)(4). Nor is there any precedent in this circuit or any other circuit that suggests otherwise. It is worth bearing in mind that the “vast majority” of litigation class actions yield settlements and not judgments on the merits. Geoffrey P. Miller, Rethinking Certification and Notice in Opt-Out Class Actions, 74 UMKC L. Rev. 637, 640 (2006); see also William B. Rubenstein, 3 NEWBERG ON CLASS ACTIONS § 9:52 (5th ed.) (“[M]ost class actions today are settlement class actions in any case[.]”). Thus, issue certification is not fundamentally incompatible with class actions that involve negotiations and produce settlements. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 34

might make some sense in the context of a litigation class or settlement class limited to resolving the narrow federal issues cited in the certification order.” Maj. Op. at 14. No court has articulated different predominance tests for classes certified for different purposes, and we should not open that Pandora’s box here. The majority also worries that certifying only the federal RICO claims and CSA issues “creates confusion surrounding the scope of negotiations[,]” specifically “whether, and how, the negotiation class representatives, empowered by the court, will address their state law claims during settlement discussions.” Id. But it is unclear how the negotiation class creates any more uncertainty about the scope of negotiations than does a litigation class, where a court certifies certain issues for litigation but negotiations during the suit may yield a global settlement, or a settlement class, where putative class members receive notice of settlement without having participated in negotiations at all.

Of course, district courts ought to exercise “caution” in finding predominance in mass torts cases, particularly “when individual stakes are high and disparities among class members great.” Amchem, 521 U.S. at 625. In the asbestos litigation, for example, the Supreme Court determined that it was impermissible to find predominance based entirely on “hundreds of thousands, perhaps millions, of individuals[’,]” Amchem, 521 U.S. at 597, “shared experience[,]” id. at 622, of asbestos exposure because putative class members “were exposed to different asbestos-containing products, in different ways, over different periods, and for different amounts of time; some suffered no physical injury, others suffered disabling or deadly diseases[,]” id. at 609. The Court buoyed these concerns in a later case when holding that a class of 1.5 million women who alleged that their employer systematically discriminated against them on the basis of sex did not present “questions of law or fact common to the class” as required by Rule 23(a)(2); the Court determined that the women “identified no ‘specific employment practice’—much less one that ties all their 1.5 million claims together[,]” Dukes, 564 U.S. at 357, and that the women “held a multitude of jobs, at different levels of Wal-Mart’s hierarchy, for variable lengths of time, in 3,400 stores, sprinkled across 50 states, with a kaleidoscope of supervisors . . . , [and] subject to a variety of regional policies that all differed[,]” id. at 359–60 (internal citation omitted). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 35

But “mass tort cases arising from a common cause or disaster may, depending upon the circumstances, satisfy the predominance requirement[.]” Amchem, 521 U.S. at 625 (citing Fed. R. Civ. P. 23 advisory committee’s note to 1966 amendment). Here, Plaintiffs are not tens of thousands (or millions) of persons who have highly individualized experiences with and “health consequences of[,]” id. at 624, asbestos exposure who attempt to satisfy predominance based on “shared experience” alone, cf. id. at 622–26. Nor are Plaintiffs unable to point to a “specific” or “uniform” practice that would provide the basis for predominance per Rule 23(b)(3) or commonality per Rule 23(a)(2). Cf. Dukes, 564 U.S. at 355–60. Plaintiffs are cities and counties that accuse Defendants of instituting an unlawful marketing scheme and an illegal supply chain that violated RICO and the CSA. See Opiate Litig., 332 F.R.D. at 542. These claims are not predicated on the highly individualistic experiences of specific persons within the cities and counties who have been impacted by the opioid crisis, nor are they predicated on the degree of harm supposedly caused by Defendants; whether these two schemes existed or not applies to all class member cities and counties with commensurate force. Because “the same evidence will suffice for each member to make a prima facie showing or the issue is susceptible to generalized, class-wide proof[,]” Bouaphakeo, 136 S. Ct. at 1045 (internal citations and alterations omitted), the district court did not abuse its discretion in finding predominance here.

Defendants home in on RICO’s causation element, arguing that “proving causation for RICO claims typically necessitates individualized proof.” Defs.’ Br. at 40. In other contexts, whether harm was caused by individuals’ reliance on alleged fraud may be an individual question that would predominate over common questions. Not so for RICO. Defendants’ argument is at odds with both the Supreme Court’s and our precedent. In Bridge v. Phoenix Bond & Indemnity Co., 553 U.S. 639 (2008), the Supreme Court held that those who bring mail fraud claims under RICO do not have to demonstrate that they relied on the fraud to establish that their injuries were caused by the fraud. See id. at 649 (“[N]o showing of reliance is required to establish that a person has violated § 1962(c) by conducting the affairs of an enterprise through a pattern of racketeering activity consisting of acts of mail fraud.”). Because “[f]or RICO purposes, reliance and proximate cause remain distinct,” Plaintiffs “need only show use of the mail in furtherance of a scheme to defraud and an injury proximately caused by that scheme.” Wallace v. Midwest Fin. & Mortg. Servs., Inc. 714 F.3d 414, 420 (6th Cir. 2013). RICO’s Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 36

softened causation element “largely dooms the Defendants’ attempt to identify individual issues of causation sufficient to preclude a finding of predominance.” Torres v. S.G.E. Mgmt., L.L.C., 838 F.3d 629, 638 (5th Cir. 2016); see, e.g., id. at 638–40 (finding predominance existed when pyramid scheme victims sought to certify RICO claims because victims’ being “necessary to the scheme[,]” “direct victims of the scheme,” and “foreseeable victims of the alleged fraud” satisfied RICO’s causation element).

The district court’s treatment of Rule 23(b)(3)’s second requirement—superiority—is another source of consternation for the majority. But the district court’s analysis of whether “a class action is superior to other available methods for fairly and efficiently adjudicating the controversy” per Rule 23(b)(3)’s four enumerated factors, see Fed. R. Civ. P. 23(b)(3)(A)–(D), does no “obvious violence to the Rule’s structural features.” Maj. Op. at 15 (quoting Dukes, 564 U.S. at 363). The district court was correct that it need not assess “[t]he likely difficulties in managing a class action” required by Rule 23(b)(3)(D) because the proposed certification “is not for litigation or trial, but simply for settlement negotiations,” Opiate Litig., 332 F.R.D. at 551– 52; this finding is in lockstep with Supreme Court precedent, see Amchem, 521 U.S. at 620 (concluding that “[c]onfronted with a request for settlement-only class certification, a district court need not inquire whether the case, if tried, would present intractable management problems, for the proposal is that there be no trial[]”) (citing Fed. R. Civ. P. 23(b)(3)(D)). Still, the majority frets that the district court seems to have bypassed the manageability analysis that Rule 23(b)(3)(D) requires of litigation classes “without halting the underlying litigation.” Maj. Op. at 14. But the majority reads too much into the district court’s permitting parties to this suit to litigate claims in tandem with the negotiation class’s existence. In any Rule 23(b)(3) class action—litigation, settlement, or otherwise—putative class members are always entitled to opt out and pursue their day in court or to conduct settlement negotiations on their own terms. See Dukes, 564 U.S. at 367 (“Because the Rules Enabling Act forbids interpreting Rule 23 to abridge, enlarge or modify any substantive right, a class cannot be certified on the premise that [Defendant] will not be entitled to litigate its statutory defenses to individual claims.”) (internal citations omitted). Because no kind of class certification “halt[s] the underlying litigation,” Maj. Op. at 14, until settlement is reached, we need not fear any “violence,” id. at 15, against Rule 23’s structure here. Moreover, we should hesitate to overstate the significance of the Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 37

manageability prong, which will rarely warrant denial of certification of a class. See In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 140 (2d Cir. 2001) (Sotomayor, J.), abrogated on other grounds by In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24 (2d Cir. 2006); see also Rubenstein, 2 Newberg on Class Actions, supra, § 4:72.

The majority further assails the district court’s analysis of the factor enumerated in Rule 23(b)(3)(A), which stipulates that “the class members’ interests in individually controlling the prosecution or defense of separate actions” is a “pertinent” matter.” Fed. R. Civ. P. 23(b)(3)(A). Disagreeing with the district court’s assessing that a negotiation class “favors individual control and involvement” by engaging class members in the negotiation and voting process, Opiate Litig., 332 F.R.D. at 551, the majority determines that the negotiation class “requir[es] class action participants to commit to the negotiation class without knowing the issue parameters or the amount or prospect of any potential recovery[,]” Maj. Op. at 15. These concerns are akin to the Six Cities’ agonies about the negotiation class’s opt-out timeline, which I address below and need not do so again here. In short, this straw man skims over the features of the negotiation class that provide class members with more input and control over settlement negotiations than they would have in a traditional litigation class or settlement class. See Part IV, infra.

Turning away from Rule 23(b)(3), we must also train our sights on the district court’s compliance with Rule 23(a), as “[a]ny class certification must satisfy Rule 23(a)’s requirement of numerosity, commonality, typicality, and adequate representation.” Clemons v. Norton Healthcare Inc. Ret. Plan, 890 F.3d 254, 278 (6th Cir. 2018).28 When certifying a class, a district court must properly find that “the representative parties will fairly and adequately protect the interests of the class[,]” as required by Rule 23(a)(4), see Fed. R. Civ. P. 23(a)(4), according to this circuit’s two-pronged adequacy-of-representation test. Young v. Nationwide Mut. Ins. Co., 693 F.3d 532, 543 (6th Cir. 2012) (“1) [T]he representative must have common interests with unnamed members of the class[;] and 2) it must appear that the representatives will vigorously prosecute the interests of the class through qualified counsel.”) (citing In re Am. Med.

28 The Appellants preserve only their arguments regarding adequacy of representation. Although we need not reach the questions of whether the district court’s numerosity, commonality, and typicality findings were proper, the record reveals that the district court did not abuse its discretion in making its determinations for these three Rule 23(a) requirements. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 38

Sys., Inc., 75 F.3d 1069, 1083 (6th Cir. 1996)). Appellants take aim at two aspects of the district court’s adequacy finding: financial conflicts and subclasses. Both missiles miss their mark.

No nefarious financial conflicts have erupted in this case yet. Defendants aver that the negotiation class representatives are “not required to prioritize prosecution of the interests of the class in negotiating a settlement over and above their own litigation interests,” Defs.’ Br. at 47, because “they remain free to pursue litigation of [their own] suits without considering the interests of absent class members,” Defs.’ Br. at 48. In sync, the Six Cities chide the district court for creating conflicts of interest by “put[ting] overwhelming financial pressure on counsel to settle the case” and “convey[ing] . . . disinclination to tolerate class litigation if the negotiations fail.” Six Cities’ Br. at 39. But Appellants err in their focus—“the adequacy requirement focuses on the desired attributes of those who seek to represent the class as opposed to the characteristics of the class.” Rubenstein, 1 Newberg on Class Actions, supra, § 3:50. Here, Appellants’ allegations of financial conflict are far too hypothetical and speculative; “[o]nly when attacks on the credibility of the representative party are so sharp as to jeopardize the interests of absent class members should such attacks render a putative class representative inadequate.” Gooch v. Life Inv’rs Ins. Co. of Am., 672 F.3d 402, 431 (6th Cir. 2012) (internal citation omitted); see also Wright, Miller, & Kane, 7B, supra, at § 1768 n.2 (collecting cases finding that mere speculation that conflicts might develop were insufficient to demonstrate inadequate representation). Appellants’ unsubstantiated accusations that class representative cities are willing to throw unnamed class members under the bus are distinct from contexts where named parties’ steep financial conflicts clearly and presently endangered the interests of other class members. Cf., e.g., Vassalle v. Midland Funding LLC, 708 F.3d 747, 757 (6th Cir. 2013) (determining financial conflict existed when settlement would forgive debts of only class representatives because representatives had interest in ensuring settlement whereas unnamed class could contest their debts only in court); Ortiz v. Fibreboard Corp., 527 U.S. 815, 852 (1999) (determining impermissible conflict of interest in Rule 23(b)(1)(B) class action when Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 39

named plaintiffs’ counsel negotiated separate settlement of 45,000 pending claims, the full payment of which was contingent on successful global settlement agreement).29

Whether the district court should have divided these cities and counties into subclasses, see Fed. R. Civ. P. 23(c)(5), demands closer scrutiny. But, query, what subclasses are Appellants proposing? In neither their memorandum opposing certification nor their brief here do the Six Cities explicate how the district court should have severed the class. By explaining that the cities and counties “operate in vastly different factual and legal contexts, spread across all fifty States,” Six Cities’ Br. at 34, perhaps Appellants suggest the class should be split on a geographic basis. Ignoring, for now, that Appellants provide no specifics about how this division should work, geography or differences in state law are not compelling reasons to use subclasses here. A class need not be subdivided merely because certain claims may be more difficult to prove in some states over others due to geography. See In re Deepwater Horizon, 739 F.3d 790, 814 (5th Cir. 2014) (finding subclasses to be unnecessary even though “‘causation becomes more difficult for a claimant [state] to establish the further one moves from the coast” and, in particular, the further one moves from the Macondo reservoir where the Deepwater Horizon incident occurred”). Likewise, “differently weighted interests” do not mandate subclasses. Gooch, 672 F.3d at 429. Here, multiple factors massage any geographical or legal rifts, see Deepwater Horizon, 739 F.3d at 813–14: the class was certified based on their federal claims—not their state claims—that nation-wide marketing and distribution schemes existed; the

29 Perhaps this is not the financial conflict that we ought to be concerned with here. Plaintiffs allege that the Six Cities “are represented by the same counsel, who previously sought and was denied a leadership position representing all cities and counties and, who, curiously, is co-counsel with [Plaintiffs’ Executive Committee] members in representing other Class Members who do not oppose the Class.” Pls.’s Reply Br. Further Supp. Renewed and Am. Mot. Certification of Rule 23(b)(3) Cities/Counties Negotiation Class, No. 1:17-md-02804-DAP, R. 2076 at 18 n.16 (Page ID #286379). If Plaintiffs’ allegations are true, the implications are deeply unpalatable. (It does appear that the Six Cities’ counsel did seek, and was denied, a position on the Plaintiffs’ Executive Committee.) See Mot. for Leave to File MSP Plaintiffs’ Motion for Appointment to a Position on the Pls.’ Executive Committee, No. 1:17-md-02804, R. 578.). Equally concerning: under its contract with counsel, at least one of the Six Cities “will not pay a legal fee unless it wins a monetary award” but “would give its [counsel] 25[%] of any settlement, verdict[,] or recovery of any kind.” Bob Sandrick, North Royalton joins litigation over opioid epidemic, CLEVELAND.COM (Jan. 30, 2019), https://www.cleveland.com/north- royalton/2018/04/north_royalton_joins_litigatio html. This is a much larger share than the 10% set aside for class counsel under the negotiation class’s allocation proposal. See Pls.’s Notice of Mot. and Mot. for Certification of Rule 23(b)(3) Cities/Counties Negotiation Class, No. 1:17-md-02804, R. 1683 at 10–12 (Page ID #46950–52). Amidst these black pots and kettles, we should refrain from finding that class counsel is the source of financial conflicts of interest in the present case. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 40

negotiation class’s proposed allocation plan addresses differences between counties in the number of persons suffering opioid use disorder, number of opioid deaths, amount of opioids distributed, see Pls.’s Corrected Notice of Mot. and Mot. for Certification of Rule 23(b)(3) Cities/Counties Negotiation Class, No. 1:17-md-02804, R. 1690-1, at 49–50 (Page ID #47150– 51), and amount spent on previously filed opioid litigation, see id. at 44 (Page ID #567144); and the class’s voting plan—which requires approval by 75% of six distinct groups—accounts for the cities’ and counties’ varying population sizes,30 see id. at 8–9 (Page ID #47091–92).

Finally, the Six Cities have an obvious champion among the named Plaintiffs. The City of Lakewood sits in Cuyahoga County, where five of the Six Cities are located; Lakewood’s income and racial diversity is also on par with that of various of the Six Cities’31 and Lakewood’s population (51,382) does not dwarf that of North Royalton (30,315).32 Cities and Counties, In Re: National Prescription Opiates Litigation, https://www.opioidsnegotiationclass.info/ (last visited Sept. 14, 2020). Indeed, the forty-nine named counties and cities are spread across nearly every state and are of every size and variety. See id. The Six Cities need not rely solely on Lakewood to campaign for their interests as they also share similar population sizes with several other named plaintiffs, including Goodling County, Idaho (15,216) and Camden County, Georgia (51,402). See id. Even in the absence of any clear articulation of what geography- or law-based subclasses would look like, it is difficult

30 The six categories are: (1) “75% of the total number of cities and counties that filed suit as of June 14, 2019”; (2) “75% of the total number of cities and counties that did not file suit as of June 14, 2019”; (3) “75% of the total voting population of all cities and counties that filed suit as of June 14, 2019”; (4) “75% of the total voting population of all cities and counties that did not file suit as of June 14, 2019”; (5) “75% of the litigating entities, weighted by their allocations as shown on the Settlement Allocation Lookup Tool to be posted at Opioidsnegotiationclass.com”; and (6) “75% of the non-litigating entities, weighted by their allocations as shown on the Settlement Allocation Lookup Tool to be posted at Opioidsnegotiationclass.com.” Pls.’s Corrected Notice of Mot. and Mot. for Certification of Rule 23(b)(3) Cities/Counties Negotiation Class, No. 1:17-md-02804, R. 1690, at 8–9 (Page ID #47091–92). 31 Compare QuickFacts, Lakewood city, Ohio, UNITED STATES CENSUS BUREAU, https://www.census.gov/quickfacts/lakewoodcityohio (last visited Sept. 14, 2020) (census data for Lakewood, OH), with Compare QuickFacts, Wickliffe city, Ohio; Huron city, Ohio; Lyndhurst city, Ohio; Mayfield Heights city, Ohio; East Cleveland city, Ohio; North Royalton city, Ohio, UNITED STATES CENSUS BUREAU, https://www.census.gov/quickfacts/fact/table/wickliffecityohio,huroncityohio,lyndhurstcityohio,mayfieldheightscity ohio,eastclevelandcityohio,northroyaltoncityohio/PST045219 (last visited Sept. 14, 2020) (census data for the Six Cities). 32 Cities and Counties, IN RE: NATIONAL PRESCRIPTION OPIATES LITIGATION, https://www.opioidsnegotiationclass.info/ (last visited Sept. 14, 2020). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 41

to comprehend how the spread of named representatives here does not protect the interests of the Six Cities or other class members.

Or perhaps the Six Cities demand that the class should be severed by interest. But the district court has already contemplated whether it was appropriate to divide class members by the most obvious axis of interests; because all named class members are “litigating entities,” Opiate Litig., 332 F.R.D. at 539 (internal quotation marks omitted), who had “filed lawsuits as of June 14, 2019[,]” id., the district court tasked Special Master Yanni to investigate whether they adequately represented “non-litigating entities,” id., who had not filed lawsuits by that date, see Order Directing Special Master Yanni To Assess Fairness of Allocation and Voting Proposals To Non-Litigating Entities, No. 1:17-md-02804, R. 2529. Acknowledging that litigating class members may want to “settle fast to recover their outlay of costs” whereas “non-litigating class members might want to settle fast and cheap[,]” Special Master Yanni found that “[t]his is true of class members in all class actions and, standing alone, does not create any meaningful, much less fundamental intra-class conflict.” Rep. of Special Master Cathy Yanni, No. 1:17-md-02804, R. 2579 (Page ID #413008). Based on the negotiation class’s transparency regarding allocation issues and the voting plan, Special Master Yanni found that “there are no conflicts of interest, much less fundamental ones, between litigating and non-litigating class members as to negotiating settlements with defendants” and that the “fact that all of the proposed class representatives are litigating entities does not render them unable to represent the interests of the non-litigating class members in settlement negotiations.” Id. at 15 (Page ID #413008). The district court adopted her “thoughtful and thorough” findings. Opiate Litig., 332 F.R.D. at 553. In short, the district court has carefully considered any conflict of interests that may have existed along the most obvious tilt—litigating versus non-litigating entities—and, without any suggestion of abuse of discretion, we must uphold the district court.

But the Six Cities do not draw their battle lines based on the district court’s deciding not to create subclasses based on litigating versus non-litigating status. Perhaps this is because all Six Cities present themselves as litigating entities—although they fail to clarify this at any point in their brief—whose interests would be adequately represented by all of forty-nine named Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 42

Plaintiffs, who are, again, all litigating entities.33 But where the Six Cities would have us draw the lines is a riddle: their brief does not explain how to slice this class by interests, and their articulation of which class members’ interests conflict with each other ambiguously relies on pure conjecture. See Six Cities’ Br. at 34–35 (“a significant aspect of any settlement may be a commitment by industry members to . . . self-regulat[e]”; “promises of industry reform may be worth very little”); “maximizing Defendants’ obligations to self-regulate may be a higher priority”) (emphasis added). All of these theoretical interests plague any class action and are not an obvious source of conflicting interests. See Geoffrey P. Miller, Conflicts of Interest in Class Action Litigation: An Inquiry into the Appropriate Standard, 2003 U. Chi. Legal F. 581, 581 (2003) (“Because of the large numbers of claims and the potential for members of the class to be differently situated with respect to particular issues, tensions among class members are common, even ubiquitous.”). Of course, objectors need not propose subclasses for their use to be proper. See, e.g., Amchem, 521 U.S. at 603 (concluding that subclasses were warranted even though “[t]he complaint delineated no subclasses; all named plaintiffs were designated as representatives of the entire class.”). But apart from litigating members’ and non-litigating members’ differing interests, it is difficult to comprehend what other interests would be so fundamentally incompatible as to undermine representation here—a difficulty that is evident in the Six Cities’ nebulous arguments.

33 In their brief, the Six Cities never clarify if they have filed a suit in the MDL, whether before or after June 14, 2019. Further, there is no accessible record in the district court docket that any of the Six Cities have filed any complaints in this matter. My understanding that they present themselves as “litigating entities” is based on the Six Cities’ counsel’s representations to the district court. See Transcript Procs. Before Hon. Dan Aaron Polster Tues., Aug. 6, 2019, No. 1:17-md-02804-DAP, R. 2147, at 15:19–16:15 (Page #ID 288093–94) (“THE COURT: All right. Which cities are you -- are the litigating cities or non-litigating cities? MR. FERRARO: Litigating cities. THE COURT: Which cities? MR. FERRARO: One is East Cleveland. THE COURT: And what other cities do you represent? MR. FERRARO: I know I don’t have the list in front of me but it’s six. THE COURT: All right. MR. FERRARO: And -- THE COURT: You mentioned Florida and I didn’t -- wasn’t following you. I thought you were representing Ohio -- you’re objecting on behalf of six -- seven Ohio cities? MR. FERRARO: I have the list, by the way. It’s the City of North Royalton, Ohio. I mentioned the City of East Cleveland, the City of Mayfield Heights, City of Lyndhurst, City of Huron, the City of Wick -- life. THE COURT: Wickliffe. MR. FERRARO: Wickliffe. Thank you, your Honor.”) (emphasis added); id. at 20:11–20:16 (Page ID #288098) (THE COURT: Have all seven of these cities filed lawsuits? MR. FERRARO: I’m sorry? THE COURT: Have all seven of the Ohio cities you represent -- MR. FERRARO: Yes.”); id. at 21:6–25 (Page ID #288099) (“THE COURT: . . . . You’re in a litigating, represent seven litigating cities. . . . So I’m just pointing out that there are roughly 2000 litigating cities that appear to have no objection to this. There’s seven who do. All right. But, those are the facts. MR. FERRARO: You look at the fraction, which is it’s less than 2000 but a small fraction of the 33,000 that may go into this blind settlement or be bound by -- THE COURT: You don’t represent any of those, do you? MR. FERRARO: Other than my seven, no, I am not. THE COURT: Okay. So I don’t think you have standing to raise an objection on behalf of the other 30,000 cities and counties. MR. FERRARO: All right.”). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 43

Failing to conjure up a cogent theory of how even one specific conflict of interest mandates subclasses, the Six Cities stress that “there will also be differences in interests between class members regarding the relative importance of addressing past versus future harm.” Six Cities’ Br. at 35. But time does not part the sea here, as all class members have both past and future harms of uniform kind. In the context of the asbestos litigation, the Supreme Court rightfully expressed horror over certifying a “single giant class,” Amchem, 521 U.S. at 626, for the “elephantine mass of asbestos cases,” Ortiz, 527 U.S. at 821. That “currently injured” asbestos victims maintained “the critical goal [of] generous immediate payments” whereas “exposure-only plaintiffs” had an interest “in ensuring an ample, inflation-protected fund for the future[,]” sprouted a deep conflict of interest. Amchem, 521 U.S. at 626; see also Ortiz, 527 U.S. at 819 (“[A] class including holders of present and future claims (some of the latter involving no physical injury and claimants not yet born) requires division into homogeneous subclasses under [then] Rule 23(c)(4)(B), with separate representation to eliminate conflicting interests of counsel.”). In the asbestos cases, there was an unambiguous wall between plaintiffs—who were either firmly in the “already injured” camp or squarely in the “not-yet-injured” camp—that created fundamentally incompatible interests in desiring present versus future relief. But here, Plaintiff counties and cities allege that nation-wide fraudulent marketing and distribution schemes have already injured their communities. Any differences lie in the magnitude of harm sustained and not in the kind of interests. Indeed, the Six Cities concede that all Plaintiffs have already sustained injury, as their analogy to the asbestos litigation’s distinction between past versus future claims is the interests held by the “jurisdictions hardest hit” by the opioid crisis and those of the “jurisdictions that have suffered comparatively less harm to date[.]” Six Cities’ Br. at 35 (emphasis added). And even if there are counties and cities in the class that, somehow, have not yet sustained any injury whatsoever from the opioid crisis—which even the Six Cities do not deign to assert—the possibility that some class members may not prevail on the merits of their individual claims does not defeat class membership. See Deepwater Horizon, 739 F.3d at 813 (“Class certification is not precluded simply because a class may include persons who have not been injured by the defendant’s conduct.”).

The only “future interest” that can be extrapolated from the Six Cities’ brief is that it “may be a higher priority” for “jurisdictions that have suffered comparatively less harm to date” Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 44

to “maximiz[e] Defendants’ obligations to self-regulate” or “restrict[] the scope of any waiver of claims for future injury.” Six Cities’ Br. at 35; see also Defs.’ Br. at 49–50. Ignoring, again, the speculative nature of these supposed future interests, Appellants’ distinguishing between past and future interests has no place in this case because all class members will suffer future harm. Unlike in the asbestos litigation, where individuals who were already injured by asbestos had no pecuniary interest whatsoever in creating a fund for those who have not been injured, every county and city in this country—even those that have been hit harder than others—share a uniform future interest in ensuring that opioid manufacturing and distribution will not affect their communities going forward. Again, any differences are of magnitude, not of kind. It is patently inconceivable that any city or county has categorically defeated the opioid epidemic and needs no financial support fighting the crisis going forward. See, e.g., Summit County Complaint, No. 1:17-md-02804, R. 513 at 225 (estimating that Summit County has already spent $66 million on costs related to the opioid crisis and will likely spend another $89 million over the next five years). This case is more analogous to Gooch v. Life Investors Insurance Co., where this court examined the adequacy of representation of a class that consisted of persons who have filed, or might file, medical claims, 672 F.3d at 429. In Gooch, the Defendant insurance company claimed that a conflict of interest persisted between persons with premium waivers, who had only an interest in past insurance payments, and persons without such waivers, who also bore an interest in attaining manageable premium rates. See id. But we rejected any overtures to divide the class by past and future interests, reasoning that those with premium waivers still had “some prospective financial interest in keeping [their] own premiums low” and that all class members “have an interest in maximal payment for [their medical] claims.” Id. at 429–30. Similarly, by virtue of the ongoing nature of the opioid epidemic, cities and counties that have already sustained monetary losses maintain a prospective financial interest in ensuring that any settlement properly accounts for future interests. In sum, even a generous extrapolation of interests from the Six Cities’ hazy arguments does not require the district court’s balkanizing this class into subclasses. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 45

IV.

Having established that history, text, and purpose empower the district court to certify a class for negotiation purposes and that the district court’s reasoning properly adhered to Rule 23’s requirements, I now inspect the Six Cities’ baseless policy and constitutional grievances about the negotiation class’s opt-out timeline.

The procedure for opting out of the negotiation class is akin to that of a litigation class— plaintiffs can decide to opt out after class certification but before a settlement is proposed. See Opiate Litig., 332 F.R.D. at 537 (“The idea is to undertake the class certification and opt-out process prior to a settlement being reached, as is done in a normal class action geared toward trial.”). Should negotiations produce a proposed settlement, Rule 23(e) gives the district court “discretion” to grant class members a second opportunity to opt out. Fed. R. Civ. P. 23(e)(4) advisory committee’s note to 2003 amendment.34 To the Six Cities, this order of operations “disarms” them of two sabers from the litigation class armory and two rapiers from the settlement class arsenal. Six Cities’ Br. at 3. First, they argue that litigation class members can “negotiate[e] from a position of strength” because they can “credibly threaten class litigation (with commensurate potential liability)[.]” Id. at 24–25. Second, they assert that plaintiffs and defendants in a litigation class action have an “incentive to ensure fair and adequate treatment for all class members” because “negotiations take place against the backdrop of the district court’s authority to allow class members another opt-out opportunity after the settlement is negotiated.” Id. at 25. Third, they aver that “[Rule 23] allows [settlement] class members to review a settlement and opt out of the class if they are dissatisfied with its terms.” Id. at 26. Fourth, they allege that “[settlement] class members are put to that opt-out choice only after the district court has determined, based on a thorough review of the final settlement proposal, that it is likely to approve the proposal and certify the class.” Id. at 26 (citing Fed. R. Civ. P. 23(e)(1)(B)); see also Maj. Op. at 16 (“The fact that negotiation class members in the present case must choose whether to opt-out before knowing the settlement amount only heightens our concern in the

34 The text of Rule 23(e)(4) reads: “New Opportunity to be Excluded. If the class action was previously certified under Rule 23(b)(3), the court may refuse to approve a settlement unless it affords a new opportunity to request exclusion to individual class members who had an earlier opportunity to request exclusion but did not do so.” FED. R. CIV. P. 23(e)(4). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 46

present case.”). The Six Cities concede that the former two features are lacking in settlement classes and that the latter two are nonexistent in litigation classes. See Six Cities’ Br. at 24–28; see also id. at 24 (“The two existing methods for settling class claims provide a balanced mix of advantages and disadvantages for plaintiffs and defendants[.]”). But they contend that the negotiation class’s opt-out timeline unacceptably divests class members of all four protections. Therefore, the Six Cities grumble, class members are trapped in the “worst of both worlds,” id. at 3—they lack both leverage in negotiations and an escape route from an unfavorable settlement, see id. at 10 (“The negotiation class is just a litigation class with no power to litigate, or a settlement class deprived of its opt-out rights.”).

This is a misleadingly simplistic illustration of how class actions operate. Negotiation class members are not wedged between leaving the class entirely and accepting a suboptimal settlement; another avenue grants material clout to all class members. Under Rule 23(e)(5), class members can file objections with the court to oppose the terms of a proposed settlement or raise other issues with the class action prior to a settlement’s approval. See Fed. R. Civ. P. 23(e)(5).35 This route is familiar to the Six Cities; they filed an objection with the district court that asserted that the negotiation class inadequately accounted for the competing interests of class members and that the opt-out procedure is unconstitutional. Opiate Litig., 332 F.R.D. at 553 (acknowledging Mem. Opp’n Renewed and Amended Mot. for Certification of Rule 23(b)(3) Cities/Counties Negotiation Class, No. 1:17-md-02804, R. 1958, as an “objection to the plan[.]”). Subsequently, the district court appointed an independent Special Master, see Order Directing Special Master Yanni, No. 1:17-md-02804, R. 2529,36 who investigated and addressed whether the proposed allocation method would distribute a settlement equitably among the class

35 Rule 23(e)(5)(A) provides that “[t]he objection must state whether it applies only to the objector, to a specific subset of the class, or to the entire class, and also state with specificity the grounds for the objection.” FED. R. CIV. P. 23(e)(5)(A). 36 Specifically, the district court asked Special Master Yanni to address “a. whether the proposed class action allocation method, allotting up to 10% of all recoveries to litigating entities’ attorney’s fees and up to another 15% to litigating entities’ expenses is, from the perspective of the non-litigating entities, consistent with Rule 23(e)(2)(D)’s requirement that a settlement be distributed equitably among the class members; b. whether the proposed voting schemes, with three sets of votes counted separately among litigating entities and non-litigating entities, treats the latter group fairly; and, c. whether the proposed Class Representatives and Class Counsel can adequately represent the entire class, including non-litigating entities, as required by Rule 23(a)(4) and Rule 23(g), respectively.” Order Directing Special Master Yanni To Assess Fairness of Allocation and Voting Proposals To Non-Litigating Entities, No. 1:17-md-02804, R. 2529, at 4–5 (Page ID #408988–89). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 47

members and whether the proposed representatives could adequately represent the whole class, see Rep. of Special Master, No. 1:17-md-02804, R. 2579. Citing the report’s findings, the district court responded to the Six Cities’ objection. See Opiate Litig., 332 F.R.D. at 553. The court found that “[n]othing in the allocation model appears to skew toward any group other than those hardest hit by the opioid epidemic[,]” id., and that the negotiation class’s proposed “Special Needs Fund” sufficiently accounted for the Six Cities’ concerns, see id. In short, the objections procedure went off without a hitch: a putative class member exercised their right to air their grievances about the class certification process via an objection and the district court— enlisting an independent special master to assist in this case—extensively investigated and addressed their complaints. This is precisely how courts should accommodate class members’ frustrations when certifying a class or when making other decisions in a class action.37 Indeed, as parties to the settlement, the Six Cities, and any other class members, can continue to file objections going forward.38 Threatening to opt out or actually opting out was not the Six Cities’ only option; we should not ignore that The Six Cities know how to wield and swing a powerful tool—objector status—that is available for class members who wish to mold the certification process.39

37 See MANUAL FOR COMPLEX LITIGATION, FOURTH, § 21.643 (“Objectors can provide important information regarding the fairness, adequacy, and reasonableness of settlements. Objectors can also play a beneficial role in opening a proposed settlement to scrutiny and identifying areas that need improvement.”). 38 See id. Those who claim the mantle of “objector” usually do so in the context of opposing the terms of a proposed settlement. See Edward Brunet, Class Action Objectors: Extortionist Free Riders or Fairness Guarantors, 2003 U. CHI. LEGAL F. 403, 407 n.22 (2003); see also Bruce D. Greenberg, Keeping the Flies Out of the Ointment: Restricting Objectors to Class Action Settlements, 84 ST. JOHN’S L. REV. 949, 950 (2010) (“‘[A]bsent class members’ are permitted to object that a proposed settlement is not fair, reasonable, and adequate.”). This form of objection is compatible with the language of Rule 23(e)(5). See FED. R. CIV. P. 23(e)(5)(A) (“Any class member may object to the proposal if it requires court approval under [Rule 23(e)].”) However, class action objectors “can emerge earlier than the exact time of a proposed settlement[,]” Brunet supra, at 474 n.22; objectors may oppose the definition of a class at the time of the Rule 23 certification hearing, see id.; object to the form, content, or timing of notice, see Greenberg, supra, at 970–72; or object to class counsel fees, see id. at 973–76. Theoretically, the Six Cities could have raised any number of objections under Rule 23(e)(5) even though a settlement has not been proposed in this case. Further, class members’ filing an adequately formal and detailed objection with the district court preserved their right to raise objections on appeal. See WRIGHT, MILLER & KANE, 7B, supra note 24, § 1797.4 (“[T]he failure to formally object to a settlement at the trial level will preclude a class member from raising objections on appeal.”). 39 See Transcript, No. 1:17-md-02804-DAP, R. 2147, at 14:23–15:8 (Page ID #288092–93) (Mr. FERRARO: [W]e filed an objection on the Plaintiff’s side on behalf of the handful of Ohio cities. . . . I was an Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 48

Even if we accept hypothetically that class members can meaningfully influence a class action only by opting out or threatening to do so, the Six Cities spin an untenable narrative about how the negotiation class’s opt-out timeline disembowels them of any of the benefits that they would have been accorded in a litigation class or in a settlement class. A consistent theme emerges: the Six Cities assert that the negotiation class takes a readily available option off the table, and they embellish the practical significance of the option’s absence. Consider the Six Cities’ claim that the negotiation class “divests the class representatives of their usual negotiating leverage by depriving them of the right to proceed to litigation as a class if Defendants’ settlement demands are unreasonable.” Six Cities’ Br. at 10. No such “right” is deprived here.40 The district court’s certifying a negotiation class does not bar the Six Cities, or any other class members, from pursuing settlement or litigation on their own terms, regardless of whether they choose to remain in or exit from the class. See Opiate Litig., 332 F.R.D. at 537; Ord. Clarifying Negotiation Class Certification Ord., Nos. 1:17-md-02804-DAP, 1:18-op-45090, R. 2713, at 2–3 (Page ID #419213–14).41 Of course, solo litigation does not carry as much weight as class litigation. But if the Six Cities really believed that a litigation class was the best vehicle for this suit, they could have proposed a litigation class before or as an alternative to the motion to certify a negotiation class. Indeed, the Six Cities could propose a litigation class for their state law claims, which are not the subject of the negotiation class, at this very moment.

Further, the Six Cities overstate the amount of leverage that is created by class members’ threatening litigation in negotiations. It is widely known that the “vast majority” of class actions

objector in the Amchem case from the very beginning all the way through the Supreme Court over 20 plus years ago. I’ve been involved with over 20, 524(g) asbestos trusts . . . .”). 40 Even if it wanted to, the district court could not prevent the Six Cities from litigating their claims. See Dukes, 564 U.S. at 367 (“Because the Rules Enabling Act forbids interpreting Rule 23 to abridge, enlarge or modify any substantive right, a class cannot be certified on the premise that [Defendant] will not be entitled to litigate its statutory defenses to individual claims.”) (internal citations omitted). 41 See Opiate Litig., 332 F.R.D. at 537 (“[The negotiation class] does not stop parties in the MDL from settling in other ways. And there is nothing intrusive about this process: it does not stop any litigation from continuing and in no way interferes with the upcoming bellwether trials in this MDL. This process simply provides an option[.]”). In a subsequent order, the district court clarified that those who chose to remain in the class could independently settle or litigate their claims, so long as they do not participate in double recovery with respect to the defendants with which they have settled or against whom they have obtained a judgment. See Ord. Clarifying Negotiation Class Certification Ord., Nos. 1:17-md-02804-DAP, 1:18-op-45090, R. 2713, at 2–3 (Page ID #419213–14). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 49

certified under Rule 23(b)(3) result in settlement, not a judgment after trial on the merits. Geoffrey P. Miller, Rethinking Certification and Notice in Opt-Out Class Actions, 74 UMKC L. Rev. 637, 640 (2006); see also Rubenstein, 3 Newberg on Class Actions, supra, § 9:52 (5th ed.) (“[M]ost class actions today are settlement class actions in any case[.]”). The threat of litigation is not a genuine source of leverage; rather, it is the possibility of certification that packs power in the punch. See Geoffrey P. Miller, supra, at 642 (“For defendants, certification of a large-scale class action can turn an insignificant case into one with potentially devastating liability exposure.”). Despite holding themselves out as litigating entities,42 the Six Cities—who have not clarified otherwise in their brief—have not referred to or cited any complaint filed in this suit. So to maintain their means to threaten litigation—which they can already pursue either on their own or as a class—the Six Cities ask this court to eliminate Plaintiffs’ primary source of bargaining power and potentially force all parties into a litigation that no one else wants to pursue.43 Because “it is difficult to see” how the “preserv[ation of] such a purely theoretic individual interest meaningfully furthers litigant autonomy,” Ryan C. Williams, Due Process, Class Action Opt Outs, and the Right Not to Sue, 115 Colum. L. Rev. 599, 617–18 (2015), we should resist the urge to capitulate to the Six Cities’ bold request to transform this 33,000-strong class into the tale of Six Cities.

Similarly unconvincing is the Six Cities’ assertion that litigation class members’ ability to opt out after a settlement is proposed guarantees fair treatment for all class members. See Six Cities’ Br. at 25. The Six Cities skim over how the district court “may”—not must—grant class members a second opportunity to opt out after a settlement is tendered; they also conveniently

42 Again, the Six Cities’ brief never clarifies whether they have filed a suit in this case, and there is no accessible record that any of the Six Cities have filed complaints. For the purposes of analysis in this section, I assume that they are litigating entities based on the Six Cities’ counsel’s representations to the district court. See Transcript, No. 1:17-md-02804-DAP, R. 2147, at 15:19–16:15 (Page #ID 288093–94), 20:11–20:16 (Page ID #288098), 21:6–25 (Page ID #288099); see also supra note 33. 43 See Appellees’ Br. at 10 (“The purpose of the Negotiation Class was only to provide a vehicle for settlement discussions with any defendant that sought to take advantage of this structure.”); Opiate Litig., 332 F.R.D. at 537 (“The Court has determined that . . . the [negotiation class] process is more likely to promote global settlement than it is . . . to impede it”); id. at 536–37 (“The Defendants have insisted throughout on the need for a ‘global settlement,’ that is, a settlement structure that resolves most, if not all, lawsuits against them arising out of the opioid epidemic.”); Six Cities’ Br. at 1 (“[A]llowing settlement class members to ‘opt out of the class after the settlement’ is reached . . . was objectionable [to the district court] because it posed the risk that ‘many of the[] Plaintiffs could opt out,’ depriving Defendants of the global settlement they desired.”) (internal quotation marks omitted) (citing Opiate Litig., 332 F.R.D. at 537). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 50

omit that Rule 23(e)(4), the source of this potential second opt out, applies in any class action in which a settlement is proposed. Fed. R. Civ. P. 23(e)(4). Should this negotiation class’s deliberations yield a settlement proposal, Rule 23(e)(4) will still be triggered. Another wrinkle is missing from the Six Cities’ brief—Rule 23(e)(4)’s “[n]ew [o]pportunity to be [e]xcluded[,]” Fed. R. Civ. P. 23(e), did not exist in the Federal Rules of Civil Procedure until 2003, see Fed. R. Civ. P. 23 advisory committee’s note to 2003 amendment; see also Wright, Miller, & Kane, 7B, supra, § 1797.5. Between 1966 and 2003, if a court did not “[keep] the opt-out window open throughout the litigation” or “beyond the time of settlement,” Kenneth S. Rivlin & Jamaica D. Potts, Proposed Rule Changes to Federal Civil Procedure May Introduce New Challenges in Environmental Class Action Litigation, 27 Harv. Envtl. L. Rev. 519, 531 (2003), members of a litigation class had no choice but to stomach whatever settlement came down the chute. Here, the Six Cities’ risk is ameliorated by the existence of Rule 23(e)(4) and by the district court’s expressly reserving the right to provide a second opt-out opportunity. See Opiate Litig., 332 F.R.D. at 540 (“Moreover, the [c]ourt always retains the option of enabling a second opt-out opportunity if circumstances require.”).

Again, the significance of a possible post-negotiation opt-out opportunity should not be exaggerated—in practice, courts are hesitant to invoke their discretion under Rule 23(e)(4) to allow a second round of opt-out rights to class members. See Low v. Trump Univ., LLC, 881 F.3d 1111, 1121 (9th Cir. 2018) (“[W]hile some class action settlements allow a second opt- out opportunity, they are unusual and probably result from the bargaining strength of the class negotiators[] rather than any due process concerns.”) (internal quotation omitted); see also 2 McLaughlin on Class Actions § 6:21 n.8 (16th ed. 2019) (collecting cases that “reject[] the suggestion that a second opt-out should be granted as a matter of course, even if the terms of the settlement change after the expiration of the initial opt-out period[]”); Theodore Eisenberg & Geoffrey Miller, The Role of Opt-Outs and Objectors in Class Action Litigation: Theoretical and Empirical Issues, 57 Vand. L. Rev. 1529, 1544 n.64 (2004) (describing the “unusual situation where, post settlement, the court allows a second round of opt-out rights to class members who previously did not exercise their right to exclude themselves”). If the district court does refuse to grant a second chance to opt out of this negotiation class, such a decision would be par for the course in the class action context. Because the creation of a second opt-out juncture is just as Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 51

unlikely and abounds with as little consequence here as in any class action, negotiation class members do not shoulder any more risk of tying themselves to the mast of an unknown settlement than litigation class members do.

The Six Cities also exaggerate the benefits afforded to class members who can opt out after a settlement has been proposed. Again, the Six Cities neglect to mention that Rule 23(e)(1)(B) requires district courts to conduct a thorough review of a final settlement proposal in any class action, Fed. R. Civ. P. 23(e)(1)(B), and that the district court has already stipulated that it will do so in this case should negotiations produce a proposed settlement, Opiate Litig., 332 F.R.D. at 539 (citing Fed. R. Civ. P. 23(e)(1) for the proposition that “the [c]ourt must preliminarily approve the settlement”). The Six Cities are correct, however, that, unless the district court permits a second opt-opt opportunity under Rule 23(e)(4), members of the negotiation class may not be allowed to opt out after a settlement is proposed. Accordingly, it is also true that members of the negotiation class are unable to opt out if they are “dissatisfied with [the settlement’s] terms.” Six Cities’ Br. at 26. But, once again, the Six Cities paint a picture of the negotiation class’s opt-out timeline that obfuscates the reality of class action proceedings. The Supreme Court has been clear—due process demands only that class members are given one opportunity to opt out of a Rule 23(b)(3) class. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 812 (1985); cf. Dukes, 564 U.S. at 363 (“In the context of a class action predominantly for money damages we have held that absence of notice and opt out violates due process.”) (citing Shutts, 472 U.S. at 812). Notwithstanding the Six Cities’ unconvincing recapitulation of their storm of arguments regarding leverage, the threat of litigation, and the benefits of reviewing a settlement before deciding the opt out, Six Cities’ Br. at 41–44, they have failed to explicate why the inability to opt out after receiving notice of settlement, which has long-defined the litigation class, is suddenly constitutionally or statutorily suspect here.

Moreover, the Six Cities rely on yet another dubious premise that does not cohere with how class actions operate in practice—that the guarantee of opting out after receiving notice of the settlement provides a realistic safeguard for class members. In practice, members of classes, including settlement classes, almost never choose to opt out. See Eisenberg & Miller, supra at 1532, 1557–59 (finding that opt-out rates for all class actions and for settlement class actions are, Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 52

on average, less than 1%).44 Exiting a settlement class is rarely realistic for most parties, who “would then be required to conduct his or her own lawsuit, the costs of which would usually exceed any expected enhancement in consideration that could be achieved by individual litigation.” Geoffrey P. Miller, supra, at 640; see also Shutts, 472 U.S. at 809 (“[T]his lawsuit involves claims averaging about $100 per plaintiff; most of the plaintiffs would have no realistic day in court if a class action were not available.”). Thus, “[t]he option value conferred on absent plaintiffs by late certification, in other words, [is] often worthless.” Geoffrey P. Miller, supra, at 640.

The risks associated with requiring class members to make an early opt out decision are undeniable: the cities and counties may not be allowed to opt out the class if negotiations yield a dissatisfactory proposed settlement. But, again, Appellants’ reducing the entire class action enterprise’s certification procedures into a narrative of “leave or remain” perpetuates a false dichotomy. The Six Cities neglect to couple their cherry-picking of these risks with a candid acknowledgement of how these hazards are uniformly latent in all class actions or how the negotiation class’ attributes mend some of these deficiencies. In the negotiation class, potential class members can help develop the plan for allocating the settlement among the class and for voting on a proposed settlement—an opportunity that they would be denied in a settlement class. See Opiate Litig., 332 F.R.D. at 538. Further, negotiation class members know the settlement distribution mechanism ahead of their opt-out decision, unlike in a litigation class. See id. Because the district court’s certification fixes the class size early, negotiation class members are prepared to calculate their share of any future settlement, which they could not do in either a settlement or litigation class. See id. at 537. After a proposed settlement, negotiation class members have the opportunity to vote on the settlement via a carefully calibrated voting plan that requires supermajorities (here, 75%) of six distinct groups. See id. at 538. This voting structure renders it highly unlikely that class members—especially the most engaged class members, like the Six Cities—would be bound to a woefully dissatisfying settlement. These protections are far more robust than those guaranteed by the ordinary settlement class, where class members are

44 For class actions where the district court approved a settlement, the mean percentage of opt-outs was 0.7% and the median was 0.1%. See Theodore Eisenberg & Geoffrey Miller, The Role of Opt-Outs and Objectors in Class Action Litigation: Theoretical and Empirical Issues, 57 VAND. L. REV. 1529, 1558 (2004). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 53

almost always forced to accept a meager amount without any chance for input at all. The Six Cities would have us believe that the negotiation class is the “worst of both worlds.” Six Cities’ Br. at 3. But such a characterization selectively highlights the pitfalls that are endemic to any kind of class action while surreptitiously concealing the ways that the negotiation class ameliorates those dangers.

It is difficult to comprehend how we can satisfy the Six Cities’ demands. We cannot force the district court to allow the Six Cities a second opportunity to opt out after they receive notice of settlement; class members are not entitled to a second opt-out opportunity under Rule 23(e)(4)’s plain text and the Supreme Court’s precedent in Phillips Petroleum Co. v. Shutts, 472 U.S. 797 (1985). If we abolish the negotiation class and impose the litigation class or settlement class structure onto Plaintiffs, we expose them to the exact same, if not worse, risks that the Six Cities raised in their briefs. In the likely scenario that Plaintiffs pursue a settlement class instead, we will have robbed Plaintiffs of their bargaining power and will obfuscate any transparency about settlement distribution. Because any deficiencies in the negotiation class will be just as, if not more, potent in a litigation or settlement class, we should not capitulate to the Six Cities’ request that we sail Plaintiffs’ claims away from the cliffs and into the whirlpool.

V.

I agree that we should not disregard the motion panel’s decision, see No. 19-0305, R. 24; No. 19-0306, R. 22, that this court may exercise our “sole,” and “unfettered,” Fed. R. Civ. P. 23, advisory committee’s note to 1998 amendment, discretion to “permit an appeal from an order granting or denying class-action certification [under Rule 23],” Fed. R. Civ. P. 23(f). But as a matter of both procedure and logic, we should decline to reach the merits of this interlocutory appeal.

Appellate standing “does not have its source in the jurisdictional limitations of Art[icle] III”45 but is “derived from the statutes granting appellate jurisdiction and the historic practices of

45 Both Defendants and Appellees cite precedent that refers to Article III standing. See, e.g., Defs.’ Br. at 55 (citing Bryant v. Yellen, 447 U.S. 352 (1980)); Appellees’ Br. at 37 (citing Hollingsworth v. Perry, 570 U.S. 693 (2013)). Article III standing is distinct from appellate standing, and the “aggrieved” standard from Deposit Guaranty National Bank v. Roper, 445 U.S. 326 (1980), is proper here. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 54

the appellate courts[,]” Deposit Guar. Nat’l Bank v. Roper, 445 U.S. 326, 333–34 (1980), and “may be a bar to an appeal even though a litigant had standing before the district court[,]” City of Cleveland v. Ohio, 508 F.3d 827, 836 (6th Cir. 2007) (internal citation omitted). “Ordinarily, only a party aggrieved by a judgment or order of a district court may exercise the statutory right to appeal therefrom[,]”46 Roper, 445 U.S. at 333, and “[f]ederal appellate jurisdiction is limited by the appellant’s personal stake in the appeal[,]” id. at 336. When a party is challenging a district court’s class certification decision, the party should “maintain[] throughout [the] appellate litigation that they retain a continuing individual interest in the resolution of the class certification question.” Id.

Let us start with Defendants’ stated aggravations with the district court’s certifying a negotiation class. Defendants assert that they have a “stake in the controversy[,]” Defs.’ Br. at 55 (citing City of Cleveland, 508 F.3d at 836–38), because they “face material harm from the improper certification” and because the certification “diverts resources away from permissible modes of resolving cases in this national MDL[,]” “is replete with conflicts of interest, cannot be relied upon to achieve meaningful results, and is structured so that any settlement that might be reached would be vulnerable to attack.” Defs.’ Br. at 55–56.47 Even if taken as true, these brief

46 In Roper, the Supreme Court credited 28 U.S.C. § 1291 as the statute that permits appellate jurisdiction. Roper, 445 U.S. at 333 (“Congress has vested appellate jurisdiction in the courts of appeals for review of final decisions of the district courts.”) (citing 28 U.S.C. § 1291). Notably, a different statute, 28 U.S.C. § 1292(e), is implicated when appellate courts review district courts’ class certifications under Federal Rule of Civil Procedure 23(f). See Fed. R. Civ. P. 23, advisory committee’s note to 1998 amendment (“Subdivision (f). This permissive interlocutory appeal provision is adopted under the power conferred by 28 U.S.C. § 1292(e).”). The Court’s decision in Roper predates the 1998 promulgation of Rule 23(f). But the Roper Court refused to limit its appellate standing analysis to “that narrow category of circumstances where appeal was allowed prior to final judgment as a matter of right under 28 U.S.C. § 1291[,]” and did not “intend[] to preclude motions under 28 U.S.C. § 1292(b) seeking discretionary interlocutory appeal for review of the [class action] certification ruling[.]” Roper, 445 U.S. at 336 n.8. Given the Court’s prior application of appellate standing analysis to a class certification appeal arising under § 1292(b)—one of the provisions used for interlocutory review prior to the 1998 revisions, see WRIGHT, MILLER & KANE, 7B, supra note 24, § 1802—the Court’s appellate standing precedent is clearly relevant to the present case. 47 Defendants also claim that they have “a concrete interest in reversal of the district court’s certification based on inaccurate and unsupported Rule 23 findings” and express concern about the “potential future application” of “those findings.” Defs.’ Br. at 56. An “asserted [] concern that [] success in some unspecified future litigation would be impaired by stare decisis or collateral-estoppel application of the [d]istrict [c]ourt’s ruling” can “suppl[y] the personal stake in [an] appeal.” Roper, 44 U.S. at 337. But Defendants concede that the district court explicitly prohibited parties from invoking its decision for precedential effect and cite only one motion by a party—who is not a member of the negotiation class—that potentially disobeyed the district court. Defs.’ Br. at 56–57. Because the Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 55

and vague concerns paint the broadest strokes about the class certification’s nature and its hypothetical effects on the multidistrict litigation. Such unsubstantiated generalities fail to articulate Defendants’ “personal stake,” Roper, 445 U.S. at 336 (emphasis added), or “continuing individual interest,” id. (emphasis added), in the district court’s certifying this negotiation class. Cf. id. (recognizing that credit card holders’ “desire to shift part of the costs of litigation to those who will share in its benefits if the class is certified and ultimately prevails” constituted “a continuing individual interest” in the issue of class certification).

Even if we were to look beyond the four corners of their brief, our conferring appellate standing on Defendants here exceeds logic. True, the existence of the negotiation class may “affect[] the state of play” and the Defendants may be “pressured, or at least strongly incentivized, to negotiate with the class.” Maj. Op. at 8. The district court itself indicated that the “[negotiation class] process is more likely to promote global settlement than it is . . . to impede it.” Opiate Litig., 332 F.R.D. at 537. But Defendants cannot be aggrieved by an entity that they have total authority to ignore. See id. at 537–39 (assuring Defendants that “[t]here is nothing coercive about this process: no Defendant has to employ it” and that “[n]o defendant is required to negotiate with the class[.]”). Negotiation with the class is not the “exclusive” way to resolve this suit; parties can choose to settle “their own cases any way they want[.]” Id. at 537. Nor can Defendants be aggrieved by something that they have been seeking from the beginning of this suit—a global settlement. See id. at 536–37 (“The Defendants have insisted throughout on the need for a ‘global settlement,’ that is, a settlement structure that resolves most, if not all, lawsuits against them arising out of the opioid epidemic.”) (emphasis added); see also Six Cities’ Br. at 1; Appellees’ Br. at 10. Defendants cannot now claim that a global settlement hangs over them when it was Defendants themselves who strove to dangle the sword to begin with. Because “[a] party who receives all that he has sought generally is not aggrieved by the judgment affording the relief and cannot appeal from it,” Roper, 445 U.S. at 333, Defendants lack appellate standing here.

certification carries no stare decisis or collateral-estoppel implications, Defendants’ alleged “concrete interest” does constitute a “stake in this controversy.” Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 56

Although the Six Cities fail to address appellate standing altogether in their brief, they might claim a “stake in the controversy” based on their objector status per Rule 23(e)(5). But their bizarre posture in the present case confounds me. Because the Six Cities have not referred to or cited any complaint filed in this suit, we cannot assess the pre-certification claims or leverage that they purportedly sacrificed to stay in the class. They had every opportunity to follow the fraction (1.6%) of the class that timely opted out to pursue settlement or judgment on their own terms. See Appellees’ Br. at 72. Instead, the Six Cities chose to join the other 98.4% of America’s 33,000 cities and counties by remaining in the class. Now that they are one with the class, the Six Cities’ options are endless. They can continue to litigate and bargain alongside the negotiation class’s existence, so long as they do not double-dip into a potential settlement yielded by the class’s negotiations. Or they could do nothing and simply reap the windfall of a transparent negotiation process to which they would not be entitled in a traditional litigation or settlement class structure. Instead, they have distinguished themselves as the only class members to have appealed class certification; curiously, the Six Cities filed their notice of appeal on November 8, 2019, see Case Opening Letter, No. 19-4099, R. 1—fifteen days before the opt-out deadline on November 8, see Opiate Litig., 332 F.R.D. at 555.

Why the Six Cities choose to demolish the negotiation class—an entity that they need not be a member of—instead of exercising their ability to settle or sue either within or outside the class’s confines is beyond comprehension. We ought to be wary of objectors who “may be seeking only personal gain” and “us[e] objections to obtain benefits for themselves rather than assisting in the settlement-review process[,]” Fed. R. Civ. P. 23, advisory committee’s note to 2018 amendment.

We should decline to hear this appeal on the merits in light of these flaws in appellate standing of Defendants and the Six Cities. As the majority apparently sees appellate standing differently, I have focused on explaining why we should uphold the certification of the negotiation class. Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 57

VI.

Finally, this suit has (somehow) waded into the family feud between Plaintiff cities and counties and the states that they sit in. Amici state attorneys general need not fear the cities and counties’ encroaching on state sovereignty in this case.

States may invoke parens patriae and sue to protect “quasi-sovereign interest[s],” such as the “the health and well-being—both physical and economic—of its residents in general.” Alfred L. Snapp & Son, Inc. v. Puerto Rico, 458 U.S. 592, 607 (1982). Amici state attorneys general— hailing from twelve states and the District of Columbia—aver that Plaintiff cities and counties “are, in effect, suing as parens patriae.” Amici States Br. at 3. Plaintiffs do no such thing. These cities and counties seek damages for the direct harm that they allegedly sustained from Defendants’ supposedly unlawful activities; their complaints detail the astronomical sums of money that these localities have spent on costs tied to the opioid crisis, including millions of dollars spent on opioid drugs, health care plans, and workers’ compensation. See, e.g., Summit County Complaint, No. 1:17-md-02804, R. 513 at 226 (estimating that county “spent roughly $66 million on costs tied to the opioid crisis,” projecting “another $89 million [in costs] over the next five years,” alleging “a total cost to the County of $155 million over the ten year period ‘simply trying to keep up with the epidemic’”); Webb County Complaint, No. 1:17-md-02804, R. 108, at 109–10 (Page ID #671–72) (alleging county “has paid hundreds of millions of dollars for opioid drugs, the treatment related to the misuse, addiction and/or overdose of opioids, and the abatement of the societal harms”); City of Chicago Complaint, No. 17-md-2804, R. 512 at 268–69 (alleging that “the City spent more than $13.9 million for over 320,000 claims for opioids” and that city “health plans alone spent more than $2.3 million for the treatment of opioid abuse and dependency”). It is perfectly within the prerogative of counties and cities to try to recoup these enormous expenditures.

Moreover, cities and counties are well-positioned to pursue litigation against opioid manufacturers and distributors.48 Counties and cities are the “unit of government that is closest

48 Lawsuits initiated by cities and counties “are now a prominent feature of the American legal landscape.” Eli Savit, States Empowering Plaintiff Cities, 52 U. MICH. J.L. REFORM 581, 587 (2019). In recent years, American counties and cities have “amplif[ied] their voices on the national stage]” by “increasingly launching lawsuits, an area Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 58

to residents,” Eli Savit, States Empowering Plaintiff Cities, 52 U. Mich. J.L. Reform 581, 584 (2019), and their role on the frontline of the opioid epidemic battle is evident from the vast sums of money that these localities have spent. The depletion of their coffers coupled with their intimate relationship with their communities drives these cities and counties to seek the best outcome for their citizens. Naturally wishing to protect state treasuries, amici attorneys general contend that the negotiation class “damage[s] [] state interests by diverting finite opioid- settlement funds to political subdivisions and away from the States.” Amici States’ Br. at 3. But history has made clear that “it is simply untrue that states generously pass along to cities the resources earned from plaintiff’s-side litigation.” Savit, supra, at 596. In the 1990s, states’ attorneys general reached a global settlement of $246 billion with tobacco companies, but “very little of that money ultimately went to cities” even though cities and counties “incurred significant financial losses as a result of tobacco’s adverse health consequences.” Id. (internal citations omitted). There is already reason to worry that funds will not trickle down from state to local treasuries in the context of the opioid epidemic; “in one of the opioid litigation cases that has already resolved, a significant portion of the money ‘recovered’ by the governmental entities has not been allocated to opioid-related expenses.” See Michelle L. Richards, Pills, Public Nuisance, And Parens Patriae: Questioning The Propriety Of The Posture Of The Opioid Litigation, 54 U. Rich. L. Rev. 405, 408 (2020).49 In sum, Plaintiff cities and counties merely “seek[] reimbursement for monies they have expended—and continue to spend—addressing the opioid crisis[,]” Opiate Litig., 332 F.R.D. at 536; this neither endangers state sovereignty nor undermines public policy.

that was long seen as the province of states” both to “protect their proprietary interests” and to “further the well- being of their residents.” Id. at 582–83. Cities and counties have “br[ought] forwards hundreds of mass-tort style claims[,]” Sarah L. Swan, Plaintiff Cities, 71 VAND. L. REV. 1227, 1227 (2019), and have sued to “protect consumers, enforc[e] environmental protections, and vindicat[e] civil rights.” Savit, supra, at 582–83, in the name of “pursu[ing] the perpetrators of the social harms that have devastated their constituents and their communities.” Swan, supra, at 1227. 49 See also Michelle L. Richards, Pills, Public Nuisance, And Parens Patriae: Questioning The Propriety Of The Posture Of The Opioid Litigation, 54 U. RICH. L. REV. 405, 408 n.14 (2020) (citing Marianne Skolek, West Virginia Uses OxyContin Settlement Money to Build Gym, NATIONAL PAIN REPORT, http://nationalpainreport.com/west-virginia-uses-oxycontin-settlement-money-to-build-a-gym-8814021.html (Apr. 30, 2012)) (citing news article that summarizes how West Virginia used $1.5 million of its $10 million settlement with Purdue Pharma to build a gym). Nos. 19-4097/4099 In re Nat’l Prescription Opiate Litig. Page 59

VII.

We should not focus exclusively on the naked words of Rule 23 at the expense of the Federal Rules of Civil Procedure’s equitable backdrop and in ignorance of the obvious precedent set by the settlement class’s creation and proliferation. All class actions certified under Rule 23(b)(3) for whatever purpose—litigation, settlement, or negotiation—must adhere to Rule 23(a)’s prerequisites and Rule 23(b)(3)’s demands; as long as the district court meticulously follows the dictates of the class action rule, what’s in a name?50 The present negotiation class certification presents no serious policy, constitutional, or state sovereignty entanglements. Accordingly, I respectfully dissent.

50 WILLIAM SHAKESPEARE, ROMEO AND JULIET act 2, sc. 2.