69 F.4th 45 (2d Cir. 2023)

From Casetext: Smarter Legal Research

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

United States Court of Appeals, Second Circuit

May 30, 2023

69 F.4th 45 (2d Cir. 2023)

Summary

approving reorganization plan's inclusion of non-consensual third-party releases for corporation's directors and officers

Summary of this case from The Fin. Oversight & Mgmt. Bd. For P.R. v. Rosa (In re The Fin. Oversight & Mgmt. Bd. for P.R.)

See 2 Summaries

Opinion

NO. 22-110-bk (L)NO. 22-113-bkNO. 22-115-bkNO. 22-116-bkNO. 22-117-bkNO. 22-119-bkNO. 22-121-bkNO. 22-299-bkNO. 22-203-bk

05-30-2023

IN RE: PURDUE PHARMA L.P., Purdue Pharma Inc, Purdue Transdermal Technologies L.P., Purdue Pharma Manufacturing L.P., Purdue Pharmaceuticals L.P., Imbrium Therapeutics L.P., Adlon Therapeutics L.P., Greenfield BioVentures L.P., Seven Seas Hill Corp., Ophir Green Corp., Purdue Pharma of Puerto Rico, Avrio Health L.P., Purdue Pharmaceutical Products L.P., Purdue Neuroscience Company, Nayatt Cove Lifescience Inc., Button Land L.P., Rhodes Associates L.P., Paul Land Inc., Quidnick Land L.P., Rhodes Pharmaceuticals L.P., Rhodes Technologies, UDF LP, SVC Pharma LP, SVC Pharma Inc., Debtors. Purdue Pharma, L. P., Purdue Pharma Inc, Purdue Transdermal Technologies L.P., Purdue Pharma Manufacturing L.P., Purdue Pharmaceuticals L.P., Imbrium Therapeutics L.P., Adlon Therapeutics L.P., Greenfield BioVentures L.P., Seven Seas Hill Corp., Ophir Green Corp., Purdue Pharma of Puerto Rico, Avrio Health L.P., Purdue Pharmaceutical Products L.P., Purdue Neuroscience Company, Nayatt Cove Lifescience Inc., Button Land L.P., Rhodes Associates L.P., Paul Land Inc., Quidnick Land L.P., Rhodes Pharmaceuticals L.P., Rhodes Technologies, UDF LP, SVC Pharma LP, SVC Pharma Inc., Debtors-Appellants-Cross-Appellees, The Official Committee of Unsecured Creditors of Purdue Pharma L.P., et al., Ad Hoc Committee of Governmental and Other Contingent Litigation Claimants, the Raymond Sackler Family, Ad Hoc Group of Individual Victims of Purdue Pharma, L.P., Multi-state Governmental Entities Group, Mortimer-Side Initial Covered Sackler Persons, Appellants-Cross-Appellees, v. The City of Grande Prairie, as Representative Plaintiff for a Class Consisting of All Canadian Municipalities, the Cities of Brantford, Grand Prairie, Lethbridge, and Wetaskiwin, the Peter Ballantyne Cree Nation, on Behalf of All Canadian First Nations and Metis People, the Peter Ballantyne Cree Nation on Behalf Itself, and the Lac La Ronge Indian Band, Appellees-Cross Appellants, The State of Washington, State of Maryland, District of Columbia, U.S. Trustee William K. Harrington, State of Connecticut, Ronald Bass, State of California, People of the State of California, By and Through Attorney General Rob Bonta, State of Oregon, State of Delaware, By and Through Attorney General Jennings, State of Rhode Island, State of Vermont, Ellen Isaacs, on Behalf of Patrick Ryan Wroblewski, Maria Ecke, Andrew Ecke, Richard Ecke, Appellees.

Marshall S. Huebner, (Benjamin S. Kaminetzky, Eli J. Vonnegut, James I. McClammy, Marc J. Tobak, Christopher S. Robertson, Gerard X. McCarthy, and Garrett L. Cardillo, on the brief), Davis Polk & Wardwell LLP, New York, New York; Gregory G. Garre, Charles S. Dameron, Eric J. Konopka, and Joshua J. Craddock, Latham & Watkins LLP, Washington, D.C. (on the brief), for Debtors-Appellants-Cross-Appellees Purdue Pharma L.P., Purdue Pharma Inc., Purdue Transdermal Technologies L.P., Purdue Pharma Manufacturing L.P., Purdue Pharmaceuticals L.P., Imbrium Therapeutics L.P., Adlon Therapeutics L.P., Greenfield BioVentures L.P., Seven Seas Hill Corp., Ophir Green Corp., Purdue Pharma of Puerto Rico, Avrio Health L.P., Purdue Pharmaceutical Products L.P., Purdue Neuroscience Company, Nayatt Cove Lifescience Inc., Button Land L.P., Rhodes Associates L.P., Paul Land Inc., Quidnick Land L.P., Rhodes Pharmaceuticals L.P., Rhodes Technologies, UDF LP, SVC Pharma LP, and SVC Pharma Inc. Mitchell P. Hurley (Erik Y. Preis and Sara L. Brauner, on the brief), Akin Gump Strauss Hauer & Feld LLP, New York, New York; Z.W. Julius Chen, Akin Gump Strauss Hauer & Feld LLP, Washington, D. C. (on the brief), for Appellant-Cross-Appellee The Official Committee of Unsecured Creditors of Purdue Pharma L.P., et al. Roy T. Englert, Jr., (Lawrence S. Robbins, on the brief), Robbins, Russell, Englert, Orseck & Untereiner LLP, Washington, D.C.; Kenneth H. Eckstein, Rachael Ringer, and David E. Blabey Jr., Kramer Levin Naftalis & Frankel LLP, New York, New York (on the brief); David J. Molton, Brown Rudnick LLP, New York, New York (on the brief); Richard J. Leveridge, Richard D. Shore, and Daniel I. Wolf, Gilbert LLP, Washington, D.C. (on the brief); Melanie L. Cyganowski, Otterbourg P.C., New York, New York (on the brief), for Appellant-Cross-Appellee Ad Hoc Committee of Governmental & Other Contingent Litigation Claimants. J. Christopher Shore (Michele J. Meises and Alice Tsier, on the brief), White & Case LLP, New York, New York; Edward E. Neiger, ASK LLP, New York, New York (on the brief), for Appellant-Cross-Appellee Ad Hoc Group of Individual Victims of Purdue Pharma, L.P. Jeffrey A. Liesemer (Kevin C. Maclay, Todd E. Phillips, and Lucas H. Self, on the brief), Caplin & Drysdale, Chartered, Washington, D.C., for Appellant-Cross-Appellee Multi-State Governmental Entities Group. Maura Kathleen Monaghan (Jasmine Ball, on the brief), Debevoise & Plimpton LLP, New York, New York, for Appellant-Cross-Appellee Mortimer-Side Initial Covered Sackler Persons. Gregory P. Joseph, Mara Leventhal, Joseph Hage Aaronson LLC, New York, New York; Gerard Uzzi, Alexander B. Lees, Milbank LLP, New York, New York, for Appellant-Cross-Appellee The Raymond Sackler Family. J. Carl Cecere, Cecere PC, Dallas, Texas; Allen J. Underwood, II, Lite Depalma Greenberg & Afanador, LLC, Newark, New Jersey (on the brief), for Appellees-Cross-Appellants The City of Grand Prairie, as Representative Plaintiff for a Class Consisting of All Canadian Municipalities, The Cities of Brantford, Grand Prairie, Lethbridge, and Wetaskiwin, The Peter Ballantyne Cree Nation, on behalf of All Canadian First Nations and Metis People, the Peter Ballantyne Cree Nation, on behalf of itself, and the Lac La Ronge Indian Band. Michael Shih, Attorney, Appellate Staff Civil Division, (Michael S. Raab and Sean R. Janda, Attorneys, Appellate Staff Civil Division, on the brief), for Sarah E. Harrington, Deputy Assistant Attorney General, (Beth A. Levene and Sumi K. Sakata, Trial Attorneys, for P. Matthew Sutko, Associate General Counsel, for Ramona D. Elliot, Deputy Director/General Counsel, Executive Office for United States Trustees, on the brief), United States Department of Justice, Washington, D.C.; Lawrence H. Fogelman, Peter Aronoff, Danielle J. Levine, Assistant United States Attorneys, for Damian Williams, United States Attorney for the Southern District of New York, New York, New York (on the brief), for Appellee U.S. Trustee William K. Harrington. Ronald Bass, pro se, North Plainfield, New Jersey. Ellen Isaacs, pro se, on behalf of Patrick Ryan Wroblewski, and all similarly situated persons, Floyd, Virginia. Maria Ecke, Andrew Ecke, Richard Ecke, pro se, West Simsbury, Connecticut. Gregg M. Galardi, Ropes & Gray LLP, New York, New York; Douglas Hallward-Driemeier, Ropes & Gray LLP, Washington, D.C.; Andrew G. Devore, Ropes & Gray LLP, Boston, Massachusetts; Ryan Preston Dahl, Ropes & Gray LLP, Chicago, Illinois, for Amici Curiae Law Professors in support of Appellants. Paul H. Zumbro, George E. Zobitz, Lauren A. Moskowitz, Cravath, Swaine & Moore LLP, New York, New York, for Amicus Curiae The Association of the Bar of the City of New York. Tara S. Morrissey, U.S. Chamber Litigation Center, Washington, D.C., for the Chamber of Commerce of the United States of America; David H. Thompson, Brian W. Barnes, John W. Tienken, Cooper & Kirk, PLLC, Washington, D.C., for Amici Curiae the Chamber of Commerce of the United States of America, the American Tort Reform Association, and Product Liability Advisory Council, Inc. Andrew K. Glenn, Jed I. Bergman, Shai Schmidt, Glenn Agre Bergman & Fuentes LLP, New York, New York, for Amici Curiae Bankruptcy Law Professors Ralph Brubaker, George W. Kuney, and Bruce A. Markell. Joshua L. Seifert, Joshua. L. Seifert PLLC, New York, New York, for Amici Curiae Law Professors in Support of Appellees Regarding the "Abuse" Standard. Robert J. Keach, Bernstein Shur Sawyer & Nelson, P.A., Portland, Maine; Albert Togut, Togut, Segal & Segal LLP, New York, New York, for Amici Curiae Commissioners of the American Bankruptcy Institute's Commission to Study the Reform of Chapter 11. Daniel R. Walfish, Walfish & Fissell PLLC, for Amicus Curiae Professor Adam J. Levitin. Harold D. Israel, Levenfeld Pearlstein, LLC, Chicago, Illinois; Scott R. Bickford, Martzell, Bickford & Centola, New Orleans, Louisiana, for Amicus Curiae the Ad Hoc Committee of NAS Children.

Eunice C. Lee, Circuit Judge

Bankruptcy is inherently a creature of competing interests, compromises, and less-than-perfect outcomes. Because of these defining characteristics, total satisfaction of all that is owed—whether in money or in justice—rarely occurs. When a bankruptcy is the result of mass tort litigation against the debtor, the complexities are magnified because the debts owed are wide-ranging and the harm caused goes beyond the financial. That is the circumstance here.

The Debtor, Purdue Pharma L.P. ("Purdue"), was owned and operated by the Sackler family for decades. In the 1990s, Purdue introduced to market—and promoted as non-addictive—OxyContin, a controlled-release semisynthetic opioid analgesic. In the years following, OxyContin has been blamed for significantly contributing to one of the largest public health crises in this nation's history: the opioid epidemic.

The district court explained that the "Sackler Family," as used in the court's opinion, "means the Mortimer D. Sackler Family (also known as 'Side A' of the Sackler family) and the Raymond R. Sackler Family (also known as 'Side B' of the Sackler family.)." In re Purdue Pharma, L.P., 635 B.R. 26, 35 n.2 (S.D.N.Y. 2021). The Mortimer D. Sackler family explains that the "Mortimer-side Initial Covered Sackler Persons include Theresa Sackler, Ilene Sackler Lefcourt, Kathe Sackler, and Mortimer D.A. Sackler, as well as trusts of which they are beneficiaries and the trustees of those trusts, and Beacon Company." Br. for Appellant-Cross-Appellee Mortimer-Side Initial Covered Sackler Persons at 1 n.1. The Raymond R. Sackler family explains that the "Raymond Sackler family is comprised of natural persons who are descendants of Raymond R. Sackler and current and former spouses of their descendants." Br. for Appellant-Cross-Appellee the Raymond Sackler Family at 1 n.1.

The fallout from this crisis led to a veritable deluge of litigation against both Purdue and individual members of the Sackler family. Claimants, spread across the United States and Canada, included many sufferers of opioid addiction and the families of those lost to opioid overdoses. To settle the mass of civil claims, the parties, including Purdue and the Sacklers, agreed that in exchange for Purdue filing for bankruptcy, the Sacklers would personally contribute billions of dollars to the bankruptcy if all civil claims against them were released.

The district court explained that the Sacklers' contribution would go "to a fund that would be used to resolve both public and private civil claims as well as both civil and criminal settlements with the federal government." In re Purdue Pharma, L.P., 635 B.R. at 70.

In accordance with that plan, Purdue and its related entities (together, the "Debtors" or "Purdue") filed for bankruptcy; the Sacklers did not. Following an intensive months-long and multi-phase mediation involving various interested parties and potential creditors of Purdue, the bankruptcy court approved the proposed bankruptcy plan. In doing so, the court limited the release of claims against the Sacklers to only claims that directly affected the Debtors' estate and for which Purdue's conduct was a legal cause, or a legally relevant factor, of any released cause of action against the Sacklers. In exchange, the Sacklers agreed to contribute a total $5.5-6.0 billion to the bankruptcy. On subsequent appeal, however, the district court for the Southern District of New York reversed the bankruptcy court and vacated the court's confirmation order, ruling that the Bankruptcy Code did not permit such releases.

This appeal followed. During the pendency of the appeal, the various parties to the mediation engaged in further negotiations, resulting in additional changes to the proposed bankruptcy plan. These changes resulted in several more parties dropping their opposition and supporting the further-revised bankruptcy plan. The Appellants, who are challenging the district court's rejection of the proposed plan, include the Debtors, various creditor and claimant groups, and certain Sackler family members. The Appellees are those parties opposed to the proposed plan, although, as noted, their number has dropped since the initial filing of this appeal. These remaining Appellees consist of the U.S. Trustee, several Canadian municipalities and indigenous nations, and several individual pro se plaintiffs.

Aside from their legal arguments, the parties contend that various policy considerations should inform whether a bankruptcy plan containing nonconsensual third-party releases of direct claims may be approved. They also raise questions about fairness and accountability, particularly as it relates to the Sacklers, in releasing parties from liability for actions that cause great societal harm. They debate the very nature of bankruptcy, including the role it is intended to serve and the parties it is intended to benefit.

But, our role in this appeal does not require us to answer all of these serious and difficult questions. Instead, we are tasked only with resolving two key questions: First, does the Bankruptcy Code permit nonconsensual third-party releases of direct claims against non-debtors, and, Second, if so, were such releases proper here in light of all equitable considerations and the facts of this case. We answer both in the affirmative.

We conclude that two sections of the Bankruptcy Code, 11 U.S.C. §§ 105(a), 1123(b)(6), jointly provide the statutory basis for the bankruptcy court's authority to approve a plan that includes nonconsensual releases of third-party claims against non-debtors. In addition, this Court has recognized that in specific circumstances—such as those presented by this appeal—bankruptcy courts are permitted to approve of restructuring plans that include such releases. We accordingly hold that the bankruptcy court's approval of the releases here is permissible both statutorily and under this Court's case law. We further hold that the bankruptcy court's inclusion of the releases is equitable and appropriate under the specific factual circumstances of this case, and we articulate several factors to guide the analysis as to when to allow similar releases in reorganization plans.

Accordingly, we REVERSE the district court's order holding that the Bankruptcy Code does not permit nonconsensual releases of third-party direct claims against non-debtors, AFFIRM the bankruptcy court's approval of the reorganization plan, and REMAND the case to the district court for such further proceedings as may be required, consistent with this opinion. We also AFFIRM the district court's denial of the Canadian Creditors' cross-appeal.

BACKGROUND

I. Factual Background

The following discussion is limited to those underlying facts that are necessary for a determination of the issues on appeal.

The decisions of the bankruptcy and district courts provide a more detailed recitation of the background facts. See In re Purdue Pharma L.P., 633 B.R. 53 (Bankr. S.D.N.Y. 2021) ("Purdue I"); In re Purdue Pharma, L.P., 635 B. R. 26 (S.D.N.Y. 2021) ("Purdue II").

A. Purdue and OxyContin

The Sackler brothers, including Mortimer and Raymond Sackler, purchased Purdue, a privately held pharmaceutical company, in the 1950s. Members of the Sackler family held various director and officer positions throughout the company and, from approximately 1993 to 2018, Purdue's Board of Directors contained at least six members of the Sackler family. Beyond the board, Sackler family members held other positions of influence in the company. For example, Mortimer and Raymond Sackler served as co-chief executive officers until their deaths, Richard Sackler served as a president, and Mortimer D.A., Ilene, and Kathe Sackler all served as officers.

Arthur Sackler sold any interest in Purdue before the development of OxyContin. He and his heirs are therefore not involved in this action.

In 1995, Purdue developed, and the Food and Drug Administration ("FDA") approved, OxyContin, a controlled-release semisynthetic opioid analgesic. At that time, and for years following, Purdue advertised that the time-release formulation prevented OxyContin from posing a threat of abuse or addiction. OxyContin's FDA label reflected a purportedly low risk of addiction. From 1996 to 2001, Purdue aggressively marketed OxyContin to patients and doctors while downplaying growing addiction concerns. Over this time-period, both prescribed and illegal use of OxyContin increased across the country.

Starting in 2000, state governments began to alert Purdue to the widespread abuse of OxyContin, and, in 2001, the FDA required Purdue to remove from its label that OxyContin had a low risk of addiction. In the years that followed, lawsuits against Purdue—brought by, among others, individuals, state governments, and federal agencies—proliferated across the United States.

B. The 2004 Indemnity Agreement

At the end of 2004, Purdue's Board of Directors voted to indemnify, among others, Purdue's directors and officers against claims made in connection with their service to the company. Bryant C. Dunaway v. Purdue Pharma L.P. (In re Purdue Pharma L.P.), No. 19-cv-10941-CM (S.D.N.Y. June 22, 2020), ECF No. 24-2 (Appellees' Suppl. App'x at SA 627-36) (the "Sackler-Purdue Indemnity Agreement" or the "Indemnity Agreement"). As part of its obligations under the agreement, Purdue agreed to:

indemnify and hold harmless each Indemnitee from and against any and all expenses (including attorneys' fees), amounts paid or incurred in satisfaction of or as part of settlements, judgments, fines, penalties, liabilities and similar or related items incurred or suffered or threatened to be incurred or suffered as a result of or in connection with such Indemnitee being made or threatened to be made a party to or participant in any pending, threatened or completed actions, suits or proceedings, whether civil, criminal, administrative, arbitrative or investigative . . . .

Id. at 628-29. Purdue further agreed to "advance all costs and expenses (including attorneys' fees and expenses) incurred by the Indemnitee in defending any one or more Proceedings." Id. at 630.

The protections conferred by the Indemnity Agreement were expansive and had no immediate time limit. The agreement ensured that "[t]he Indemnitee's rights under these provisions shall continue after the Indemnitee has ceased to serve" in his or her official capacity at Purdue, and "shall be binding on and inure to the benefit of successors, assigns, legatees, distributees, heirs, executors, guardians, administrators, estates and other legal representatives." Id. at 635.

At the same time, the Indemnity Agreement contained a bad faith carveout. Purdue's indemnification obligations did not extend to matters where "a final decision by a court . . . establishe[d] that the Indemnitee did not act in good faith." Id. at 629.

C. Sackler Conduct Between 2007 and 2019

Starting in 2007, the Sacklers anticipated that the effects of litigation against Purdue would eventually impact them directly. See, e.g., Deferred Joint App'x at 5059 (David Sackler emailed Jonathan and Richard Sackler, "We will be sued . . . . [A]sk yourself how long it will take these lawyers to figure out that we might settle with them if they can freeze our assets and threaten us."). From 2008 to 2016, Purdue distributed a significant proportion of the company's revenue—an approximated $11 billion in total—to Sackler family trusts and holding companies. This represented an increase in the distribution pattern from years prior and "drained Purdue's total assets by 75% and Purdue's 'solvency cushion' by 82%" during that same time period. Special App'x at 40. By 2018, Purdue defended the many lawsuits against it from a significantly weakened financial position, and, by 2019, all Sacklers had stepped down from Purdue's Board of Directors.

D. The DOJ Suit

In 2019, the United States Attorneys' Offices for the Districts of New Jersey and Vermont, and the United States Department of Justice ("DOJ") brought federal criminal and civil charges against Purdue. The criminal counts alleged that Purdue defrauded the government by inducing healthcare providers to prescribe OxyContin and violated the federal anti-kickback statute. The DOJ also brought civil claims under various federal statutes and common law doctrines (such as mistake, unjust enrichment, fraud, nuisance, and negligent entrustment).

In 2020, after filing for bankruptcy, Purdue entered into a plea agreement with the DOJ, the terms of which created future obligations on Purdue. First, in exchange for Purdue pleading guilty to violations of the federal anti-kickback statute, the DOJ agreed it would "not initiate any further criminal charges against Purdue." Deferred Joint App'x at 4798. Second, regarding its civil liability, Purdue agreed to a forfeiture judgment of $2 billion; the judgment gave the DOJ "superpriority" to collect on the forfeiture judgment in the event of a liquidation of Purdue's estate. Deferred Joint App'x at 4804. Thus, in any future bankruptcy proceedings, the plea required that Purdue satisfy the DOJ's $2 billion claim ahead of all other creditors' claims.

However, the plea agreement also stipulated that the DOJ would agree to release $1.775 billion of its $2 billion claim so long as a future distribution plan met certain requirements, specifically that an abatement trust for the public benefit would be established and a document repository created. Finally, while the plea agreement released Purdue from any additional civil or administrative monetary claims by the government for the covered conduct, it expressly did not release criminal liability.

E. Purdue Files for Bankruptcy

On September 15, 2019, Purdue and its related entities declared bankruptcy; the Sacklers did not. The Estate of the Debtors (the "Estate" or the "res") is estimated at approximately $1.8 billion.

Purdue consists of Purdue Pharma L.P., Purdue Pharma Inc., Purdue Transdermal Technologies L.P., Purdue Pharma Manufacturing L.P., Purdue Pharmaceuticals L.P., Imbrium Therapeutics L.P., Adlon Therapeutics L.P., Greenfield BioVentures L.P., Seven Seas Hill Corp., Ophir Green Corp., Purdue Pharma of Puerto Rico, Avrio Health L.P., Purdue Pharmaceutical Products L.P., Purdue Neuroscience Company, Nayatt Cove Lifescience Inc., Button Land L.P., Rhodes Associates L.P., Paul Land Inc., Quidnick Land L.P., Rhodes Pharmaceuticals L.P., Rhodes Technologies, UDF L.P., SVC Pharma L.P., and SVC Pharma Inc.

Three days after the bankruptcy filing, the Debtors sought an injunction halting all other lawsuits (almost 3,000 actions against Purdue and over 400 actions against the Sacklers concerning liability for OxyContin). On October 11, 2019, the bankruptcy court enjoined all litigation. At the time, claims against the Debtors and Sacklers were estimated at more than $40 trillion.

II. Procedural History

A. The Mediation and Confirmation Process

Following discovery, as is typical in Chapter 11 bankruptcy, the bankruptcy court ordered mediation to reach a plan of reorganization and avoid liquidation of the Estate. In addition to Purdue and the Sacklers, there were a number of groups that participated in the mediation.

The Debtors, Official Committee of Unsecured Creditors of Purdue Pharma L.P., et al. (the "UCC"), Ad Hoc Committee of Governmental and Other Contingent Litigation Claimants ("AHC"), Ad Hoc Group of Non-Consenting States ("NCSG"), Multi-State Governmental Entities Group ("MSGE"), Ad Hoc Group of Individual Victims of Purdue Pharma, L.P. ("PI Ad Hoc Group"), Ad Hoc Committee of NAS Children ("NAS Children"), Ad Hoc Group of Hospitals ("Hospitals"), Third-Party Payor Group ("TPP Group"), and Ratepayer Mediation Participants ("Ratepayers") all participated in the mediation as official Mediation Parties. The Native American Tribes Group ("Tribes Group"), Public School District Claimants ("Public Schools"), the National Association for the Advancement of Colored People, and others also participated in mediation, although not as official Mediation Parties.

The first phase of the mediation addressed the allocation of the Estate's available funds to non-federal public claimants, such as states and political subdivisions, and private claimants. The second phase largely focused on determining what the Sacklers would contribute to the Debtors' estate. While this second phase resulted in an agreement in principle among the Sacklers, the Debtors, and several creditors, a group of twenty-five non-consenting states, among others, rejected the agreement.

That agreement guaranteed that the Sacklers would contribute at least $4.275 billion to the Debtors' estate over approximately nine years. In exchange, the Debtors' plan of reorganization contained several nonconsensual releases (the "Shareholder Release," the "Release," or the "Releases") that, in effect, permanently enjoined certain third-party claims against the Sacklers. As initially proposed, the Release provisions were extremely broad and included the release of claims pertaining to, inter alia, the same subject matter as any claim treated in the plan; any business or other contractual arrangements including transfers; any employment-related conduct; any pending opioid actions and opioid-related activities; and the bankruptcy process.

In the third phase of the mediation, the Sacklers reached a modified agreement with fifteen out of the twenty-five non-consenting states. The new terms of the modified settlement included additional payments of $50 million by the Sacklers, and the accelerated payment of an additional $50 million from a previously agreed-upon settlement payment. These modifications raised the Sacklers' aggregate contribution to the proposed plan to $4.325 billion. At that time, no changes were made to the Shareholder Release.

The majority of the non-consenting states (California, Connecticut, Delaware, Maryland, Oregon, Rhode Island, Vermont, Washington, and the District of Columbia) (the "Nine") maintained their objections to the plan and were parties to the appeal to the district court.

Following mediation, a vote on the proposed plan was set in motion. Notice of the confirmation hearing was published in the summer of 2021, with votes for or against confirmation due by mid-July 2021, and reached 98% of adults in the United States and 86% of adults in Canada. More than 120,000 votes were cast, and each voting class voted "overwhelmingly" in favor of the plan. Special App'x at 150-51 ("In the aggregate, the vote was over 95 percent in favor of confirmation . . . . In each class the percent voting in favor of the plan was above 93 percent with the exception of the class of hospital claims, which was over 88 percent . . . .").

Ultimately, on September 1, 2021—after a confirmation hearing that included the live testimony of 41 witnesses and extensive oral argument—the bankruptcy court rendered an oral ruling stating that it would confirm the proposed plan, but with a few changes. Most relevantly, the court modified the Shareholder Release to ensure that the Debtors' conduct must be a legal cause or a legally relevant factor of any released cause of action against the Sacklers:

I . . . require that the shareholder releases . . . be further qualified than they now are. To apply [only] where . . . a debtor's conduct or the claims asserted against it [are] a legal cause or a legally relevant factor to the cause of action against the shareholder released party . . . .

Deferred Joint App'x at 1330-31. The new Shareholder Release thus read in pertinent part:

[T]he Shareholder Released Parties . . . shall be conclusively, absolutely, unconditionally, irrevocably, fully, finally, forever and permanently released . . . from any and all Causes of Action, including any derivative claims [and future claims] . . . (x) based on or relating to, or in any manner arising from, in whole or in part, (i) the Debtors, . . . (ii) the Estates or (iii) the Chapter 11 Cases and (y) as to which any conduct, omission or liability of any Debtor or any Estate is the legal cause or is otherwise a legally relevant factor.

Special App'x at 920.

B. Bankruptcy Court Order Confirming the Plan

This opinion describes the bankruptcy court's opinion and the subsequent district court opinion only to the extent required to explain our reasoning today.

The bankruptcy court confirmed its modified version of the proposed plan ("the Plan") on September 17, 2021, and issued an extensive opinion memorializing its decision. See In re Purdue Pharma L.P. ("Purdue I"), 633 B.R. 53 (Bankr. S.D.N.Y. 2021) (Robert D. Drain, Bankr. J.).

The bankruptcy court order began by describing its task as "resolv[ing] the collective problem presented by an insolvent debtor and a large body of creditors competing for its insufficient assets . . . especially when there are mass claims premised on . . . massive harm." Purdue I, 633 B.R. at 58. The court found that the confirmation hearing established that the Plan was the only "reasonably conceivable" way to resolve the issues in the case, id. at 59, and, in doing so, grounded its opinion on the principle that, in bankruptcy, courts "focus the solution away from individual litigations to a fair collective result subject to the unique ability under bankruptcy law to bind holdouts under well-defined circumstances who could not otherwise be bound under non-bankruptcy law." Id. at 58.

1. Equitable Considerations

From there, the bankruptcy court asked whether the terms of the Plan created an equitable plan and answered in the affirmative. Purdue I, 633 B.R. at 84-95. The court explained that to approve a settlement, a bankruptcy court must determine whether the proposed terms are fair, equitable, and in the estate's best interest. Id. at 84. Here, in exchange for the Shareholder Release, the terms included:

$4.325 billion, coupled with the Sackler[s'] other agreements, including the dedication of the two charities worth at least $175 million for abatement purposes, the Sacklers' agreement to a resolution on naming rights, their agreement not to engage in any business with NewCo [Purdue's successor company], their agreement to exit their foreign companies within a prescribed time, their agreement to various 'snap back' protections to ensure the collectability of their settlement payments, and their agreement to an unprecedented extensive document depository accessible to the public that will archive in a comprehensive way the Debtors' history, including as it relates to the development, production, and sale of opioids.

Id. The bankruptcy court also highlighted the extensive mediation and discovery processes that led to the development of these terms. Id. at 85-87.

As a legal framework for balancing the equities and determining whether to approve the plan, the court was guided by the factors from In re Iridium Operating LLC, 478 F.3d 452, 464-66 (2d Cir. 2007):

(1) The probability of success, should the issues be litigated, versus the present and future benefits of the settlement; (2) the likelihood of complex and protracted litigation if the settlement is not approved, with its attendant expense, inconvenience and delay, including the difficulty of collecting on a judgment; (3) the interests of the creditors, including the degree to which creditors support the proposed settlement; (4) whether other interested parties support the settlement; (5) the competence and experience of counsel supporting, and the experience and knowledge of the court in reviewing, the settlement; (6) the nature and breadth of the releases to be obtained by officers and directors or other insiders; and (7) the extent to which the settlement is the product of arms-length bargaining.

Purdue I, 633 B.R. at 85.

In applying the Iridium factors, the bankruptcy court observed that, in this case, counsel on both sides were experienced and formidable. Id. at 86-87. Over 95% of the voters approved the Plan, showing clear creditor support, and the potential difficulty in collecting from the Sacklers and their related entities on any successfully litigated claims was an issue of "significant concern." Id. at 89. The court noted that while the Sacklers are worth approximately $11 billion, they are a large family whose assets are "widely scattered and primarily held" in spendthrift trusts—both offshore and in the United States—that are largely unreachable via bankruptcy proceedings.Id. at 88. Moreover, certain members of the Sackler family live "outside of the territorial jurisdiction of the United States and might not have subjected themselves sufficiently to the U.S." such that a U.S. court would have personal jurisdiction over them. Id. And, perhaps most importantly, according to the court, continued litigation—even if it were limited to the claims at issue—would be extremely expensive and lead to delays. Id. at 89-90. Thus, the court reasoned, an order against confirmation would not only destroy the entire settlement but would also result in a major escalation of costs and time. Id.

Spendthrift trusts in the United States may be recovered from, however, if the transfers to such trusts are fraudulent. Id. at 88-89.

The bankruptcy court also noted that, in exchange for the Shareholder Release, the Sacklers were contributing "the largest amount that shareholders have ever paid in such a context of these types of third party claims and closely related claims" and that "the non-monetary consideration under the settlement also is substantial." Id. at 107. And, according to the bankruptcy court's findings, without approval of the Plan including the Release, Purdue would be forced into liquidation, the DOJ would recover its $2 billion claim first, and recovery by all other creditors would be extremely limited because it would not be supplemented with Sackler funds. Id. at 108-09; see also id. at 84 ("Without the $4.325 billion being paid by the Sacklers under the plan and the other elements of the Sackler settlements, those other elements of the plan would not happen. The record is clear on that."). Thus, the court concluded that, in a world without the Plan, the Sacklers would likely be mired in litigation, but it would also be likely that they could successfully shield much of their estimated $11 billion fortune from creditors through spendthrift trusts and offshore accounts, and broader creditor recovery from Purdue's estate would be extremely limited due to the DOJ's superpriority. Id. at 84, 109.

2. The Authority of the Bankruptcy Court to Release Third-Party Claims Against the Sacklers

The bankruptcy court next turned to the Plan's release of third-party claims against the Sacklers, which included all claims that had by then been asserted in litigations against the Sacklers by third parties. The Release encompassed both those based on a direct injury to the third-party claimant and those where the claim properly lay with the Debtors (including, for example, whether the Sacklers fraudulently transferred Purdue funds to family spendthrift trusts and other offshore accounts). Purdue I, 633 B.R. at 91-95. This overview of claims led the court to the thornier legal issue: whether direct claims by a third-party against a non-debtor (here, the Sacklers) could ever be released through the bankruptcy process. Id. at 95.

In addressing the question of its own authority, the bankruptcy court first evaluated threshold arguments and determined that it had subject-matter jurisdiction over the released claims. Id. at 95-98. But, in so finding, the court narrowed the Release even further to cover only those claims that directly affect the res—these claims included "insurance rights" and "the shareholder released parties' rights to indemnification and contribution" from the Debtors. Id. at 97. Likewise, the court noted that "the Debtors' ability to pursue the estates' own closely related, indeed fundamentally overlapping, claims" against the Sacklers also directly affected the res. Id. at 97-98. The court did not exclude derivative claims from the Release, reasoning that those claims were similarly likely to affect the res. Id. at 98.

The bankruptcy court next addressed the objectors' due process arguments and found that they reduced to two claims—neither of which it found meritorious. Id. at 98-99. The first due process claim argued that the Release was an impermissible "adjudication of the claim." Id. at 98. The court disagreed, and instead characterized a release as "part of the settlement of the claim that channels the settlement funds to the estate." Id. As such, the bankruptcy court held that the Release did not rule on the underlying merits of the claims being released. Id. The objectors' second due process argument claimed that there was inadequate notice. Id. at 98-99. The court found adequate notice because the holders of claims against the Debtors had received notice of "the plan's intention to provide a broad release of third-party claims against the shareholders" and other "entities related to the Debtors." Id. at 98. As the final part of its due process analysis, the bankruptcy court also found that the claims released by the Plan were constitutionally core claims, so the bankruptcy court had the constitutional power to issue "a final order under Article III of the Constitution." Purdue I, 633 B.R. at 99-100.

After clearing the constitutional hurdles, the bankruptcy court began its analysis of statutory authority by noting that the majority of Circuits permit nonconsensual third-party releases, while only three Circuits—the Fifth, Ninth, and Tenth—do not. Id. at 100-01. The bankruptcy court concluded that the provision of the Bankruptcy Code relied upon by that minority, 11 U.S.C. § 524(e), is not a statutory impediment to third-party releases. Id. at 101-02.

The bankruptcy court instead looked to 11 U.S.C. § 105(a) and § 1123(b)(6) as two potential sources of a bankruptcy court's equitable authority to approve the releases. Id. at 102-05. Following a review of pertinent case law, the bankruptcy court held that so long as the releases are limited to those claims legally intertwined with the Debtors' conduct, they are appropriately subject to settlement under both statutory and common law frameworks. Id. at 103-05.

The bankruptcy court then looked to this Court's decision in In re Metromedia Fiber Network, Inc., 416 F.3d 136 (2d Cir. 2005), and other case law from this Circuit, to determine which factors a bankruptcy court should consider when determining whether third-party releases are appropriate. Id. at 105-06. The court identified the following factors: (1) the third-party releases were narrowly tailored; (2) monetary contributions were critical to the Plan; (3) the success of the Plan hinged on the third-party releases; (4) the affected class or classes overwhelmingly accepted the Plan; (5) the amount being paid under the Plan was substantial (which, the court noted, is not determined by the Sacklers' net worth because defendants' wealth should not dictate settlement terms); and (6) claimants would be compensated fairly under the Plan. Id. at 106-09.

Evaluating those factors, the bankruptcy court found that they supported approval of the Plan. It pointed to the significant overlap in third-party claims against both the Debtors and the Sacklers, chiefly that: (1) claims against both derived from the Debtors' conduct, and (2) to the extent that one or more of the Sacklers could be said to have directed that conduct, or to have possessed the knowledge and power to do so, the Sacklers' and Debtors' defenses would be the same. Id. at 108. And it added that the potential difficulty, as discussed above, of collecting on any judgment, the existence of spendthrift trusts, and the Estate's limited resources that the litigation process would likely deplete, also weighed in favor of approval of the Plan. Id. at 108-09.

In sum, the bankruptcy court predicated confirmation of the Plan on a few limitations to the third-party releases (namely that the Debtors' conduct amount to a legally relevant factor to a released cause of action and that the settled claims affect the res), but otherwise—having established its authority to do so—confirmed the Plan. Id. at 115.

3. Canadian Creditors' Objections

The bankruptcy court also rejected the objections of certain Canadian municipalities and First Nations (the "Canadian Creditors") to the Plan, which were essentially based on an argument that the Plan improperly classified their claims. Id. at 69-72. Specifically, they objected on the basis that those claims should have been classified like the claims of American non-federal governmental creditors and tribal entities, such that they could participate in abatement trusts. Id. at 69. Yet, the bankruptcy court observed that, even if the Canadian claims had been otherwise classified, notwithstanding the resulting change in the Canadian Creditors' voting status, the Plan still would have been approved. Id. The bankruptcy court's reasons for classifying the Canadian claims separately boiled down to: (1) different regulatory regimes of the United States and Canada, and (2) that the Canadian Creditors did not participate in the mediation process. Id. at 70. The bankruptcy court also noted that certain decisions to recognize or confirm the Plan would be left to the Canadian courts. Id. at 71.

4. Pro SeObjections

Several pro se parties also objected to the Plan, but the bankruptcy court similarly found their objections to be without merit. For example, one pro se objector asserted that it was improper and unfair that the Plan provided only $700-$750 million to a particular claimant group's personal injury claims. Id. at 78. The bankruptcy court looked to the length of the mediation, rigor of the legal analysis and negotiation, and quality of mediators and lawyers, all to support that the valuation of personal injury claims was reasonable. Id. at 78-79. Another pro se objector claimed that releasing the Sacklers from civil liability under the Plan was unfair and should not be approved because this plan is "the Sacklers' plan." Id. at 82. However, the bankruptcy court disagreed and emphasized that the Plan was "not the Sacklers' plan" because it involved an arms-length negotiation among all interested parties with three experienced mediators. Id. at 82-83 (emphasis in original).

C. District Court Order Rejecting the Plan

In a December 16, 2021 opinion, the district court vacated the bankruptcy court's decision to confirm the Plan. In re Purdue Pharma, L.P. ("Purdue II"), 635 B.R. 26 (S.D.N.Y. 2021) (Colleen McMahon, J.). Principally, the court ruled that no statutory authority permits third-party releases such as the ones found in the Plan. Id. at 89-90. The court based its reasoning on two propositions: first, that the Bankruptcy Code does not expressly allow such releases; and second, that this Circuit's case law "has not yet been required to identify any source [in the Bankruptcy Code] for [the] authority" to grant such releases. Id.

1. Subject-Matter Jurisdiction

The district court's analysis of statutory authority was preceded by the preliminary question of the bankruptcy court's jurisdictional reach under the Bankruptcy Code to release the claims encompassed by the Shareholder Release. The district court agreed that the bankruptcy court had subject-matter jurisdiction over all claims because: (1) the third-party claims raised questions as to the distribution of the Estates' property, id. at 85; (2) the third-party claims might have altered the liabilities of the Debtors and changed the amount available from the res, id. at 85-86; (3) the claims had a high degree of interconnectedness with claims against the Debtors, id. at 86-87; and (4) Purdue's insurance obligations to members of the Sacklers who were officers of Purdue could have burdened the res. Id. at 87-88. Accordingly, having found that the release of the third-party claims "might have some conceivable effect on the estate of a debtor," the district court concluded that they fell within the bankruptcy court's jurisdiction. Id. at 89 (emphasis in original).

2. Statutory Power to Release Third-Party Claims

Turning to the primary issue in this appeal, the district court next ruled that the bankruptcy court did not have statutory authority to release third-party direct claims against the Sacklers because the Sacklers were not the Debtors, and the Bankruptcy Code does not authorize the "non-consensual" release of "direct/particularized claims asserted by third parties against non-debtors." Purdue II, 635 B.R. at 90 (emphasis in original).

The district court's analysis on this issue considered the case law from this Court, the Supreme Court, and other circuit courts. It characterized this Court's holding in Metromedia as indicating that third-party releases could be permissible, but as being inconclusive as to whether "such releases [a]re consistent with or authorized by the Bankruptcy Code." Purdue II, 635 B.R. at 101. And, to the extent that Metromedia suggested that such releases would be permissible in "unique instances," the district court viewed the opinion as having failed to identify what those instances are. Id. (internal quotation marks omitted). Due to this perceived lack of clarity, the district court concluded that "while Metromedia said a great deal, the case did not hold much of anything," id., and thus a bankruptcy court's statutory authority to impose third-party releases is "questionable." Id. at 89.

Moving on to the Supreme Court, the district court acknowledged that although the Court has never spoken directly on whether the Bankruptcy Code provides authority for these types of releases, it has held, "albeit in contexts different from the one at bar, that a bankruptcy court lacks the power to award relief that varies or exceeds the protections contained in the Bankruptcy Code," and it lacks such power "even in 'rare' cases, and [ ] even when those orders would help facilitate a particular reorganization." Id. at 94-96 (citing Law v. Siegel, 571 U.S. 415, 134 S.Ct. 1188, 188 L.Ed.2d 146 (2014) and Czyzewski v. Jevic Holding Corp., 580 U.S. 451, 137 S.Ct. 973, 197 L.Ed.2d 398 (2017)).

At bottom, the district court concluded that no section of the Bankruptcy Code expressly or impliedly provided the requisite statutory authority for the Releases. Id. at 115. The district court also rejected the argument that the bankruptcy court possessed residual equitable authority to impose the Releases. Id. at 112-14. The district court further ruled that the fact that the Plan required the Releases for confirmation did not vest the bankruptcy court with authority to approve them. Id. at 108-09.

3. Classification of Canadian Claims

Finally, the district court agreed with the bankruptcy court that the Canadian Appellants' claims were properly classified differently than those of the domestic claimants, and that all the Bankruptcy Code requires is a reasonable basis for differentiation. Id. at 116-17. The equal treatment mandate applies only to creditors within the same class, and the district court held that, under this Court's precedent, there was a reasonable basis to differentiate the Canadian creditors' claims because different regulatory regimes apply, and because the mediation solely involved U.S.-based claimants. Id.

This Appeal followed.

D. This Appeal

The Appellants include a variety of interests unified in favor of the confirmation of the Plan: the Debtors, the Official Committee of Unsecured Creditors (the "UCC"), the Ad Hoc Committee of Governmental and Contingent Litigation Claimants (the "AHC"), the Ad Hoc Group of Individual Victims of Purdue Pharma, L.P. ("Pl. Ad Hoc Group"), the Multi-State Governmental Entities Group (the "MSGE"), the Mortimer-side Initial Covered Sackler Persons (the "Mortimer Sacklers"), and the Raymond Sackler Family (the "Raymond Sacklers," and together with the Mortimer Sacklers, the "Sacklers" or "Sackler family").

The UCC is composed of eight dedicated members, including individuals who are themselves (or whose loved ones are) victims of the opioid epidemic, representatives of a trade association for 35 independent health insurance companies collectively insuring 110 million members, a member of one of the largest hospital systems in the United States, the Pension Benefit Guaranty Corporation (the federal entity responsible for insuring defined benefit pension plans), a co-defendant in opioid litigation that has asserted indemnification claims against the Debtors, and three ex officio members that represent political subdivisions, tribes, and public school districts.

The AHC is composed of ten States, the court-appointed Plaintiffs' Executive Committee in the multi-district litigation captioned In re National Prescription Opiate Litigation, Case No. 17-md-02804 (DAP) (N.D. Ohio), six counties, cities, parishes, or municipalities, and one federally recognized American Indian Tribe.

This group comprises over 60,000 individuals who were injured by direct exposure to Purdue's opioid products, who together make up approximately one-half of those who filed personal injury claims in Purdue's Chapter 11 Cases.

Members of the MSGE Group are creditors of the Debtors, and many filed prebankruptcy lawsuits against them for their role in fostering the nationwide opioid crisis.

While this Appeal was pending, eight states—California, Connecticut, Delaware, Maryland, Oregon, Rhode Island, Vermont, and Washington—and the District of Columbia (the "Nine") that had appealed the confirmation of the original settlement, the Debtors, and the Sacklers filed a new settlement agreement with the bankruptcy court that provided for an additional $1.175-$1.675 billion in Sackler contributions (resulting in an aggregate $5.5 to $6.0 billion contribution to the Plan). See Order Pursuant to 11 U.S.C. §§ 105 and 363(b) Authorizing and Approving Settlement Term Sheet, In re Purdue PharmaL.P., No. 19-23649 (Bankr. S.D.N.Y. Mar. 10, 2022), ECF No. 4503. The bankruptcy court granted the motion to confirm the revised plan but noted that its confirmation would require one or more orders by this Court or the district court. Id. As part of the revised settlement agreement, the Nine agreed to withdraw their opposition to the Plan, including the Shareholder Releases. Id.

As a result, the Appellees currently left defending the district court's decision include only U.S. Trustee William K. Harrington ("the Trustee"), several Canadian municipalities and indigenous nations (the "Canadian Creditors"), and several individual pro se personal injury claimants (Ronald Bass, Ellen Isaacs, Maria Ecke, Richard Ecke, Andrew Ecke, the Estate of David Jonathan Ecke, and Peter Sottile).

Congress has authorized the Attorney General to appoint U.S. Trustees, who are Department of Justice officials, to supervise the administration of bankruptcy cases. 28 U.S.C. §§ 581-589a. U.S. Trustees "serve as bankruptcy watch-dogs to prevent fraud, dishonesty, and overreaching in the bankruptcy arena." H.R. Rep. No. 95-595, at 88 (1977). They "may raise and may appear and be heard on any issue in any case or proceeding" brought under the Bankruptcy Code. 11 U.S.C. § 307. Congress specifically empowered U.S. Trustees to comment on proposed disclosure statements and Chapter 11 plans of reorganization. 28 U.S.C. § 586(a)(3)(B).

DISCUSSION

I. Standard of Review

A. The Bankruptcy Court's Adjudicatory Authority

As stated by the district court, to the extent claims encompassed by the third-party releases are non-core under Stern v. Marshall, the bankruptcy court was required to submit "proposed findings of fact and conclusions of law to the district court, for that court's [de novo] review and issuance of final judgment." 564 U.S. 462, 471, 131 S.Ct. 2594, 180 L.Ed.2d 475 (2011). Stern defines core claims as those stemming "from the bankruptcy itself" or those which "would necessarily be resolved in the claims allowance process." Id. at 499, 131 S.Ct. 2594. For substantially the same reasons articulated by the district court, see Purdue II, 635 B.R. at 79-83, we agree that the bankruptcy court lacked constitutional authority to finally approve of the releases, and, therefore, that the district court correctly construed the bankruptcy court's decision as setting forth its proposed findings of fact and conclusions of law for the district court's de novo review. In short, the released claims at issue here—which, pursuant to the Plan, are permanently enjoined, have res judicata effect, and, as such, are effectively finally resolved—do not stem "from the bankruptcy itself," Stern, 564 U.S. at 499, 131 S.Ct. 2594, but are direct claims, arising under state law, against non-debtors held by third parties who have not sought to recover on those claims in bankruptcy, or otherwise consented to a bankruptcy court's adjudication of those claims.

It is true, as Debtors note, that the resolution of the third-party claims might impact the res of the Estate—a fact determinative of the district court's statutory jurisdiction under the Code—but the same was true for the counterclaims held in Stern to be beyond the bankruptcy court's constitutional reach to finally determine. To the point, had the debtor in Stern been successful on her counterclaim against the creditor, the value of the estate would have been impacted; she would have had a property interest in the resulting damages award, which would have, in turn, increased the value of her estate. See id. The focus of the constitutional analysis in Sterndoes not turn on the extent to which the non-core claim might alter the creditor-debtor relations in a given bankruptcy. That said, we agree with the district court that the practical import of the Stern issue is nonexistent given that only conclusions of law are at issue here, requiring our de novo review under any standard. See Purdue II, 635 B.R. at 82 n.54.

B. Appellate Review of the District Court

In an appeal from a district court's review of a bankruptcy court's decision, this Court "independently" reviews the bankruptcy court's conclusions of law de novo and its factual findings for clear error. Morning Mist Holdings Ltd. v. Krys (In re Fairfield Sentry Ltd.), 714 F.3d 127, 132 (2d Cir. 2013). A factual finding is clearly erroneous when "the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." United States v. U.S. Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948). "[I]n reviewing factual findings for clear error, an appellate court is not confined to evidence cited in a lower court's opinion, but must instead review all of the record evidence." Bankr. Servs, Inc. v. Ernst & Young (In re CBI Holding Co.), 529 F.3d 432, 449 (2d Cir. 2008).

This Court may uphold a bankruptcy court decision on any ground—even one not relied upon by the district court. Resol. Tr. Corp. v. Best Prods. Co. (In re Best Prods. Co.), 68 F.3d 26, 30 (2d Cir. 1995). As such, we decide all pertinent issues necessary to confirm the Plan and do not limit our analysis solely to the issues addressed below.

II. Nonconsensual Third-Party Releases of Direct Claims

The two primary questions posed on appeal are: (1) whether the bankruptcy court had the authority to approve the nonconsensual release of direct third-party claims against the Sacklers, a non-debtor, through the Plan; and (2) whether the text of the Bankruptcy Code, factual record, and equitable considerations support the bankruptcy court's approval of the Plan. We answer both in the affirmative.

To explain our reasoning, we begin by describing the scope of the Shareholder Releases (including the types of claims covered and the claims at issue here). We then address the various statutory and constitutional arguments raised by the parties. Finally, we evaluate the bankruptcy court's findings regarding the fairness and equitable nature of the Plan, and we articulate factors to help guide future courts evaluating similar issues.

A. The Scope of the Releases

The original version of the Release from the September 2, 2021 Plan of Reorganization settles

any and all Causes of Action, including . . . [present and future claims], (x) based on or relating to, or in any manner arising from, in whole or in part, (i) the Debtors, . . . (including the Debtors' Opioid-Related Activities, manufacture, marketing and sale of Products, interaction with regulators concerning Opioid-Related Activities or Products, and involvement in the subject matter of the Pending Opioid Actions, and the past, present or future use or misuse of any opioid by a Releasing Party) . . . and (y) as to which any conduct, omission or liability of any Debtor or any Estate is the legal cause or is otherwise a legally relevant factor.

Special App'x at 920. As discussed supra, the bankruptcy court subsequently limited the releases such that they only "apply where . . . a debtor's conduct or the claims asserted against it [are] a legal cause or a legally relevant factor to the cause of action against the shareholder released party," Deferred Joint App'x at 1330-31, and the released claims directly affect the res, Purdue I, 633 B.R. at 97-98.

The released claims can be grouped into two categories: direct claims and derivative claims. In this context, direct claims are causes of action brought to redress a direct harm to a plaintiff caused by a non-debtor third party. See Marshall v. Picard (In re Bernard L. Madoff Inv. Secs. LLC), 740 F.3d 81, 89 n.9 (2d Cir. 2014). By contrast, derivative claims are "ones that arise from harm done to the estate and that seek relief against [the] third part[y] that pushed the debtor[s] into bankruptcy." Id. at 89 (internal quotation marks and alterations omitted); see also Tronox Inc. v. Kerr-McGee Corp. (In re Tronox Inc.), 855 F.3d 84, 100-04 (2d Cir. 2017) (explaining the law of derivative claims in the bankruptcy context). The potential claims released against the Sacklers include, inter alia, fraudulent transfer, constructive fraudulent transfer, deceptive marketing, public nuisance, unfair competition, fraudulent misrepresentation, violation of state consumer protection acts, civil conspiracy, negligence, and unjust enrichment. Some of these claims are direct, and some are derivative. As conceded by the parties, fraudulent transfer claims, for example, are typically derivative claims in that the real injury is to the Debtors' estate, and it is well-settled that a bankruptcy court may approve not only third-party releases which are consensual, but also third-party releases of derivative claims because those claims really belong to the estate of the debtor. See, e.g., 11 U.S.C. § 1123(b)(3)(A) (permitting release of claims as to the estate's property); Madoff, 740 F.3d at 88 ("A claim based on rights derivative of, or derived from, the debtor's typically involves property of the estate. By contrast, a bankruptcy court generally has limited authority to approve releases of a nondebtor's independent claims." (internal citation and quotation marks omitted)). The more controversial issue, however, is this Plan's likely release of some direct claims against the Sacklers.

Although the Plaintiff Ad Hoc Group contends that the district court erred in concluding the claims against the Sacklers are not all derivative, we find no error because certain consumer protection act claims at a minimum constitute direct claims in that the injury belongs directly to the claimant, and not to the Debtors. We need not define the exact claims which fall under the umbrella of direct claims but note that certain state law claims under consumer protection acts likely do.

The bankruptcy court's ability to release claims at all derives from its power of discharge. See generally 11 U.S.C. § 524(a). Under the Bankruptcy Code, a bankruptcy discharge releases a debtor from personal liability with respect to any debt by enjoining creditors from attempting to collect on that debt, so long as the debtor discloses all its financial information and puts those assets towards its estate. 11 U.S.C. § 524; Tenn. Student Assistance Corp. v. Hood, 541 U.S. 440, 447, 124 S.Ct. 1905, 158 L.Ed.2d 764 (2004) ("The discharge order releases a debtor from personal liability with respect to any discharged debt by voiding any past or future judgments on the debt and by operating as an injunction to prohibit creditors from attempting to collect or to recover the debt."); see also 11 U.S.C. §§ 521-523. This extraordinary remedy is based on bankruptcy courts' in rem jurisdiction over the property of the debtor. While the Bankruptcy Code forbids a discharge of a non-debtor's claim under 11 U.S.C. § 524(e), the releases at issue on appeal do not constitute a discharge of debt for the Sacklers because the releases neither offer umbrella protection against liability nor extinguish all claims. See MacArthur Co. v. Johns-Manville Corp. ("Manville I"), 837 F.2d 89, 91 (2d Cir. 1988) (ruling that the bankruptcy court had the authority to enjoin third-party claims because "the injunctive orders d[id] not offer the umbrella protection of a discharge in bankruptcy" and were instead limited to suits "that ar[o]se out of or relate[d] to" specific issues central to the bankruptcy).

Thus, the primary dispute is whether direct claims brought by creditors of Purdue against the Sacklers (for which the Debtors' conduct is legally relevant) can be released. As described in the following sections, we conclude that the bankruptcy court possessed both jurisdiction and statutory authority to approve the Releases because the limitations on the scope of the releases are significant and no other argument bars their imposition.

B. Subject-Matter Jurisdiction

As an initial matter, we must ensure the bankruptcy court had subject-matter jurisdiction, pursuant to the Bankruptcy Code, over the released claims. See Joseph v. Leavitt, 465 F.3d 87, 89 (2d Cir. 2006) ("[W]e have an independent obligation to consider the presence or absence of subject matter jurisdiction sua sponte.").

A bankruptcy court's subject-matter jurisdiction under the Code is broad. It extends to all civil actions so long as "the action's outcome might have any conceivable effect on the bankrupt estate." Parmalat Cap. Fin. Ltd. v. Bank of Am. Corp., 639 F.3d 572, 579 (2d Cir. 2011) (internal quotation marks omitted); see also 28 U.S.C. §§ 157(a), 1334. However, that jurisdictional reach is not endless: a bankruptcy court may only "enjoin third-party non-debtor claims that directly affect the res of the bankruptcy estate." Johns-Manville Corp. v. Chubb Indemnity Ins. Co. ("Manville III"), 517 F.3d 52, 66 (2d Cir. 2008). That limitation is in line with the goal that "extending bankruptcy jurisdiction to actions against certain third parties, as well as suits against debtors themselves, is to protect the assets of the estate so as to ensure a fair distribution of those assets at a later point in time." Pfizer Inc. v. Law Offices of Peter G. Angelos (In re Quigley Co.), 676 F.3d 45, 57 (2d Cir. 2012) (internal quotation marks and alteration marks omitted).

A direct claim brought against non-debtors, such as the Sacklers, "that nevertheless poses the specter of direct impact on the res of the bankrupt estate may just as surely impair the bankruptcy court's ability to make a fair distribution of the bankrupt's assets as a third-party suit alleging derivative liability." Id. at 58. Accordingly, if, for example, the litigation of the settled claims "would almost certainly result in the drawing down of . . . the bankruptcy estate of [the debtor], the exercise of bankruptcy jurisdiction to enjoin [third-party direct claims is] appropriate." Id. Thus, as to statutory jurisdiction, our key inquiry is into the likely impact on the res.

We agree with both the bankruptcy court and the district court that the bankruptcy court had statutory jurisdiction to impose the Releases because it is conceivable, indeed likely, that the resolution of the released claims would directly impact the res.

First, as both courts below noted, at least some of the third-party claims, although directly asserted against the Sacklers, are closely related to the derivative claims which the Estate might bring against the Sacklers. For example, many of the states that, below, objected to the Plan (but have since withdrawn their claims in favor of settlement) have laws which impose direct liability on individuals who, as officers of a corporation, personally participated in acts of corporate fraud. See, e.g., U.S. Trustee's App'x at 2644-47, 2765, In re Purdue Pharma L.P., No. 21-07532 (S.D.N.Y. Sept. 9, 2021), ECF Nos. 91-7, 91-8. However, although the various state statutes ensure that managerial personnel can be held independently liable for the same conduct that subjects the corporation to liability, those claims often "rely on detailed and virtually identical set of facts to make the claims" against both Purdue and the Sacklers. Purdue II, 635 B.R. at 86. As a result of that substantial overlap, the litigation of third-party direct claims against the Sacklers would likely impact the Debtor's ability to pursue, and the likelihood of recovering on, the Estate's own claims against the Sacklers.

Second, the Sacklers are covered by the Sackler-Purdue Indemnity Agreement, and, therefore, depending on the outcome of any given claims against them, would have a reasonable basis to seek indemnification from the Debtors. That possibility is enough to implicate the bankruptcy court's "related to" jurisdiction under our precedent. See SPV Osus Ltd. v. UBS AG, 882 F.3d 333, 341-42 (2d Cir. 2018).

In addition to indemnification claims, the Sacklers might also assert claims against the Estate for either insurance coverage or contribution. See generally Appellees' Suppl. App'x at 627-35, Bryant Dunaway v. Purdue Pharma L.P. (In re Purdue Pharma L.P.), No. 19-10941 (CM) (S.D.N.Y. June 22, 2020), ECF No. 24-2.

In SPV, the plaintiffs asserted direct claims against, among other defendants, UBS AG, alleging principally that UBS had aided and abetted the infamous fraud perpetrated by the debtor, Bernard L. Madoff Investment Securities LLC. See SPV, 882 F.3d at 338. Although the plaintiffs sought recovery from UBS itself, UBS, in turn, had viable claims for indemnification and contribution against the debtor. See id. at 340-42. The possibility that those claims might have succeeded—and the fact that the debtor would incur expense in litigating such claims—was enough to confer jurisdiction on the bankruptcy court to enjoin the plaintiff's direct claims against UBS. See id. at 341-42.

To be sure, the Indemnity Agreement plainly bars any indemnification obligation flowing from the Debtors to the Sacklers where a court determines the Sacklers "did not act in good faith." SA 629. Consequently, as to any successful claims against the Sacklers sounding in fraud (such as the state consumer protection claims), the Sacklers would not have any reasonable basis to seek indemnification. Yet, as the district court noted, "the question of bad faith in this case is hotly disputed." Purdue II, 635 B.R. at 88. In the end, the jurisdictional issue does not require us to resolve that question; the relevant inquiry is whether the claims for indemnification "might have any conceivable effect on the bankrupt estate." SPV, 882 F.3d at 339-40 (emphasis added) (internal citation omitted). That standard is plainly satisfied here.

C. Bankruptcy Code Authority

The ultimate authority for the imposition of nonconsensual releases of direct third-party claims against non-debtors is rooted—as it must be—in the Bankruptcy Code, specifically 11 U.S.C. §§ 105(a) and 1123(b)(6). Further bolstering this statutory authority is this Circuit's caselaw stating that a bankruptcy court has authority to impose such releases.

1. Statutory Authority

The bankruptcy court correctly grounded its authority for approving the Releases in §§ 105(a) and 1123(b)(6), which provide the statutory basis for the bankruptcy court's equitable authority and permit the bankruptcy court's approval of the Plan. 11 U.S.C. § 105(a) states that "[t]he court may issue any order, process, or judgment that is necessary or appropriate to carry out the provisions of [the Bankruptcy Code]." 11 U.S.C. § 1123(b)(6) states that "a plan may . . . include any other appropriate provision not inconsistent with the applicable provisions of this title." We deem Appellees' arguments—that since the Bankruptcy Code does not explicitly authorize third-party releases, they are outside of a bankruptcy court's statutory authority—unpersuasive.

First, although we have stated that § 105(a) gives "broad equitable power to the bankruptcy courts to carry out the provisions of the Bankruptcy Code," Adelphia Bus. Sols., Inc. v. Abnos, 482 F.3d 602, 609 (2d Cir. 2007) (emphasis added), we reject Appellants' suggestion that § 105(a) alone supports the imposition of the releases in this action. Indeed, our case law, and that of the majority of our sister circuits, support the proposition that § 105(a) alone cannot justify the imposition of third-party releases. See New England Dairies, Inc. v. Dairy Mart Convenience Stores, Inc. (In re Dairy Mart Convenience Stores, Inc.), 351 F.3d 86, 92 (2d Cir. 2003) (ruling that an exercise of § 105(a) power must "be tied to another Bankruptcy Code section and not merely to a general bankruptcy concept or objective"); see, e.g., Brown v. Viegelahn (In re Brown), 960 F.3d 711, 719-20 (5th Cir. 2020) (ruling that bankruptcy courts must link Section 105(a) with another provision of the Bankruptcy Code); Bird v. Carl's Grocery Co. (In re NWFX, Inc.), 864 F.2d 593, 595 (8th Cir. 1989) (same); Southern Ry. Co. v. Johnson Bronze Co. (In re Johnson Bronze Co.), 758 F.2d 137, 141 (3d Cir. 1985) (same). Thus, at least one other provision of the Bankruptcy Code must provide the requisite statutory authority. Section 1123(b)(6) does.

As previously stated, 11 U.S.C. § 1123(b)(6) permits the inclusion of "any other appropriate provision" in a plan so long as it is "not inconsistent" with other sections of the Bankruptcy Code. In United States v. Energy Resources Co., Inc., the Supreme Court held that this provision—acting in tandem with § 105(a)—grants bankruptcy courts a "residual authority" consistent with "the traditional understanding that bankruptcy courts, as courts of equity, have broad authority to modify creditor-debtor relationships." 495 U.S. 545, 549, 110 S.Ct. 2139, 109 L.Ed.2d 580 (1990) (emphasis added). Thus, in Energy Resources, the Court, relying on § 1123(b)(6), permitted bankruptcy courts "to approve reorganization plans designating tax payments as either trust fund or nontrust fund"—even absent express authorization from the Bankruptcy Code. Id.at 545, 110 S.Ct. 2139. Appellees, however, nevertheless argue that Energy Resources does not permit reliance on § 1123(b)(6) because the third-party releases at issue here are "not specifically authorized by the Code." Trustee Br. at 48. Appellees further maintain that Energy Resources only speaks to the ability of bankruptcy courts to modify "creditor-debtor" relationships, and that these releases go beyond such relationships. Trustee Br. at 54.

Energy Resources refers to 11 U.S.C. § 1123(b)(5). That provision was later recodified as § 1123(b)(6).

We are not persuaded by Appellees' arguments. First, as the Court's language in Energy Resources indicates, § 1123(b)(6) is limited only by what the Code expressly forbids, not what the Code explicitly allows. Second, and as the Seventh Circuit convincingly has held, bankruptcy courts' equitable powers under § 1123(b)(6) include the power "to release third parties from liability." Airadigm Commc'ns, Inc. v. FEC (In re Airadigm Commc'ns, Inc.), 519 F.3d 640, 657 (7th Cir. 2008). The Sixth Circuit has also ruled that the residual authority grounded in §§ 105(a) and 1123(b)(6) supports a bankruptcy court's power to impose third-party releases. Class Five Nev. Claimants (00-2516) v. Dow Corning Corp. (In re Dow Corning Corp.), 280 F.3d 648, 656-58 (6th Cir. 2002) (concluding that third-party releases can be appropriate, but that the factual findings presented did not support them). Although our case law has never expressly cited § 1123(b)(6) to support the imposition of third-party releases, we now explicitly agree with these Circuits and conclude that § 1123(b)(6), with § 105(a), permit bankruptcy courts' imposition of third-party releases.

Our sister circuits that have held that the Bankruptcy Code does not support the imposition of nonconsensual third-party releases rely upon the provisions limiting the discharge of debt under 11 U.S.C. § 524(e). See Bank of N.Y. Tr. Co. v. Official Unsecured Creditors' Comm. (In re Pac. Lumber Co.), 584 F.3d 229, 251-53 (5th Cir. 2009); Resorts Int'l, Inc. v. Lowenschuss (In re Lowenschuss), 67 F.3d 1394, 1401-02 (9th Cir. 1995); Landsing Diversified Props.-II v. First Nat'l Bank and Tr. Co. of Tulsa (In re W. Real Estate Fund, Inc.), 922 F.2d 592, 600-02 (10th Cir. 1990). Section 524(e) states that "discharge of a debt of the debtor does not affect the liability of any other entity on, or the property of any other entity for, such debt." 11 U.S.C. § 524(e).

This language assures that an entity also liable with a bankruptcy debtor for "such debt" remains liable notwithstanding the debtor's discharge of its obligation. For example, the entity might be jointly liable for the debt.

The circuits that have read § 524(e) as a bar to third-party releases have reasoned that "it is the debtor[ ] who has invoked and submitted to the bankruptcy process, that is entitled to its protections; Congress did not intend to extend such benefits to third-party bystanders." In re W. Real Estate Fund, Inc., 922 F.2d at 600-02 (quoting 11 U.S.C. § 524(e)); In re Pac. Lumber Co., 584 F.3d at 252 ("In a variety of contexts, this court has held that Section 524(e) only releases the debtor, not co-liable third parties. These cases seem broadly to foreclose non-consensual non-debtor releases and permanent injunctions." (internal citations omitted)); In re Lowenschuss, 67 F.3d at 1401 ("This court has repeatedly held, without exception, that § 524(e) precludes bankruptcy courts from discharging the liabilities of nondebtors.").

In contrast to these holdings, we do not consider 11 U.S.C. § 524(e) to be a bar to such releases. As explained by the Seventh Circuit in Airadigm:

§ 524(e) does not purport to limit the bankruptcy court's powers to release a non-debtor from a creditor's claims. If Congress meant to include such a limit, it would have used the mandatory terms "shall" or "will" rather than the definitional term "does." And it would have omitted the prepositional phrase "on, or . . . for, such debt," ensuring that the "discharge of a debt of the debtor shall not affect the liability of another entity"—whether related to a debt or not.

519 F.3d at 656. Moreover, "where Congress has limited the powers of the bankruptcy court, it has done so clearly—for example, by expressly limiting the court's power . . . or by creating requirements for plan confirmation." Id. (citing to 11 U.S.C. § 105(b) ("a court may not appoint a receiver in a case under this title") and 11 U.S.C. § 1129(a) ("The court shall confirm a plan only if the following requirements are met") as illustrative examples). Following this logic, we see no reason grounded in the text of the Bankruptcy Code to bar the inclusion of third-party releases in plans of reorganization.

2. Second Circuit Case Law

Despite the district court's pronouncement to the contrary, Purdue II, 635 B.R. at 89, this Court's precedents permit the imposition of nonconsensual third-party releases. Appellants uniformly agree that our precedents support the approval of a plan containing nonconsensual third-party releases. See, e.g., AHC Br. at 18 ("This Court has held on multiple occasions that third-party releases are allowed in appropriate circumstances."); Debtors Br. at 32 ("For more than three decades, this Court has held that bankruptcy courts are authorized to enjoin and release third-party claims against non-debtors, as part of a plan of reorganization, in appropriate circumstances."). But Appellees contend that such releases are the equivalent of an inappropriate discharge, that this Circuit at no point has permitted the release of direct third-party claims in non-asbestos actions, and that no case supports a plan doing so here. Trustee Br. at 69-77; Canadian Creditors Br. at 27-35. That reading is incorrect in the face of our case law, most explicitly Drexel, where we concluded: "In bankruptcy cases, a court may enjoin a creditor from suing a third party, provided the injunction plays an important part in the debtor's reorganization plan." In re Drexel Burnham Lambert Group, Inc. ("Drexel"), 960 F.2d 285, 293 (2d Cir. 1992). Our opinions in Manville I and Metromedia further confirm that such releases are neither discharges nor allowable only in the context of asbestos cases.

Manville I stated that injunctive orders barring third-party claims are not necessarily impermissible discharges. 837 F.2d at 91. There, we were presented with a Chapter 11 bankruptcy plan that released over $2 billion in asbestos victims' claims against the insurers of Manville, a distributor of asbestos products. 837 F.2d at 90. While Manville was a debtor in the bankruptcy, its insurers were not. Id. at 91. Thus, to obtain the releases, the insurers paid Manville a $770 million settlement. Id. at 94. Before this Court, the appellant (a distributor of Manville's products) challenged the bankruptcy court's jurisdiction and authority by arguing that the third-party releases operated as a bankruptcy discharge that cannot be granted to non-debtors under the Bankruptcy Code. Id. at 91. We disagreed and ruled that the releases did not constitute a bankruptcy discharge because they (1) did not offer the umbrella protection of a discharge, and (2) did not extinguish the claims against the insurer, but rather "channeled" them "away from the insurers and redirected [them to] the proceeds of the settlement." Id. at 91. Moreover, the insurers' rights were "completely derivative of" and "inseparable from" the debtor's rights. Id. at 92-93. Thus, plaintiffs' released claims fell well-within the bankruptcy court's jurisdiction over the debtor's estate. Id.

We also stated in Manville I that the bankruptcy court properly imposed the releases under the Bankruptcy Code. While the bankruptcy court primarily relied on § 363(f)—which permits channeling orders (the funneling of claims into one proceeding to preserve the debtors' estate) under certain circumstances applicable to Manville I—it also looked to § 105(a) for additional support. Id. at 93; id. at 94 (noting both statutory and equitable powers to dispose of the debtor's property free from third-party interests). Moreover, the releases there were "essential" to a "workable reorganization." Id. at 94. Thus, although Manville I was in the asbestos context, its premise that this Circuit permits third-party releases in bankruptcy still stands. See In re Metromedia Fiber Network, Inc. ("Metromedia"), 416 F.3d 136, 142 (2d Cir. 2005) (citing Manville I); Drexel, 960 F.2d at 293 (recognizing the propriety of third-party releases in a reorganization).

Appellees argue, however, that it is significant that Manville I, unlike the current appeal, concerned asbestos products because the Bankruptcy Code now explicitly authorizes releases in such circumstances. Trustee Br. at 41-42. That is because in 1994 Congress enacted 11 U.S.C. § 524(g), which expressly allows for the injunction of third-party claims against non-debtors in "actions seeking recovery for damages allegedly caused by the presence of, or exposure to, asbestos or asbestos-containing products." 11 U.S.C. § 524(g)(2)(B)(i)(I); see Bankruptcy Reform Act of 1994, Pub. L. No. 103-394, 108 Stat. 4106 (1994). Thus, under Appellees' view, "[h]ad Congress intended to allow bankruptcy courts to adjust the relationship between non-debtors and other non-debtors in this manner, it would have said so expressly—as it did when it authorized narrow non-debtor releases in the context of bankruptcies involving asbestos." Trustee Br. at 3.

The first blow to Appellees' restrictive reading of the statute comes from the text of the Bankruptcy Reform Act of 1994 itself, which states:

RULE OF CONSTRUCTION.—Nothing in [the language since enacted as § 524(g)], shall be construed to modify, impair, or supersede any other authority the court has to issue injunctions in connection with an order confirming a plan of reorganization.

Pub. L. 103-394, § 111(b), 108 Stat. 4106, 4117 (1994). Thus, in enacting § 524(g), Congress expressly intended not to change the pre-existing powers of bankruptcy courts. Therefore, neither Manville I nor the subsequent adoption of § 524(g) supports a limitation of its reasoning to asbestos claims.

More importantly, this Court's opinion in Metromedia flatly rejects a restrictive interpretation of the Bankruptcy Code by stating that third-party releases can be valid outside of the asbestos context. 416 F.3d at 141. In that case, the debtor Metromedia's reorganization plan allowed certain non-debtor directors and officers of the company to "receive a full and complete release, waiver and discharge from . . . any holder of a claim of any nature . . . arising out of or in connection with any matter related to" Metromedia or its subsidiaries. Id. at 141 (alterations in original). Creditors challenged the imposition of these types of releases generally on statutory grounds, and specifically on equitable grounds. Although this Court ultimately rejected the imposition of the releases, we did so based on insufficient factual findings, not because we found that such releases could not ever be approved. Id. at 143.

Regarding the third-party releases themselves, the Metromedia court faced many of the same arguments we are presented with today. There, appellants had primarily contended that the non-debtor releases were unauthorized by the Bankruptcy Code, at least on the findings made by the bankruptcy court. Id. at 141. But in Metromedia, we did not accept those arguments. Instead, we noted that "[w]e have previously held that 'in bankruptcy cases, a court may enjoin a creditor from suing a third party, provided the injunction plays an important part in the debtor's reorganization plan.' " Id. at 141 (alterations adopted) (quoting Drexel, 960 F.2d at 293); see also Metromedia, 416 F.3d at 141 ("it is clear that such a release is proper only in rare cases"). And, while we acknowledged that some circuits have permitted such releases only in the asbestos context, id., we focused on the circumstances under which other circuits "have approved non-debtor releases," such as when: "the estate received substantial consideration," the plan channeled enjoined claims to a settlement fund as opposed to extinguishing them, "the enjoined claims would indirectly impact the debtor's reorganization" due to factors like indemnification, "the plan otherwise provided for the full payment of the enjoined claims," and affected creditors consent to such releases. Id. at 142. Following this review, we then articulated two requirements for the imposition of such releases in this Circuit. First, in order for the inclusion of a release to be approved, the release "itself" must be "important to the Plan." Id. at 143 (emphasis in the original). Second, the "breadth" of the release must also be "necessary to the Plan." Id.

Thus, while we ultimately ruled that the bankruptcy court's findings were insufficient for the imposition of releases under the facts of that case, Metromedia nevertheless rests upon the premise that such releases may be permitted so long as bankruptcy courts make sufficient factual findings and satisfy certain equitable considerations. Id. at 143.

For these reasons, our precedents permit the imposition of third-party releases jointly under 11 U.S.C. § 105(a) and 11 U.S.C. § 1123(b)(6).

D. Factors Relevant to Releasing Direct Third-Party Claims Against Non-Debtors

Having upheld the bankruptcy court's statutory authority and jurisdiction to impose such releases, we now turn to the circumstances under which releases may be approved. The Trustee appears to take issue with the fact that the Releases were approved despite their failing to satisfy certain factors stated in Metromedia. The Debtors, by contrast, contend that this is exactly the sort of case that epitomizes when third-party nonconsensual releases are proper because (1) the releases are essential to the confirmation of the Plan (including serving as its primary financing); (2) litigation of the settled claims would negatively impact the res of the Debtors' estates; (3) the bankruptcy court already narrowed the scope of the releases; and (4) this case is highly unusual and complex given the "inextricable interrelation between the claims against the Debtors and against the Sacklers," Debtors Br. at 65.

We now clarify any ambiguity and identify the factors that should be considered in order for a bankruptcy court to approve of nonconsensual third-party releases of direct claims against a non-debtor and to include them in a plan. In doing so, we remain conscious of the "heightened" "potential for abuse" posed by such releases, and our analysis of pertinent factors is informed by that risk.Metromedia, 416 F.3d at 140. We wholeheartedly endorse the view that "third-party releases are not a merit badge that somebody gets in return for making a positive contribution to a restructuring," nor are they "a participation trophy" or "gold star for doing a good job." In re Aegean Marine Petroleum Network Inc., 599 B.R. 717, 726-27 (Bankr. S.D.N.Y. 2019).

This Court has also observed that it is abusive for a bankruptcy court to enjoin third-party claims against a non-debtor based solely on the non-debtor's financial contribution to the estate. Manville III, 517 F.3d at 66. "It is . . . precisely this conditioning of financial participation by non-debtors on releases that is subject to the sort of abuse foreseen in Metromedia." Id. (internal quotation marks omitted).

With that said, bankruptcy courts should look to the following seven factors before imposing nonconsensual third-party releases:

First, courts should consider whether there is an identity of interests between the debtors and released third parties, including indemnification relationships, "such that a suit against the non-debtor is, in essence, a suit against the debtor or will deplete the assets of the estate." Dow Corning, 280 F.3d at 658; see also In re Master Mortgage Investment Fund, 168 B.R. 930, 935 (Bankr. W.D. Mo. 1994) (same). This requirement reflects our observation in Metromedia that nondebtor releases have been allowed in circumstances including those where "the enjoined claims would indirectly impact the debtor's reorganization by way of indemnity or contribution." Metromedia, 416 F.3d at 142 (internal quotation marks omitted).

The multifactor test articulated in In re Master Mortgage Investment Fund has been widely cited by courts in other circuits. See, e.g., Monarch Life Ins. Co. v. Ropes & Gray, 65 F.3d 973, 980 (1st Cir. 1995); Gillman v. Continental Airlines (In re Cont'l Airlines), 203 F.3d 203, 217 n.17 (3d Cir. 2000); In re Chicago Invs., LLC, 470 B.R. 32, 95 (Bankr. D. Mass. 2012); In re U.S. Fidelis, Inc., 481 B.R. 503, 519 (Bankr. E.D. Mo. 2012).

Second, courts should consider whether claims against the debtor and nondebtor are factually and legally intertwined, including whether the debtors and the released parties share common defenses, insurance coverage, or levels of culpability. We note that although the bankruptcy court did not list this as a factor, it discussed that releases limited to those claims legally intertwined with the Debtors' conduct are appropriately subject to settlement. Purdue I, 633 B.R. at 104. We agree.

Third, courts should consider whether the scope of the releases is appropriate. This is the second factor evaluated in Metromedia. 416 F.3d at 143. In our view, a release is proper in scope when its "breadth" is "necessary to the Plan." Id.

Fourth, courts should consider whether the releases are essential to the reorganization, in that the debtor needs the claims to be settled in order for the res to be allocated, rather than because the released party is somehow manipulating the process to its own advantage. In other words, it must be the case that, without the releases, "there is little likelihood of [a plan's] success." Master Mortg. Inv. Fund, 168 B.R. at 935. This factor also reflects the first factor required by Metromedia—that the release be important to the plan. 416 F.3d at 143.

Fifth, courts should consider whether the non-debtor contributed substantial assets to the reorganization. This factor was mentioned by this Court in Metromedia, 416 F.3d at 142-43, and is emphasized in Dow Corning, 280 F.3d at 658, and Master Mortgage Investment Fund, 168 B.R. at 935.

Sixth, courts should consider whether the impacted class of creditors "overwhelmingly" voted in support of the plan with the releases. Master Mortg. Inv. Fund, 168 B.R. at 935. A reference point to define "overwhelmingly" can be found in 11 U.S.C. § 524(g)(2)(B)(ii)(IV)(bb), which requires approval by a minimum of 75% of voting creditors in favor of the plan. However, we consider that threshold to be the bare minimum, and instead express approval for requiring overwhelming approval of the plan.

Seventh, and finally, courts should consider whether the plan provides for the fair payment of enjoined claims. In Metromedia, we noted that other courts have found such releases permissible when "the plan . . . provided for the full payment of the enjoined claims." 416 F.3d at 142; see also Dow Corning, 280 F.3d at 658 (requiring that "[t]he plan provides a mechanism to pay for all, or substantially all, of the class or classes affected by the injunction"). While the full payment of the enjoined claims would of course tend to favor the approval of a plan containing such releases, we are concerned with the fairness of the payment, as opposed to the final amount of payment. Because the amount of the payment does not necessarily indicate its fairness, the determinative question is not whether there is full payment, but rather whether the contributed sum permits the fair resolution of the enjoined claims.

Although consideration of each factor is required, it is not necessarily sufficient—there may even be cases in which all factors are present, but the inclusion of third-party releases in a plan of reorganization should not be approved. Further, as contemplated by Dow Corning, the bankruptcy court is required to support each of these factors with specific and detailed findings. 280 F.3d at 658. For the bankruptcy court to make such findings, extensive discovery into the facts surrounding the claims against the released parties will most often be required.

Finally, as with any term in a bankruptcy plan, a provision imposing releases of claims like that at issue here must be imposed against a backdrop of equity. See Energy Resources, 495 U.S. at 549, 110 S.Ct. 2139 (describing the authority conferred by § 1123(b)(6) as deriving from bankruptcy courts' status as "courts of equity"); see also Adelphia Bus. Sols., Inc. v. Abnos, 482 F.3d 602, 609 (2d Cir. 2007) ("Section 105(a) grants broad equitable power to the bankruptcy courts to carry out the provisions of the Bankruptcy Code so long as that power is exercised within the confines of the Bankruptcy Code."). Given the potential for abuse, courts should exercise particular care when evaluating these types of releases.

E. Application of These Factors Based Upon the Bankruptcy Court's Findings

In light of these factors, we now evaluate the bankruptcy court's findings supporting its approval of the Plan. The thorough bankruptcy court opinion, which indicated that it grounded its findings in the tens of millions of documents produced in discovery, informs our analysis.

The extensive discovery provided by the parties is exactly the sort that bankruptcy courts should expect when permitting broad third-party releases.

Factor 1. Identity of Interests Between Debtors and Released Parties

We have described supra the identity of interests between the Debtors and those Sacklers named as defendants in the litigations, chiefly that the named Sacklers were directors and officers of the Debtors. Purdue was a closely held corporation, and, according to the bankruptcy court, the record tended to show that the Sacklers "took a major role in corporate decision-making, including Purdue's practices regarding its opioid products that was more akin to the role of senior management." Purdue I, 633 B.R. at 93. This overlap constitutes a sufficient identity of interests between the Debtors and the Sacklers.

Factor 2. Factual and Legal Overlap Between Claims Against Debtors and Settled Third-Party Claims

In the prior sections, we also discussed the factually and legally intertwined nature of the claims against both the Debtors and the Sacklers. More importantly, the bankruptcy court required that the releases only "apply where . . . a debtor's conduct or the claims asserted against it [are] a legal cause or a legally relevant factor to the cause of action against the shareholder released party," Deferred Joint App'x at 1330-31, and the released claims directly affect the res, Purdue I, 633 B.R. at 97-98. Cf. Metromedia, 416 F.3d at 141 (ruling that factual circumstances and equitable considerations did not support a broad release that included the "waiver and discharge from . . . any holder of a claim of any nature . . . of any and all claims . . . arising out of or in connection with any matter related to [the Debtor] or one or more subsidiaries . . . based in whole or in part upon any act or omission or transaction" (alterations in original, emphasis added)). By so narrowing the Releases, the bankruptcy court ensured sufficient overlap between claims against the Debtors and the settled third-party claims.

Factors 3. and 4. The Releases are Essential to the Reorganization & Proper in Scope

We next evaluate, in tandem with our analysis of the Releases' scope, whether the Releases are essential to reorganization.See Metromedia, 416 F.3d at 143. The Releases are essential to reorganization for two reasons. First and foremost, as described supra, the Releases are required to ensure that the valuation of the res is settled. Otherwise, the Debtors would, in all likelihood, be required to litigate indemnity and contribution claims brought against them by the Sacklers, which would likely deplete the res, no matter the ultimate outcome of those claims. The bankruptcy court limited the Releases extensively in order not to exceed its jurisdiction, restricting their scope to ensure that the released claims related to the Debtors' conduct and the Estate. Second, the res itself amounted to only approximately $1.8 billion. Without the Plan, the government would recover its $2 billion first, thereby depleting the res completely. As a result, many victims of the opioid crisis would go without any assistance and face an uphill battle of litigation (in which a single claimant might disproportionately recover) without fair distribution.

Although we describe these as two separate factors, following Metromedia, we analyze them together in this case because the two factors are interrelated. We nevertheless acknowledge that a case with a different factual record might require them to be considered separately.

On the question of what is essential to the Plan, the Trustee argues that the Sacklers themselves created the conditions that make these releases essential, and that, as a term of their contribution, the Sacklers had insisted upon these releases before the Debtors even entered bankruptcy. Per the Trustee, these facts demonstrate the Sacklers' unworthiness of receiving the benefit of the releases. First, we are not called upon to determine whether the Sacklers are worthy of receiving the benefit of the releases. As noted supra, the various equities of the Plan were carefully considered by the bankruptcy court. However, to the extent that there is a fear that this opinion could be read as a blueprint for how individuals can obtain third-party releases in the face of a tsunami of litigation, we caution that the key fact regarding the indemnity agreements at issue is that they were entered into by the end of 2004—well before the contemplation of bankruptcy. Acts taken " 'in contemplation of' bankruptcy ha[ve] long been, and continue[ ] to be, associated with abusive conduct." Milavetz, Gallop & Milavetz, P.A. v. United States, 559 U.S. 229, 240, 130 S.Ct. 1324, 176 L.Ed.2d 79 (2010). We would be far less persuaded if the party seeking to be released entered into this type of indemnity agreement in contemplation of such a third-party release in bankruptcy. Of course, this similar restriction falls in line with our decision in Manville I, where we approved of releases in favor of insurance companies. 837 F.2d at 90. Similarly, in that action, there was no suggestion that the insurance policies were taken out in contemplation of bankruptcy. See id. at 90-91.

As our precedents have suggested, and as we make clear today, if the only reason for the inclusion of a release is the non-debtor's financial contribution to a restructuring plan, then the release is not essential to the bankruptcy. See Manville III, 517 F.3d at 66 (cautioning against this type of situation as abusive). But that is not the present case. Here, the Releases are both needed for the distribution of the res and to ensure the fair distribution of any recovery for claimants. Thus, we deem the scope of the Releases—as limited by the bankruptcy court—appropriate and the Releases essential to the reorganization.

Factor 5. Substantial Contribution to the Reorganization

When evaluating the substantial nature of the released parties' contribution, our primary focus is on the impact of the financial contribution. The bankruptcy court found the financial contribution by the Sacklers, which totaled approximately $4.325 billion, to be substantial and of course did not change its mind when the Sacklers agreed, after the initial approval of the Plan and during the pendency of this appeal, to increase their contribution to make the settlement equal approximately $5.5-6.0 billion. Order Pursuant to 11 U.S.C. §§ 105 and 363(b) Authorizing and Approving Settlement Term Sheet, In re Purdue Pharma L.P., No. 19-23649-shl (Bankr. S.D.N.Y. Mar. 10, 2022), ECF No. 4503. The bankruptcy court stated its belief that this is one of the largest contributions to bankruptcy anywhere in the country. Purdue I, 633 B.R. at 107; cf. In re Mallinckrodt PLC, 639 B.R. 837, 852 (Bankr. D. Del. 2022) (approving of bankruptcy plan with releases where the non-debtor third-party contributed $1.6 billion).

The Trustee primarily argues that the Plan is inequitable because it improperly provides a quid pro quo to the Debtors, and that if the Sacklers had declared bankruptcy, under the Bankruptcy Code they would have had to dedicate substantially all of their net worth (an estimated $11 billion) to the Estate—well more than the approximately $5.5-6.0 billion they have agreed at this point to fund. It is not for this Court to determine whether a greater contribution from the Sacklers would be desirable, but rather our role is simply to decide whether the bankruptcy court erred in finding the Sacklers' contribution substantial. It did not. Five and a half billion dollars—purportedly the largest contribution in history for such releases—is a significant sum.

At oral argument, answering a question from the Court, the Trustee conceded that it would oppose the releases even if the Sacklers contributed $10 billion. Oral Arg. Hr'g at 1:27:45-58.

Factor 6. Overwhelming Approval by Creditors

The claimants voted overwhelmingly to approve the Plan. Over 95% of the personal injury classes voted to accept the plan, which is well above the 75% benchmark. Moreover, with the Nine no longer pursuing their objection, the main challenge to this appeal is not by creditors, but by the Trustee—a government entity without a financial stake in the litigation.

Factor 7. Fair Payment of Enjoined Claims

Finally, the Plan provides for the fair payment of claims. As Appellees concede, the valuation of the claims—estimated at $40 trillion—far exceeds the total funds available, as well as the Sacklers' personal wealth. The bankruptcy court also acknowledged that although "in a vacuum the ultimate judgments that could be achieved on the estates' claims (and the closely related third-party claims that are being settled under the plan) might well be higher than" the Sacklers' contribution to the plan, "the vast size of the claims against the Debtors and the vast number of claimants creates the need for the plan's intricate settlements." Purdue I, 633 B.R. at 93. Thus, as it is not possible to require the full payment of all claims, we do prioritize fair allocation over the full payment of any one claim. The Trustee has not alleged any unequal treatment of claimants, and no party gives us reason to disturb the bankruptcy court's findings that the settlements and allocations were "fair and equitable." Purdue I, 633 B.R. at 84 (internal quotation marks omitted).

* * *

For the reasons stated, the bankruptcy court's detailed findings support approval of the Plan under each of the seven factors that we announce in this opinion. We would also note the additional concessions made by the Sacklers—including governance requirements, abatement trusts, the public document archive, and divestment of the Sacklers from the opioid business worldwide—contribute to the Plan's equity. Purdue I, 633 B.R. at 107. We therefore find no error with the bankruptcy court's weighing of the equitable considerations.

III. Due Process

Although the bankruptcy court found that there was adequate notice to impose the releases, on appeal, the Trustee asserts that the releases in this action did not comply with due process. We, however, find no due process violation.

The district court did not reach this issue.

A procedural due process claim entails a two-part inquiry: whether claimants were deprived of a protected interest and, if so, whether claimants received adequate notice and a meaningful opportunity to be heard. Spinelli v. City of New York, 579 F.3d 160, 168 (2d Cir. 2009). The releases extinguish causes of action, which, as the parties impliedly concede, are a constitutionally protected property interest. See Logan v. Zimmerman Brush Co., 455 U.S. 422, 428, 102 S.Ct. 1148, 71 L.Ed.2d 265 (1982) ("a cause of action is a species of property protected by the Fourteenth Amendment's Due Process Clause"); Rosu v. City of New York, 742 F.3d 523, 526 (2d Cir. 2014) ("[T]he cause of action itself constitutes a cognizable property interest."). Thus, the only remaining question is whether claimants lacked adequate notice or a meaningful opportunity to be heard. Spinelli, 579 F.3d at 168.

In this respect, the Trustee is correct that the Release "permanently extinguish[es] virtually all opioid-related claims against the Sacklers and other non-debtors without the consent of every affected claimant." Trustee Br. at 50. Certainly, that aspect of the Release raises due process concerns—but it does not resolve them. "Once due process is triggered, the question becomes what process is due." In Matter of Motors Liquidation Co., 829 F.3d 135, 158 (2d Cir. 2016). The Trustee's focus on the effect of the Release only gets it so far.

The Trustee argues that there was a denial of due process because the bankruptcy court failed to provide adequate notice of the confirmation hearing and because the language of the Release is dense. "Due process requires notice reasonably calculated . . . to apprise interested parties of the pendency of the action." Burda Media, Inc. v. Viertel, 417 F.3d 292, 303 (2d Cir. 2005) (alteration in original, internal quotation marks omitted). "There is no rigid formula as to the kind of notice that must be given; notice required will vary with circumstances and conditions." Baker v. Latham Sparrowbush Assocs., 72 F.3d 246, 254 (2d Cir. 1995) (internal quotation marks omitted). Here, the bankruptcy court made detailed findings that notice of the confirmation hearing was widespread through a variety of media and that direct notice was provided to any creditors of the Debtors (potential claimants here). The bankruptcy court further observed that although legal training may have been helpful to understanding the initial wording of the releases, the narrowed releases were written more clearly and in "simple . . . plain English." Purdue I, 633 B.R. at 60. The Trustee has given no reason to consider such findings error. See also Mallinckrodt, 639 B.R. at 876-77 (rejecting similar arguments by the Trustee because of the extensive notice, the representation of the victims by a UCC, the lack of a deadline on claims that can access the opioid trusts, and the fact that the court considered those who might not have received or understood notice). Moreover, the bankruptcy court gave process—i.e., meaningful opportunity to be heard—at the confirmation hearing, which lasted for six days.

The Trustee also questions whether such a release, without an ability to opt-out, can comply with due process because it effectively denies claimants their day in court. But, again, the Due Process Clause does not absolutely protect against the deprivation of property; it instead ensures that a deprivation does not occur without due process. In bankruptcy, the sufficiency of process turns on the adequacy of notice and a meaningful opportunity to be heard, both of which, as explained above, occurred here. The Trustee's argument would essentially call into question all releases through bankruptcy, including bankruptcy discharges (which are one of the most important features of bankruptcy). We decline to so undermine such a critical component of bankruptcy. As described supra, the bankruptcy court here acted within its jurisdiction over the bankruptcy estate—even if the third-party claims were not actually the property of the estate—and therefore did not violate due process.

Whatever other constitutional concerns might be raised by the extinguishing of state law claims in bankruptcy, the parties have not argued them here.

* * *

In sum, we reverse the district court's holding that the bankruptcy court lacked the authority to approve the Plan that included the nonconsensual third-party releases. We instead hold that the bankruptcy court properly approved the Plan and made the requisite detailed factual findings to approve of the Shareholder Releases. IV. The Canadian Creditors' Foreign Sovereign Immunity Act Claim

The Canadian Creditors raise various arguments based upon their contention that Section 10.7(b) of the Plan imposes liability personal to the Canadian Creditors in a manner that violates their sovereign immunity.

As a threshold matter, it is not clear that sovereign immunity is even implicated by the releases. To the contrary, at least in the context of discharging claims against a debtor, "[a] debtor does not seek monetary damages or any affirmative relief from a State by seeking to discharge a debt; nor does he subject an unwilling State to a coercive judicial process. He seeks only a discharge of his debts." Tenn. Student Assistance Corp. v. Hood, 541 U.S. 440, 450, 124 S.Ct. 1905, 158 L.Ed.2d 764 (2004). The releases here do not require a suit to be maintained against the Canadian Creditors. Nor do they seek to impose personal liability on the Canadian Creditors. The Canadian Creditors also cannot be described as unwilling with regard to this judicial process, in which they have fully and voluntarily participated. S.G. Phillips Constructors, Inc. v. City of Burlington (In re S.G. Phillips Constructors, Inc.), 45 F.3d 702, 707 (2d Cir. 1995) ("The Supreme Court and this court have consistently held that in filing a proof of claim the petitioner submits to the bankruptcy court's equitable jurisdiction."). Moreover, the Foreign Sovereign Immunities Act also does not protect the Canadian municipalities because 28 U.S.C. § 1605(a)(1) provides that a foreign state will not be immune from jurisdiction of the courts where the foreign state has waived its immunity either explicitly or by implication. For these reasons, we find that the Plan does not violate the sovereign immunity of the Canadian Creditors.

V. The Cross-Appeal

The bankruptcy court and the district court both determined that the Plan properly differentiated the Canadian objectors' claims from their domestic counterparts. The Canadian Creditors contend it is inequitable that they do not have access to the abatement trusts, but domestic creditors do. Thus, in their view, because § 1129(a)(1) requires equal treatment, the Plan fails. We do not find those arguments persuasive and affirm the district court.

Section 1123(a)(1) provides that "[n]otwithstanding any otherwise applicable non-bankruptcy law, a plan shall designate, subject to section 1122 of this title, classes of claims." 11 U.S.C. § 1123(a)(1). Under 11 U.S.C. § 1122(a), plans may classify claims in a particular class so long as those claims are "substantially similar to the other claims or interests of such class." Yet, the statute itself "does not explicitly address whether similar claims must be placed in the same class." Boston Post Rd. Ltd. P'Ship v. FDIC (In re Boston Post Rd. Ltd. P'ship), 21 F.3d 477, 481 (2d Cir. 1994). Looking to other circuits, which "have generally held that separate classification of similar claims is permissible only upon proof of a legitimate reason for separate classification, and that separate classification to gerrymander an affirmative vote is impermissible," id., this Court has held that "similar claims may not be separately classified solely to engineer an assenting impaired class," id. at 482. Instead, the separation of similar claims can only be justified by a legitimate reason. Id. at 483; see also In re W.R. Grace & Co., 729 F.3d 311, 329 (3d Cir. 2013) (ruling that the separate classification of Canadian claims is appropriate because the "Canadian and U.S. property damage claimants . . . operate under separate tort regimes[ ] and reached separate settlement agreements"); Dow Corning, 280 F.3d at 662 (approving the separate classification of foreign claims because "the bankruptcy court determined that the evidence supported the factual assumptions upon which the classifications are based," including clear expert witness testimony about tort recovery in other nations). Here, both the bankruptcy court and the district court found that the claims were properly differentiated in the Plan because the claims are subject to different regulatory regimes that result in different types of recovery and the Canadian creditors did not participate in the mediation allocation. Purdue I, 633 B.R. at 70; Purdue II, 635 B.R. at 117.

The Cross-Appellants argue regulatory differences do not suffice to account for the different classification. However, we see no reason to disturb the conclusions of the bankruptcy court and the district court. There are substantive differences among the claims, including both the types of claims and elements of causes of action. Moreover, the Canadian objectors have another source of recovery: Purdue Canada. We believe those reasons alone provide enough support to differentiate the claims, and thus to affirm the district court's holding on the cross-appeal.

Of note, Purdue Canada reached a separate settlement of $150 million. See Settlement reached with Purdue Pharma (Canada) for opioid damages, British Columbia Government News (June 29, 2022), https://news.gov.bc.ca/releases/2022AG0044-001031.

CONCLUSION

For the reasons set forth above, we REVERSE the district court's order holding that the Bankruptcy Code does not permit nonconsensual third-party releases of direct claims, and AFFIRM the bankruptcy court's approval of the Plan, including the modification made on March 10, 2022, and the case is REMANDED to the district court for such further proceedings as may be required, consistent with this opinion. We also AFFIRM the district court's denial of the Canadian Creditors' cross-appeal.

Judge Wesley concurs in the judgment in a separate opinion.

Richard C. Wesley, Circuit Judge, concurring in the judgment:

Does a bankruptcy court have the power to release direct or particularized claims asserted by third parties against nondebtors without the third parties' consent? Yes—this Court said so in In re Drexel Burnham Lambert Grp., Inc., 960 F.2d 285, 293 (2d Cir. 1992). Drexel has not been overruled either by the Supreme Court or by this Court sitting en banc. It is binding. Consequently, although the parties have sacrificed a forest on the matter—and rightly so, weighty as it is—that ship has, for better or worse, sailed. I therefore reluctantly concur with the majority's conclusion that a bankruptcy court has the authority to approve a Chapter 11 reorganization plan that includes nonconsensual nondebtor releases. Again: Drexel says so.

That said, neither Drexel, nor our subsequent discussion of nonconsensual non-debtor releases in Metromedia, traces that power back to any provision of the Bankruptcy Code. See In re Metromedia Fiber Network, Inc., 416 F.3d 136, 142 (2d Cir. 2005). In fact, although Metromedia acknowledged that Drexel had already crossed the bridge, it also appreciated its questionable structure, and was wary to traverse it once more. To the point, Judge Jacobs carefully explained "the reluctance to approve nondebtor releases," and cautioned that nowhere—apart from asbestos-related bankruptcies—does the Code authorize them. See id. The majority concedes as much; it recognizes that "whatever equitable powers remain in the bankruptcy courts must and can only be exercised within the confines of the Bankruptcy Code." Law v. Siegel, 571 U.S. 415, 421, 134 S.Ct. 1188, 188 L.Ed.2d 146 (2014). Today, it fills that gap with §§ 105 and 1123(b)(6).

Those provisions of the Bankruptcy Code say nothing about nondebtor releases, and I am not convinced that statutory footing is up to the task. Accordingly, although mindful that, for this Court, the issue has already been settled (albeit without any basis in the Code), I write separately to highlight my concerns.

Those concerns are, in brief: extinguishing direct, particularized claims against nondebtors without the claimholder's consent, and without compensating the claimholder, is an extraordinarily powerful tool for a bankruptcy court to wield—indeed, for any court to wield. Congress once before provided clarity on the propriety of third-party releases in bankruptcy. It could do so again, but, since 1994, has not. Absent any movement on that front, the question, which has divided the courts of appeals for decades, would benefit from nationwide resolution by the Supreme Court. In that event, a uniform view of the problem would emerge.

I

The majority's overview of the facts, procedural history, and opinions below, is thorough and well stated. For present purposes, it is sufficient to emphasize exactly what the Shareholder Release purports to do.

Defined terms here coincide with the those in the majority opinion.

Prior to Purdue's Chapter 11 filing, widespread efforts to hold Purdue legally accountable for its role in the opioid epidemic eventually revealed, at least in the eyes of countless plaintiffs, that certain members of the Sackler family were heavily involved with unlawful efforts to boost Purdue's opioid sales. See In re Purdue Pharma, L.P. ("Purdue II"), 635 B.R. 26, 50-51 (S.D.N.Y. 2021). Seeking to hold the Sackler family members directly liable for their part in perpetuating the opioid epidemic, both private litigants as well as state Attorneys General turned to various state statutes, including consumer protection laws, which, notwithstanding considerable factual overlap with allegations of corporate liability, impose a separate and independent duty on individuals who, by virtue of their role as either officer, manager, or director of a corporation, personally participated in corporate wrongdoing. See id. As Judge McMahon aptly put it:

[I]t is undisputed that these laws impose liability, and even penalties, on such persons independent of any corporate liability (or lack of same), and independent of any claim the corporation could assert against them for faithless service as a result of those same acts.

Id. at 91. These claims "arise out of the Sacklers' own conduct." Id.

The Shareholder Release forever halts those proceedings in their tracks. It permanently enjoins the private and state litigants, as well as all future plaintiffs, from pursuing those claims against the Sacklers—indeed, any claim "of any kind, character[,] or nature whatsoever, Special App'x 798—so long as the Debtors' "conduct, omission, or liability" is "the legal cause or is otherwise a legally relevant factor."Id. at 920. No carveout exists for claims based on fraud—claims from which a debtor could not seek a discharge under the Code. See 11 U.S.C. § 523(a)(2)(A); see also Archer v. Warner, 538 U.S. 314, 321, 123 S.Ct. 1462, 155 L.Ed.2d 454 (2003) ("[The Code] ensure[s] that all debts arising out of fraud are excepted from discharge no matter their form." (quotation omitted)). Appellants seek a release broader than that which Congress decided was wise to make available to a debtor in bankruptcy.

The limiting effect of the "legally relevant" requirement is elusive, and its precise reach has, understandably, not been articulated either by the parties, the bankruptcy court, or the majority. Their failure to do so is no fault of their own; it is difficult to predict the various claims which might be asserted directly against the Sacklers, and future litigation will determine whether any given claim falls within the provision. Still, one can envision an exceedingly broad understanding of "legal relevance," and I, for one, am skeptical of the requirement's limiting effect. To illustrate, at issue in Manville III were direct claims against Manville's primary insurer alleging that the insurer violated purported state-imposed duties to disclose certain asbestos-related information it learned from Manville. See In re Johns-Manville Corp. ("Manville III"), 517 F.3d 52, 66 (2d Cir. 2008) rev'd and remanded on other grounds sub nom. Travelers Indem. Co. v. Bailey, 557 U.S. 137, 129 S.Ct. 2195, 174 L.Ed.2d 99 (2009). We held that notwithstanding the factual overlap of those claims with claims which might be asserted against Manville, or by Manville against the insurer, the bankruptcy court was without jurisdiction to release the direct claims against the insurer. See id. As to those direct claims, Manville's "conduct" or "omission" might be described as legally relevant: they were based on what the insurer learned from Manville. I am concerned that "legal relevance" might release claims mirroring those which we have previously held did not fall within bankruptcy jurisdiction.

On top of that, the Release does not "channel[ ]" the enjoined claims "to a settlement fund" for compensation, Metromedia, 416 F.3d at 142, but instead mandates that any value paid to personal injury claimants regarding, for example, the opioid-related death of another person, be based only upon claims "held against the Debtors, and not to any associated . . . Channeled Claim against a non-Debtor party." Special App'x 634, 693, 734-35. In other words, the value of a channeled claim is only the value of claims against the estate.

Consider this example. Someone has a claim against only Purdue and it's worth $100,000. They file a proof of claim and receive a check for some percentage of that claim. Another person has the same claim for the same amount, and a direct claim against the Sacklers worth another $100,000. Under the Plan, that party receives only the same amount as the first claimant; they receive no payout on their direct claim against the Sacklers, even though the Sacklers are released from that claim.

This aspect of the Release substantially broadens its reach as compared with the release approved in Manville I. See MacArthur Co. v. Johns-Manville Corp. ("Manville I"), 837 F.2d 89 (2d Cir. 1988). There, we rejected the notion that a release constituted a bankruptcy discharge because the released claims were not extinguished, but were "channeled away from the insurers and redirected at the proceeds of the settlement." Id. at 91. Here, the Plan expressly disallows value being paid based on claims against non-debtors, that is, the Sacklers.Manville I therefore does not lay the groundwork for the Release's approval.

Appellants dispute that characterization; they point to the Plan's language that any distribution "is deemed to be a distribution in satisfaction of all [personal injury] Channeled Claims," and argue that payments from the personal injury trust is in satisfaction of claims both against the Debtors and Sacklers. Mortimer Side Br. at 49 (quoting Special App'x 693). Yet simply stating as much does not make it so where, as here, the amount of distribution is based only upon claims as against the Debtors.

Finally, the Release is non-consensual; it binds consenting and objecting parties, without providing an opt-out option to those who object.

In summary, the Release enjoins a broader swath of claims than a debtor himself could seek to discharge under the Bankruptcy Code, and it does so without providing any compensation to the claimholders, who must abide by its terms whether they like it or not. The Release encompasses a potentially wide range of claims and cloaks the Sacklers with blanket immunity. It is "in effect . . . a [ ] discharge." Metromedia, 416 F.3d at 142.

In exchange, the Sacklers have agreed, under the Plan, to offer a substantial sum of money to the Debtors' estate. No doubt, those funds help make possible (a) a more meaningful distribution of the Debtors' estate to its creditors and (b) recovery for those who hold claims against the Debtors. It is equally apparent that the Sacklers mean what they've said: no release, no money. However, our task today is not to decide whether, as a policy matter, the Release is justified. Instead, without ignoring that the Sacklers' substantial contribution will likely play a meaningful role in providing some measure of finality to the countless families who have suffered as a result of the opioid crisis, the dispositive question is whether, under the Bankruptcy Code, a bankruptcy court is authorized to approve the Release.

Again, however, their contribution is not directed at the satisfaction of third parties' direct claims against them in their individual capacity, but, instead, at the satisfaction of either claims against the Debtors, or claims held by the estate against the Sacklers. As to the latter set of claims, the Sacklers have, in essence, settled derivative claims belonging to the estate and, in return, received a release not just from those derivative claims, but also from claims independently held by third parties.

Of course, the majority correctly recognizes that the antecedent question to the statutory authority analysis is whether the bankruptcy court had jurisdiction under the Code to approve the Release. I do not dispute its conclusion that it did; it is settled law in this Circuit that a bankruptcy court has broad "related to" jurisdiction over any civil proceedings that "might have any conceivable effect" on the estate. See SPV OSUS Ltd. v. UBS AG, 882 F.3d 333, 339-40 (2d Cir. 2018). In the easy case, that effect can be direct, as it was in Manville I. There, the claims asserted against the insurer sought recovery from the res itself. See Manville I, 837 F.2d at 93. The bankruptcy court had jurisdiction to prevent the third party from "collect[ing] out of the proceeds of Manville's insurance policies . . . ." Id. In the harder case, the effect is less direct. In SPV, for example, the plaintiffs asserted direct claims against, among other defendants, UBS AG, alleging principally that UBS had aided and abetted the infamous fraud perpetrated by the debtor, Bernard L. Madoff Investment Securities LLC. See SPV, 882 F.3d at 338. Although the plaintiffs sought recovery from UBS itself, UBS, in turn, had viable claims for indemnification and contribution against the debtor. See id. at 340-42. The possibility that those claims might succeed—and the fact that the debtor would incur expense in litigating such claims—was enough to confer jurisdiction on the bankruptcy court to enjoin the plaintiff's direct claims against UBS. See id. at 341-42. This case looks more like SPV, and the majority's explanation as to how the direct claims against the Sacklers might affect the Debtors' estate is sound.

II

The Bankruptcy Code is silent on the matter. That is no surprise. Bankruptcy is the "subject of the relations between a[ ] . . . debtor[ ] and his creditors, extending to his and their relief." Wright v. Union Cent. Life Ins. Co., 304 U.S. 502, 513-14, 58 S.Ct. 1025, 82 L.Ed. 1490 (1938). To that end, Congress created a "comprehensive federal system . . . to govern the orderly conduct of debtors' affairs and creditors' rights." Eastern Equip. & Servs. Corp. v. Factory Point Nat'l Bank, 236 F.3d 117, 120 (2d Cir. 2001). In short, the Bankruptcy Code's central focus is on the adjustment of the debtor-creditor relationship. Of course, that adjustment can implicate the interests of third-party non-debtors. But as to their own independent obligations, third-party nondebtors are, simply, a nonconcern.

For example, the automatic stay triggered by a debtor's bankruptcy filing can apply to nondebtors in certain circumstances. See, e.g., Queenie, Ltd. v. Nygard Int'l, 321 F.3d 282, 287 (2d Cir. 2003).

Against that backdrop, there is little to glean from Congressional silence where, as Judge McMahon put it, "one would not expect Congress to speak." Purdue II, 635 B.R. at 110. Appellants ask us to accept the remarkable premise that Congress, while believing it wise to except certain claims (i.e., claims for fraud) from a debtor's discharge, took no issue with the idea that such claims could be effectively discharged for nondebtors, who might contribute funds to settle claims against the debtor, but who would face no consequences from their own, independent liability—even though state laws mandate otherwise. Not only that, appellants ask us to ground this grant of authority in congressional silence, as, again, the Bankruptcy Code does not expressly authorize the practice.

And yet that silence is, effectively, what the majority sees as granting the bankruptcy court a power that is nothing short of extraordinary. It points to 11 U.S.C. § 1123(b)(6), which it says encompasses a bankruptcy court's residual equitable authority, and empowers a bankruptcy court to do all but that which the Code expressly forbids. Maj. Op. at 73-74.

The majority recognizes that the Release cannot be justified solely by § 105. See Metromedia, 416 F.3d at 142 ("Any power that a judge enjoys under § 105 must derive ultimately from some other provision of the Bankruptcy Code." (internal citation omitted)). In other words, the Release turns on § 1123(b)(6). I focus my analysis there.

To be sure, the Court in Energy Resources characterized § 1123(b)(6) as Congress's recognition of a bankruptcy court's residual equitable authority. But it did so in connection with its observation that "bankruptcy courts, as courts of equity, have broad authority to modify creditor-debtor relationships." United States v. Energy Resources Co., Inc., 495 U.S. 545, 549, 110 S.Ct. 2139, 109 L.Ed.2d 580 (1990) (emphasis added). That case concerned whether a debtor's tax liabilities could be satisfied in an order as determined by the bankruptcy court, over the objection of the Internal Revenue Service. Nothing in Energy Resources suggests that within § 1123(b)(6)'s equitable repository is the power to extinguish an individual's claims against a nondebtor without their consent, and without providing them any value in return. Indeed, that case says nothing about a nondebtor's obligations under the Bankruptcy Code whatsoever.

Instead, Energy Resources reminds us that bankruptcy courts are courts of equity, and that their ability to carry out the Code's provisions must be understood with that principle in mind. But it does not answer whether under that umbrella of equitable authority exists the power to release, on a nonconsensual basis, nondebtors from direct claims held by third parties. Nor does Energy Resources suggest that a bankruptcy court's well of residual equitable authority, so long as it does not run up against a more specific provision of the Code, is bottomless.

Again: that case concerned the adjustment of a creditor-debtor relationship, which, as provided above, is a bankruptcy court's raison d'etre. Courts should understand any congressional grant of equitable authority to the bankruptcy court with that principal purpose in mind. Releasing nondebtors from their own liability—provided for under state law—over the objection of a claimholder and without compensating that claimholder is so far afield from that purpose that plugging-and-playing Energy Resources' description of § 1123(b)(6) can't be right.

The decisions from our sister circuits cited by the majority are no more persuasive. Those decisions also rely on Energy Resources' characterization of § 1123(b)(6). See, e.g., In re Airadigm Commc'ns, Inc., 519 F.3d 640, 657 (7th Cir. 2008). In any event, in Airadigm itself, the release did not cover, as the Release here does, claims for willful misconduct, a fact emphasized by the Seventh Circuit as justifying its confirmation. See id. That case does not signal a green light to the approval of the Shareholder Release. Neither does In re Dow Corning Corp., 280 F.3d 648 (6th Cir. 2002). There, the Sixth Circuit rejected the third-party release because it did not provide an opportunity for objecting claimholders to recover in full. See id. at 659-61.

Moreover, the Court has, in other contexts, explained that a bankruptcy court's equitable authority is not "unlimited," but instead incorporates "traditional standards in equity practice," and that courts can look to "cases outside the bankruptcy context" to help understand the limits of that authority. Taggart v. Lorenzen, — U.S. —, 139 S. Ct. 1795, 1801-02, 204 L.Ed.2d 129 (2019). The majority does not liken the equitable authority recognized today to anything traditionally recognized at equity. I too am at a loss. Indeed, the idea that bankruptcy courts can order the involuntary release of direct claims against non-debtors is "an extraordinary thing" that is "different . . . from what courts ordinarily do." In re Aegean Marine Petroleum Network Inc., 599 B.R. 717, 723 (Bankr. S.D.N.Y. 2019).

In Taggart, the Court drew on traditional equitable standards for civil contempt sanctions outside the bankruptcy context to define a bankruptcy court's authority to hold a party in civil contempt for violating § 523(a)(2)'s discharge injunction. If, as the majority and appellants would have us believe, a bankruptcy court's ability to enforce its injunction were limited only by that which the Code did not forbid, then Taggart's invocation of traditional civil contempt standards would seem misplaced.

At bottom, if Congress intended so extraordinary a grant of authority, it should say so. See Czyzewski v. Jevic Holding Corp., 580 U.S. 451, 465, 137 S.Ct. 973, 197 L.Ed.2d 398 (2017) (requiring "more than simple statutory silence if, and when, Congress were to intend a major departure" from the Code). It has before; in 1994, it amended the Bankruptcy Code to provide express authorization for nondebtor releases in asbestos-related bankruptcies, subject to a stringent set of requirements. See 11 U.S.C. 524(g). That amendment occurred when, at that time, courts, such as in Drexel, were then approving nondebtor releases in non-asbestos bankruptcies. Yet Congress endorsed nondebtor releases in only the asbestos context. The parties debate whether Congress' express but limited approval in § 524(g) was an implicit rejection of nondebtor releases in non-asbestos contexts. The majority says no. Regardless of the right answer, the majority's answer pins this Circuit firmly on one side of a weighty issue that, for too long, has split the courts of appeals.

Even there, however, the injunction may extend to nondebtors only where the nondebtor is "directly liable or indirectly liable for the conduct of, claims against, or demands on the debtor . . . ." 11 U.S.C. § 524(g)(4)(A)(ii). The Release here is broader; it covers claims aimed at the Sacklers' liability even if it is independent from the Debtors' liability. Even under § 524(g), it's far from clear the Release would survive.

This difference in views has consequences. As it stands, a nondebtor's ability to be released through bankruptcy turns on where a debtor files. Forum-dependent results are anathema to the establishment of "uniform Laws on the subject of Bankruptcies throughout the United States." U.S. Const. art. I, § 8, cl. 4. Finding implicit grants of extraordinary powers in congressional silence is at cross purposes with the Code's "comprehensive scheme." RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 566 U.S. 639, 645, 132 S.Ct. 2065, 182 L.Ed.2d 967 (2012). Absent direction from Congress—and, since 1994, there has been none—or the High Court, the answer is a function of geography.

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Summary

Opinion

BACKGROUND

I. Factual Background

II. Procedural History

1. Equitable Considerations

2. The Authority of the Bankruptcy Court to Release Third-Party Claims Against the Sacklers

3. Canadian Creditors' Objections

4. Pro SeObjections

1. Subject-Matter Jurisdiction

2. Statutory Power to Release Third-Party Claims

3. Classification of Canadian Claims

DISCUSSION

I. Standard of Review

II. Nonconsensual Third-Party Releases of Direct Claims

1. Statutory Authority

2. Second Circuit Case Law

Factor 1. Identity of Interests Between Debtors and Released Parties

Factor 2. Factual and Legal Overlap Between Claims Against Debtors and Settled Third-Party Claims

Factors 3. and 4. The Releases are Essential to the Reorganization & Proper in Scope

Factor 5. Substantial Contribution to the Reorganization

Factor 6. Overwhelming Approval by Creditors

Factor 7. Fair Payment of Enjoined Claims

* * *

III. Due Process

* * *

V. The Cross-Appeal

CONCLUSION

I

II

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Case Details

Citations

Citing Cases

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Summary

Opinion

BACKGROUND

I. Factual Background

II. Procedural History

1. Equitable Considerations

2. The Authority of the Bankruptcy Court to Release Third-Party Claims Against the Sacklers

3. Canadian Creditors' Objections

4. Pro SeObjections

1. Subject-Matter Jurisdiction

2. Statutory Power to Release Third-Party Claims

3. Classification of Canadian Claims

DISCUSSION

I. Standard of Review

II. Nonconsensual Third-Party Releases of Direct Claims

1. Statutory Authority

2. Second Circuit Case Law

Factor 1. Identity of Interests Between Debtors and Released Parties

Factor 2. Factual and Legal Overlap Between Claims Against Debtors and Settled Third-Party Claims

Factors 3. and 4. The Releases are Essential to the Reorganization & Proper in Scope

Factor 5. Substantial Contribution to the Reorganization

Factor 6. Overwhelming Approval by Creditors

Factor 7. Fair Payment of Enjoined Claims

* * *

III. Due Process

* * *

V. The Cross-Appeal

CONCLUSION

I

II

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Purdue Pharma. v. The City of Grande Prairie (In re Purdue Pharma.)

Case Details

CitationsCopy Citation

Citing Cases

Citations