Lindsay Cooper v. Tokyo Elec. Power Co. Holdings ( 2020 )


Menu:
  •                FOR PUBLICATION
    UNITED STATES COURT OF APPEALS
    FOR THE NINTH CIRCUIT
    LINDSAY R. COOPER; JAMES R.         No. 19-55295
    SUTTON; KIM GIESEKING; CHARLES
    A. YARRIS; ROBERT M. MILLER;           D.C. No.
    CHRISTOPHER G. BITTNER; ERIC        3:12-cv-03032-
    MEMBRILA; JUDY GOODWIN;                JLS-MSB
    JENNIFER L. MICKE; JOHN W.
    SEELBACH; MAURICE D. ENIS; JAIME
    L. PLYM; NATHAN J. PIEKUTOWSKI;       OPINION
    CAROLYN A. WHITE; LOUIE
    VIERNES; MICHAEL L. SEBOURN;
    K.S., an infant by his father and
    natural guardian Michael L.
    Sebourn; CHRISTIAN M. EBUENG;
    PAUL J. ENCINIAS; DANIEL E. HAIR;
    ADAM W. KRUTZLER; DAVID K.
    MALONE; ROBERT SELIGMAN; ELOI
    A. WHITEMAN; JASON D. HENRY;
    NELLIE ALLEN-LOGAN; JAMI
    BESCHORNER; NATHAN CANCHE;
    NATHAN CRISWELL; JASON TROY
    FRIEL; OSCAR GONZALEZ; DAVID
    HAHN; JAMES JACKSON; JARRETT
    BRADY JOHNSTON; JONATHAN
    MEDINA; ADAM MINTZ; MALLORY
    K. MORROW; WILLIAM NETHERTON;
    MICHELLE ODEN; DONALD RAIRIGH;
    CHRISTOPHER RICKARD; ANDREW
    RIVERA; STEVEN RAY SIMMONS;
    AKEEM SMITH; JUSTIN SPENCER;
    2        COOPER V. TOKYO ELEC. POWER CO.
    ALAN SPURLING; ANGEL TORRES;
    ANTHONY GARCIA; JASMINE ALLEN;
    RHONDA ANBERT; SUSAN ASH;
    ADAM ARMENTA; JINKY M.A.,
    individually and as the
    Administrator of the Estate of
    Charliemagne T.A.; J.C.A., a minor
    by his mother as guardian ad litem
    Jinky M.A.; J.A., a minor by his
    mother as guardian ad litem Jinky
    M.A.; DANA AUSTIN; RENAR AWA;
    JOSH BANE; ARAMIS BARRIOS;
    TREVOR BECK; MARKUS BEGAY;
    JORDAN BENOIT; JORDAN
    BETTENCOURT; BRETT A. BINGHAM;
    GUNNAR BORTHICK; KENNETH CLEO
    BOSWELL; JAMES P. BOWEN;
    MATTHEW BRADLEY; NICOLAS
    BREWTON; NICOLAUS BROOKS;
    RYAN S. BROWN; CASEY
    BRUCKLACHER; REBECCA BRUNET;
    GERARDO BRUING; ROBIN
    CALCATERRA; ROBBY CANLAS;
    CARLISI; COURTNEY CARMICHAEL;
    MATTHEW CARTWRIGHT; WAYNE
    CASSAR; FABIAN CERVANTES;
    MELVIN A. CHAMBERLAIN;
    TERANCE CHAPMAN; WILLIAM
    CHAPMAN, JR.; ANNMARIE
    CHESSARI; DAVID CHITWOOD;
    GEORGE COBB; LORI LYNN CODY;
    KEONDICE W. COOK; ANGELA
    CRABTREE; CHAD CROFT; BRIAN
    COOPER V. TOKYO ELEC. POWER CO.   3
    CROSS; NICOLAS CROUCH; THOMAS
    CULBERSON; VICENT CURCI; HONDA
    DAGAN; JAMES DARNELL; JANELLE
    DARNELL; JASON DASILVA; JOHN
    DAVIS; MARK DECASA; NICHOLE M.
    DECATUR; MARTIN DELGARDILLO;
    TINA DIBERNARDO; BRANDON
    DOCKERY; J. D., a minor by his
    father as guardian ad litem Jeremy
    D.; JEREMY D.; CHRISTIAN DOERR;
    IAN W. DOVE; JESSE DUNN;
    CHRISTINA DUVALL; CHRISTIAN JOY
    NAGUI EBUENG; ANGEL ESCRIBANO;
    SETH ESLIN; NICHOLAS J. FELLER;
    KYLE E. FELT; TERI FORZA; JOEL
    FUDGE; PAUL GABBY; SHANE
    GALLAGHER; ZACH GARNER; JOHN
    OLIVER GOOCH IV; KATE GRACE;
    JENNIFER GUANA; M. H., a minor by
    Allison D. Eyring her mother as
    guardian ad litem; ANDREW HAJNY;
    ROBERT HAREWOOD; DANIEL
    PATRICK HARREN; JOSHUA
    HARRIGAN; CHAD HARRIS; ROBERT
    C. HARTAGE; TIFFANY HARTMAN;
    NICHOLAS HELMSTADT; ASHTON
    HEMPHILL; ERIN HERRING; CORA E.
    HILL; CHAD HOLT; NEIL HOPKINS;
    DYLAN IMGRAM; NICK INCA;
    JEDIDIAH IRONS; GERARDO IRVING;
    BLAKE ISAACS; THE ESTATE OF
    THEODORE H., by Manuel Leslie as
    the administrator of the estate of
    4       COOPER V. TOKYO ELEC. POWER CO.
    Theodore H; JOANNA ILOILO;
    DARIUS JACKSON; JESSICA JACKSON;
    CHRISTIAN A. JESSUP; WILLIAM
    JONES; WINSTON JONES; LEON
    JULIAN; CHARLES D. KAISER;
    DANIEL KREGSTEIN; ZACKARY
    KUBE; SHANE M. LANGNES; DANIEL
    LAWVIER; ROBERT LEHRMAN;
    JULIAN LEON; MARY LOKKA;
    NICOLE LOOK FANG; ALYSSA LOPEZ;
    ZACKERY LOUVERS; CHRISTOPHER
    LOWE; ALEJANDRO LUSK; CORA
    MAE; BILLY MARKHAM; ALEX
    MATIN; ALFRED MCALLISTER;
    DIANNA MCCANTS; THOMAS
    MCCANTS; CHENEIL MCCARTER;
    TYLER MCDONALD; PETINA
    MCINTOSH; RYAN MENENDEZ;
    MICHAEL MESIGH; SAMY MOHANIE;
    JOEL MONSALUD; LETICIA
    MORALES; KEVIN MORRIS; COLIN
    MORRISON; TIMOTHY MUIS; JON
    NEUMANN; MARK NEWMAN; DANIEL
    OLSEN, Officer; CHAD YARBROUGH;
    ANTHONY J. YOVANOVIC;
    JONATHAN ZAVITZ; WILLIAM
    ZELLER; MICHAEL ZITELLA; MIKE
    TISOY ORMAN; CHRISTOPHER
    PETERSON; MATTHEW PETERSON;
    ALYSSA PETTERWAY; KEITH
    POTTRATZ; DANIEL PRETTO; ASHLEY
    RAMIREZ; TYLER RAY RANDRUP;
    SUSAN RODRIQUEZ; BRANDEN
    COOPER V. TOKYO ELEC. POWER CO.   5
    RUCKER; W. RUSHBY; ERICA RYAN;
    DAVID SANCHEZ; DANE SANTO;
    DAISY M. SARSLOW; ROBERT
    SEELIGMAN; BENITO G. SERENTAS,
    JR.; KELLI SERIO; CHRISTOPHER
    SHAMRELL; MICHAEL B. SHANNON;
    RYAN SIVELS; JOSHUA C. SMILEY;
    BRANDON SMITH; FRANCES FISTER
    STOGA; BYRON SY; KELLY
    TANNEHILL; NIGEL THOMPSON;
    CHAD THORTON; MICHAEL TIMKO;
    PATRICIA TOTEMEIER; JAMES TRICE;
    DARREL USRY; MARK VALDEZ;
    OSVALDO VERA; KAILEE VICTRUM;
    ANDREW VODOPIJA; ANDREW
    VROOMAN; SKYLER STEVEN
    WARNOCK; TAWNY WATSON;
    TIMOTHY J. WENDAL; IAN LEE
    WHEATON; TIM WHITE; EDWARD
    JOSEPH WICKLE; PATRICK WIGHT;
    KRISTIAN WILLIAM; TIM WOELKY;
    JUSTIN WOMMACK; RONALD
    WRIGHT; KIOCHI YAMAZAKI; A. G.,
    an infant by her mother and natural
    guardian Kim Gieseking,
    Plaintiffs-Appellants,
    6          COOPER V. TOKYO ELEC. POWER CO.
    v.
    TOKYO ELECTRIC POWER COMPANY
    HOLDINGS, INC., AKA TEPCO;
    GENERAL ELECTRIC COMPANY;
    DOES, 1–200, inclusive,
    Defendants-Appellees.
    Appeal from the United States District Court
    for the Southern District of California
    Janis L. Sammartino, District Judge, Presiding
    Argued and Submitted March 10, 2020
    Pasadena, California
    Filed May 22, 2020
    Before: A. Wallace Tashima, Kim McLane Wardlaw,
    and Jay S. Bybee, Circuit Judges.
    Opinion by Judge Bybee
    COOPER V. TOKYO ELEC. POWER CO.                           7
    SUMMARY*
    International Comity / Choice-of-Law
    The panel affirmed the district court’s dismissals of
    claims brought in California by United States
    servicemembers against Tokyo Electric Power Company
    (“TEPCO”) and General Electric Company (“GE”), alleging
    they were exposed to radiation from the Fukushima Daiichi
    Nuclear Power Plant.
    The Japanese Act on Compensation for Nuclear Damage
    provides that the operator of a nuclear power plant is strictly
    liable for any damage caused by the operation of the power
    plant but no other person shall be liable.
    GE argued that Japanese law should apply, and that under
    Japanese law only the plant operator, TEPCO, could be liable
    for injuries resulting from the power plant’s failure. The
    panel held that Japan’s Compensation Act was a liability-
    limiting statute with outcome-determinative implications, and
    was substantive for Erie purposes, and subject to a choice-of-
    law analysis. The panel also held that the district court did not
    err by conducting a choice-of-law analysis at the motion-to-
    dismiss stage of the litigation.
    The panel applied California’s three-step “governmental
    interest” test in deciding the choice-of-law questions. First,
    the panel held that the laws of California and Japan differed
    because under Japanese law, the Compensation Act would
    *
    This summary constitutes no part of the opinion of the court. It has
    been prepared by court staff for the convenience of the reader.
    8           COOPER V. TOKYO ELEC. POWER CO.
    limit liability to TEPCO and require dismissal of all claims
    against GE; but under California law, GE, as manufacturer,
    would be strictly liable for design defects. Second, the panel
    held that both jurisdictions had a legitimate interest. In
    addition, the panel concluded, as did the district court, that
    there was a “true conflict” where California had an interest in
    holding manufacturers of defective products liable in tort to
    insure compensation for its residents, while Japan had an
    interest in consistent application of its liability-limiting
    statute to businesses participating in its nuclear industry.
    Third, the panel held that Japan’s interests would be more
    impaired if its law were not applied. Because there was no
    dispute on appeal that application of Japanese law required
    dismissal of all claims against GE, the panel affirmed the
    dismissal of those claims with prejudice.
    TEPCO argued for dismissal on international-comity
    grounds. Plaintiffs raised the same challenges to the choice-
    of-law analysis for TEPCO’s claims. The panel held that for
    the reasons previously stated, the choice-of-law analysis was
    not premature or inappropriate at this stage. As to the merits
    of the choice-of-law analysis, the district court correctly
    found that Japanese law applied to the plaintiffs’ claims
    against TEPCO.
    The panel proceeded to the question of whether, given
    that Japanese law must apply to the proceedings in the
    Southern District of California, the district court abused its
    discretion in dismissing the case on international comity
    grounds. In deciding whether the doctrine of adjudicative
    comity applied, courts weigh several factors. The panel held
    that it was not an abuse of discretion for the district court to
    take the applicability of Japanese law into consideration. The
    panel also held that the district court did not abuse its
    COOPER V. TOKYO ELEC. POWER CO.                   9
    discretion in considering Japan’s strong interests in the case
    being litigated in Japan. The panel concluded that the district
    court did not abuse its discretion when it dismissed the claims
    against TEPCO on international comity grounds.
    COUNSEL
    A. Cabral Bonner (argued) and Charles A. Bonner, Law
    Offices of Bonner & Bonner, Sausalito, California; John R.
    Edwards (argued) and Catharine E. Edwards, Edwards Kirby
    LLP, Raleigh, North Carolina; Paul C. Garner, Rancho
    Mirage, California; for Plaintiffs-Appellants.
    Mark R. Yohalem (argued) and Gregory P. Stone, Munger
    Tolles & Olson LLP, Los Angeles, California, for Defendant-
    Appellee Tokyo Electric Power Company Holdings, Inc.
    David J. Weiner (argued) and Sally L. Pei, Arnold & Porter
    Kaye Scholer LLP, Washington, D.C.; Michael D. Schissel,
    Arnold & Porter Kaye Scholer LLP, New York, New York;
    for Defendant-Appellee General Electric Company.
    OPINION
    BYBEE, Circuit Judge:
    In the aftermath of a massive earthquake and tsunami in
    Japan, the Fukushima Daiichi Nuclear Power Plant (FNPP)
    was damaged. Hundreds of United States servicemembers,
    deployed to provide relief to the victims, allege that they were
    exposed to radiation from the FNPP. The plaintiffs,
    servicemembers and their families, brought suit in California
    10          COOPER V. TOKYO ELEC. POWER CO.
    for negligence and strict products liability against Tokyo
    Electric Power Company (TEPCO), the power plant’s owner
    and operator, and General Electric Company (GE), the
    manufacturer of the plant’s boiling water reactors.
    This is the second time we have heard an appeal in this
    case. In 2017, we affirmed the district court’s denial of
    TEPCO’s motion to dismiss. Cooper v. Tokyo Elec. Power
    Co., 
    860 F.3d 1193
    (9th Cir. 2017) (“Cooper III”). On
    remand, GE and TEPCO both moved to dismiss. GE argued
    that Japanese law should apply to the case and that, under
    Japanese law, only the plant operator could be liable for
    injuries resulting from the power plant’s failure. TEPCO
    argued for dismissal on international-comity grounds. The
    district court granted both motions to dismiss. We affirm.
    I. FACTS AND PROCEDURAL HISTORY
    A. The Act on Compensation for Nuclear Damage
    In the early 1960s, the Japanese National Diet enacted
    legislation “to establish a basic system pertaining to
    compensation for damages in the case where nuclear damage
    has occurred in connection with the operation . . . of a nuclear
    reactor” and to “provid[e] protection for injured parties and
    contribut[e] to the sound development of nuclear reactor
    operations.” Act on Compensation for Nuclear Damage, Act
    No. 147, ch. I, art. 1 (June 17, 1961) (Compensation Act).
    The Compensation Act encouraged participation in Japan’s
    nuclear industry while ensuring compensation for any persons
    injured through operation of nuclear power plants. Articles
    3 and 4 of Chapter II of the Compensation Act provide that
    the operator of a nuclear plant is strictly liable for any
    damage caused by the operation of the power plant but that
    COOPER V. TOKYO ELEC. POWER CO.                        11
    “no other person shall be liable to compensate for damages.”
    Id. at ch.
    II, arts. 3–4. This is referred to as the “channeling
    provision.”
    These provisions, along with others that provide for the
    creation of a national insurance pool and financial backing
    from the Japanese government to fund compensation, work
    to facilitate recovery for accident victims by eliminating the
    need to prove fault and ensuring recovery of damages.
    Id.
    at ch.
    III, arts. 6–9.
    B. The 2011 Earthquake
    On March 11, 2011, a 9.0-magnitude earthquake and
    massive tsunami struck Japan, causing enormous and
    widespread destruction.1 Some 15,000 people died. The
    FNPP was damaged by the earthquake and tsunami and
    released over 300 tons of contaminated water into the sea. In
    response to the disaster, the United States joined in a
    humanitarian relief effort known as “Operation Tomodachi.”
    The day following the earthquake, the servicemember
    plaintiffs arrived off the coast of Fukushima on the U.S.S.
    Ronald Reagan and other vessels participating as part of the
    U.S. 7th Fleet’s Reagan Strike Force.
    Defendant TEPCO owns and operates the FNPP. After
    the FNPP meltdown, the Japanese government provided
    billions of dollars in financial support to TEPCO. It also
    developed a comprehensive scheme to deal with the
    thousands of claims resulting from the FNPP leak, giving
    1
    We have taken the facts from plaintiffs’ Third Amended Complaint.
    Additional details may be found in our prior opinion. Cooper 
    III, 860 F.3d at 1197
    –98.
    12         COOPER V. TOKYO ELEC. POWER CO.
    claimants the option to submit a claim (1) directly to TEPCO,
    (2) to the newly established Nuclear Damage Claim Dispute
    Resolution Center, or (3) to a Japanese court. The plaintiffs,
    however, chose to sue in the Southern District of California.
    Subject matter jurisdiction was asserted under the district
    court’s diversity jurisdiction. 28 U.S.C. § 1332(a)(2).
    C. Initial Complaints and the First Appeal
    The plaintiffs’ second amended complaint (SAC) alleged
    that TEPCO was negligent in operating and maintaining the
    FNPP. Six months later, the plaintiffs moved to amend their
    complaint to name GE and three other manufacturer
    defendants, claiming they had only recently learned of their
    involvement. Shortly thereafter, the district court granted in
    part TEPCO’s motion to dismiss the SAC. It found that the
    plaintiffs’ claims were not barred by the political-question
    doctrine, forum non conveniens, or the doctrine of
    international comity, but granted the motion in part because
    the plaintiffs failed to state claims for strict design-defect
    liability and intentional infliction of emotional distress.
    Cooper v. Tokyo Elec. Power Co., 
    2014 WL 5465347
    (S.D.
    Cal. Oct. 28, 2014) (Cooper I). The court granted leave to
    amend, including leave to name GE and the other
    manufacturers as defendants.
    The plaintiffs then filed their third-amended complaint
    (TAC) against TEPCO, GE, and three other named
    defendants. The TAC asserts claims for negligence, strict
    liability for ultrahazardous activities, res ipsa loquitur,
    negligence per se, loss of consortium, and survival and
    wrongful death against all defendants. The TAC also raises
    strict-liability claims for manufacturing and design defects
    against GE and the other named manufacturers.
    COOPER V. TOKYO ELEC. POWER CO.                   13
    GE and TEPCO separately moved to dismiss.
    Meanwhile, TEPCO moved for reconsideration of the denial
    of its motion to dismiss the SAC. The district court granted
    the motion for reconsideration but again denied TEPCO’s
    motion to dismiss the SAC. Cooper v. Tokyo Elec. Power
    Co., 
    166 F. Supp. 3d 1103
    (S.D. Cal. 2015) (Cooper II). The
    district court certified the issues for interlocutory appeal and
    denied the pending motions to dismiss as moot.
    On appeal, we affirmed the denial of TEPCO’s motion to
    dismiss the SAC. Cooper 
    III, 860 F.3d at 1218
    . We held that
    none of the arguments raised in TEPCO’s appeal warranted
    dismissal at that stage of the litigation, but allowed that
    “[f]urther developments . . . may require the district court to
    revisit some of the issues.”
    Id. at 1197;
    see also
    id. at 1210
    n.12.
    D. Proceedings on Remand
    After remand from Cooper III, and in light of a ruling in
    a parallel case filed by the same plaintiffs’ counsel raising
    similar issues, the district court relieved the defendants of the
    requirement to respond to the TAC and allowed the plaintiffs
    to file a fourth-amended complaint. The plaintiffs informed
    the court they would not do so, leaving the TAC as the
    operative complaint in this case.
    GE and TEPCO filed new motions to dismiss the TAC.
    GE argued that it could not be held liable because, under
    California’s choice-of-law rules, Japan’s Compensation Act
    applied and channeled all liability to TEPCO as the FNPP’s
    14           COOPER V. TOKYO ELEC. POWER CO.
    operator.2 TEPCO argued that the court lacked personal
    jurisdiction, and that, even if the court had jurisdiction over
    it, the doctrine of international comity required dismissal.3
    The district court granted both motions. As to GE, the
    district court first concluded that it had subject-matter
    jurisdiction over the suit. The court then, over the plaintiffs’
    objection, conducted a choice-of-law analysis. It determined
    that Japanese law governed with respect to third-party
    liability and, under that law, GE could not be held liable. As
    to TEPCO, the district court concluded that TEPCO had
    waived the personal-jurisdiction defense because it had not
    raised the issue in previous Rule 12 motions and there was no
    intervening change in the law that affected TEPCO’s ability
    to raise the defense. But the district court ultimately
    dismissed the claims against TEPCO without prejudice on
    international-comity grounds.
    II. ANALYSIS
    The plaintiffs appealed the district court’s ruling on both
    GE’s and TEPCO’s motions to dismiss. We begin with GE’s
    motion to dismiss before discussing TEPCO’s.
    2
    GE also argued that the TAC presented a political question, the
    claims were time-barred, the complaint should be dismissed under forum
    non conveniens, and the doctrine of international comity required
    dismissal. The district court did not reach these arguments.
    3
    TEPCO also revived its forum non conveniens argument before the
    district court, but the district court did not reach it.
    COOPER V. TOKYO ELEC. POWER CO.                     15
    A. GE
    The district court, after concluding under California’s
    choice-of-law rules that Japanese law applied, dismissed all
    claims against GE with prejudice. Plaintiffs do not dispute
    that if Japan’s Compensation Act applies, their claims must
    be dismissed because TEPCO, as the operator, is exclusively
    liable under the channeling provision for any damages.
    Instead, they raise three challenges to the district court’s
    choice-of-law ruling: first, that the channeling provision in
    the Compensation Act is procedural, not substantive, and
    therefore not subject to a choice-of-law analysis; second, that
    it was premature to decide choice-of-law questions at this
    stage of litigation; and, third, that the district court’s analysis
    was wrong and California’s strict products liability law
    should apply to the plaintiffs’ claims against GE. We review
    choice-of-law questions de novo, but review underlying
    factual findings for clear error. Daewoo Elecs. Am., Inc. v.
    Opta Corp., 
    875 F.3d 1241
    , 1246 (9th Cir. 2017). We apply
    California’s choice-of-law rules to this claim. See Klaxon
    Co. v. Stentor Elec. Mfg. Co., 
    313 U.S. 487
    , 496 (1941).
    1. Procedural Versus Substantive
    A federal district court sitting in diversity jurisdiction,
    28 U.S.C. § 1332, applies substantive state or foreign law.
    Erie R.R. Co. v. Tompkins, 
    304 U.S. 64
    , 78 (1938). “The nub
    of the policy that underlies [Erie] is that for the same
    transaction the accident of a suit by a non-resident litigant in
    a federal court instead of in a State court a block away,
    should not lead to a substantially different result.” Guar. Tr.
    Co. of N.Y. v. York, 
    326 U.S. 99
    , 109 (1945); see Van Dusen
    v. Barrack, 
    376 U.S. 612
    , 639 (1964) (a “change of
    courtrooms” should not lead to a change in the law applied to
    16          COOPER V. TOKYO ELEC. POWER CO.
    the parties). Accordingly, a federal district court will apply
    its own rules of procedure, but state or foreign substantive
    law. See, e.g., Abogados v. AT&T, Inc., 
    223 F.3d 932
    , 935
    (9th Cir. 2000).
    Unfortunately, “[c]lassification of a law as ‘substantive’
    or ‘procedural’ for Erie purposes is sometimes a challenging
    endeavor.” Gasperini v. Ctr. for Humanities, Inc., 
    518 U.S. 415
    , 427 (1996). “In determining whether a state law is
    substantive or procedural, we ask whether the law is outcome
    determinative,” that is, whether not applying the law would
    significantly affect the result of the litigation. Cuprite Mine
    Partners LLC v. Anderson, 
    809 F.3d 548
    , 555 (9th Cir. 2015).
    We have explained that “[a] substantive rule is one that
    creates rights or obligations” while a procedural rule “defines
    a form and mode of enforcing the substantive right or
    obligation.” County of Orange v. U.S. District Court (In re
    County of Orange), 
    784 F.3d 520
    , 527 (9th Cir. 2015)
    (internal quotation marks omitted). Our inquiry is “guided by
    ‘the twin aims of the Erie rule: discouragement of forum-
    shopping and avoidance of inequitable administration of the
    laws.’” Cuprite 
    Mine, 809 F.3d at 555
    (quoting 
    Gasperini, 518 U.S. at 428
    ). It is important to the fair administration of
    law that “‘the outcome of the litigation in the federal court
    should be substantially the same, so far as legal rules
    determine the outcome of a litigation, as it would be if tried
    in a State court.’” 
    Gasperini, 518 U.S. at 427
    (quoting Guar.
    
    Tr., 326 U.S. at 109
    ). The same logic applies when a foreign,
    rather than state, forum is at issue.
    The plaintiffs argue that the district court should not have
    conducted a choice-of-law analysis at all because the
    channeling provision in the Compensation Act is procedural
    and not substantive. According to the plaintiffs, the provision
    COOPER V. TOKYO ELEC. POWER CO.                        17
    effectively “strips a court of jurisdiction” over claims against
    anyone other than an operator of a nuclear plant. But the
    plaintiffs do not explain how that is the case. While it is a
    well-established principle that jurisdictional rules are not
    substantive, see McGee v. Int’l Life Ins. Co., 
    355 U.S. 220
    ,
    224 (1957), the plaintiffs make no meaningful argument
    about how the channeling provision is jurisdictional. The
    channeling provision makes no reference to jurisdiction or,
    indeed, to any court.4 It simply provides that only the
    operator of a nuclear reactor will be liable for any damages
    caused by the operation of the facility. That is not a
    jurisdictional provision. See Reed Elsevier, Inc. v. Muchnick,
    
    559 U.S. 154
    , 160–61 (2010).
    The channeling provision is much more akin to state
    statutes that limit liability for certain injuries.          If
    “[j]urisdiction is the power to declare law,” Ex parte
    McCardle, 74 U.S. (7 Wall.) 506, 514 (1869), the channeling
    provision is the law itself, not an assignment of the power to
    declare it. Importantly, California courts have routinely
    treated such liability-limiting provisions as substantive law
    and applied them under California’s choice-of-law rules. See,
    e.g., Offshore Rental Co. v. Cont’l Oil Co., 
    583 P.2d 721
    , 729
    (Cal. 1978) (applying Louisiana law foreclosing employer’s
    cause of action for negligent injury to key employee); Castro
    v. Budget Rent-A-Car Sys., Inc., 
    65 Cal. Rptr. 3d 430
    , 444
    (Ct. App. 2007) (applying Alabama law precluding liability
    against vehicle owner for negligence of a permissive user).
    4
    Under Japanese procedures put in place after the disaster, claims
    may be filed, but do not have to be filed, in a Japanese court. They may
    also be filed with TEPCO directly or with the Nuclear Damage Claim
    Dispute Resolution Center, created after the FNPP incident.
    18          COOPER V. TOKYO ELEC. POWER CO.
    Moreover, application of the channeling provision ends
    the case as to GE’s liability, making the provision, as the
    plaintiffs concede, outcome-determinative. This weighs
    heavily in favor of finding that the provision is substantive
    rather than procedural. See 
    Gasperini, 518 U.S. at 427
    –28.
    A liability-limiting statute with such outcome-determinative
    implications is substantive and subject to a choice-of-law
    analysis.
    2. Propriety of Choice-of-Law Analysis at the Motion-
    to-Dismiss Stage
    The plaintiffs next contend that the district court erred by
    conducting a choice-of-law analysis at this stage of the
    litigation. They argue that they needed additional time and
    discovery to fully develop the arguments and factual issues
    related to the choice-of-law analysis. The district court
    rejected this argument, finding that it was appropriate to
    analyze choice-of-law at this stage because the issue was
    fully briefed and discovery would not affect the analysis.
    The plaintiffs argue that a choice-of-law determination
    “requires a level of factual development and detailed legal
    briefing that was not and is not present in” this case at this
    time. But the plaintiffs cite no principle to support the
    argument that a court cannot decide choice-of-law issues in
    a motion to dismiss. And although some district courts
    reserve ruling on the issue until later in the litigation, the
    district court here had all the argument and facts necessary to
    make its decision.
    As to the contention that additional legal briefing was
    necessary, choice of law was one of the primary issues
    presented to the district court in the motion to dismiss. It was
    COOPER V. TOKYO ELEC. POWER CO.                   19
    fully briefed by both parties, and the district court was able to
    engage in a complete analysis. This is unlike some of the
    cases the plaintiffs cited, in which the parties had provided
    little or no briefing when asking the district court to decide
    the choice-of-law question. See, e.g., Dean v. Colgate-
    Palmolive Co., 
    2015 WL 3999313
    , at *11 (C.D. Cal. June 17,
    2015) (no analysis at all); Czuchaj v. Conair Corp., 
    2014 WL 1664235
    , at *9 (S.D. Cal. Apr. 18, 2014) (one page); Brazil
    v. Dole Food Co., 
    2013 WL 5312418
    , at *11 (N.D. Cal.
    Sept. 23, 2013) (no analysis at all).
    We are also not persuaded that the district court needed a
    more expansive factual record to decide the choice-of-law
    issue. The plaintiffs first argue that the terms of the contract
    between GE and TEPCO, particularly its choice-of-law or
    venue provisions, could influence the analysis. But those
    provisions have no bearing on tort claims filed by third
    parties, like the plaintiffs. Sutter Home Winery, Inc. v.
    Vintage Selections, Ltd., 
    971 F.2d 401
    , 407 (9th Cir. 1992)
    (“Claims arising in tort are not ordinarily controlled by a
    contractual choice of law provision.”); see also Paracor Fin.,
    Inc. v. Gen. Elec. Capital Corp., 
    96 F.3d 1151
    , 1165 (9th Cir.
    1996) (finding third parties cannot be bound by a choice-of-
    law provision in a contract in which they had no interest).
    The plaintiffs also contend that discovery could shed light on
    information about the operational responsibilities of GE and
    TEPCO, as well as TEPCO’s knowledge about the incoming
    U.S. naval ships. But the plaintiffs offer no explanation about
    why this information would be important to the analysis of
    whether California or Japan has a greater interest in the
    application of its substantive laws to the claims against GE.
    That information could be important to determining GE’s
    ultimate liability, but it has no bearing on the choice-of-law
    analysis.
    20          COOPER V. TOKYO ELEC. POWER CO.
    The district court did not err in proceeding with the full
    choice-of-law analysis.
    3. Choice of Law
    The final question is whether the district court erred when
    it decided that the laws of Japan, not California, govern
    plaintiffs’ claims against GE. California courts decide
    choice-of-law questions by means of the “governmental
    interest” test, which proceeds in three steps. Offshore 
    Rental, 583 P.2d at 723
    . First, the court must determine whether the
    substantive laws of California and the foreign jurisdiction
    differ on the issue before it. McGhee v. Arabian Am. Oil Co.,
    
    871 F.2d 1412
    , 1422 (9th Cir. 1989). Second, if the laws do
    differ, then the court must determine what interest, if any, the
    competing jurisdictions have in the application of their
    respective laws.
    Id. “If only
    one jurisdiction has a legitimate
    interest in the application of its rule of decision, there is a
    ‘false conflict’ and the law of the interested jurisdiction is
    applied.”
    Id. But if
    more than one jurisdiction has a
    legitimate interest, “the court must move to the third stage of
    the analysis, which focuses on the ‘comparative impairment’
    of the interested jurisdictions.”
    Id. This third
    step requires
    the court to “identify and apply ‘the law of the state whose
    interest would be the more impaired if its law were not
    applied.’”
    Id. (quoting Offshore
    Rental, 583 P.2d at 726
    ).
    a. The Laws Differ
    The parties agree that the laws of California and Japan
    differ. Under Japanese law, the Compensation Act would
    apply and channel liability for nuclear damage exclusively to
    the licensed operator of the nuclear installation—here,
    TEPCO. If Japanese law applies, it requires dismissal of all
    COOPER V. TOKYO ELEC. POWER CO.                   21
    claims against GE. Under California law, on the other hand,
    a manufacturer such as GE is strictly liable if its product is
    defectively manufactured, defectively designed, or distributed
    without adequate instructions or warnings. See Hufft v.
    Horowitz, 
    5 Cal. Rptr. 2d 377
    , 379 (Ct. App. 1992). If
    California law applies and the plaintiffs prove that GE
    defectively manufactured or designed the reactor, GE would
    be strictly liable. See
    id. b. Both
    Jurisdictions Have a Legitimate Interest
    We must next determine what interest, if any, Japan and
    California have in the application of their respective laws to
    this case. Offshore 
    Rental, 583 P.2d at 724
    –25. Only if each
    jurisdiction involved has a legitimate but conflicting interest
    in applying its own law will there be a “true conflict,”
    requiring us to move on to step three of the analysis.
    Id. at 725–26.
    The plaintiffs agree that there is a true conflict,
    but contend that Japan’s interests are not “strong.” GE argues
    that there is no true conflict because, while Japan has
    substantial, legitimate interests in applying its laws,
    California’s interests are “minimal at best.” It asserts that the
    plaintiffs’ claims directly implicate conduct that occurred in
    Japan and is subject to a Japanese liability-limiting statute,
    giving Japan strong legitimate interests in having its law
    applied. In contrast, GE contends that California’s only
    interest is in ensuring compensation for California-resident
    victims, which would be equally served under Japanese law.
    At this point in the analysis, our only consideration is
    whether each jurisdiction has legitimate interests in seeing its
    own law applied in this case and whether those interests
    conflict. Weighing the strength of California’s interests
    22          COOPER V. TOKYO ELEC. POWER CO.
    against Japan’s occurs at the third step, which we need only
    reach if there is a true conflict.
    (1) Japan’s Interests. The parties generally agree that
    Japan has legitimate interests in having its law applied to this
    case. These interests are: (1) adjudicating claims arising from
    a natural disaster that occurred in Japan, (2) adjudicating
    claims arising from injuries that occurred in Japan, and
    (3) providing consistent allocation of liability for nuclear
    disasters under the Compensation Act. The final interest is of
    particular importance. As the California Supreme Court has
    stated:
    When a state adopts a rule of law limiting
    liability for commercial activity conducted
    within the state in order to provide what the
    state perceives is fair treatment to, and an
    appropriate incentive for, business enterprises,
    we believe that the state ordinarily has an
    interest in having that policy of limited
    liability applied to out-of-state companies that
    conduct business in the state, as well as to
    businesses incorporated or headquartered
    within the state.
    McCann v. Foster Wheeler LLC, 
    225 P.3d 516
    , 530 (Cal.
    2010). Here, Japan’s Compensation Act limits liability for
    participants in its nuclear industry, in part as an incentive for
    businesses to participate. This is a “real and legitimate
    interest” in having Japanese law apply to the case.
    Id. at 531–32.
    (2) California’s Interest.    California law holds
    manufacturers strictly liable for products defectively
    COOPER V. TOKYO ELEC. POWER CO.                 23
    manufactured or designed. 
    Hufft, 5 Cal. Rptr. 2d at 379
    .
    California courts have described the state’s interest
    underlying this law:
    [It] is to [e]nsure that the costs of injuries
    resulting from defective products are borne by
    the manufacturers that put such products on
    the market rather than by the injured persons
    who are powerless to protect themselves. The
    other purposes, or public policies, behind the
    creation of the doctrine of strict products
    liability in tort as a theory of recovery are:
    (1) to provide a short cut to liability where
    negligence may be present but difficult to
    prove; (2) to provide an economic incentive
    for improved product safety; (3) to induce
    allocation of resources towards safer products;
    and (4) to spread the risk of loss among all
    who use the product.
    Barrett v. Superior Court, 
    272 Cal. Rptr. 304
    , 309 (Ct. App.
    1990) (internal quotation marks and citations omitted). These
    interests are certainly legitimate.
    GE, however, contends that these interests are
    insignificant in this case, arguing that the plaintiffs “cannot
    manufacture a true conflict by invoking irrelevant policies
    and interests.” It argues that policies underlying California’s
    strict-products-liability law “are immaterial here because
    there are no ‘products’ at issue.” But that is not the case.
    While it is true that the district court previously commented
    that “[t]he FNPP was evidently not a product ‘placed on the
    market,’” Cooper 
    II, 166 F. Supp. 3d at 1129
    , the claims
    against GE do not arise out of the FNPP as a single entity.
    24          COOPER V. TOKYO ELEC. POWER CO.
    GE manufactured particular parts of the Fukushima Daiichi
    facility—the reactors. The fact that reactors were not
    marketed broadly to consumers does not detract from the fact
    that they were designed and built for the FNPP. That is
    sufficient, in a proper case, to be subject to California’s
    products liability rules. See Rawlings v. D.M. Oliver, Inc.,
    
    159 Cal. Rptr. 119
    , 122 (Ct. App. 1979) (holding that the fact
    that a product was not mass produced has no effect on the
    manufacturer’s responsibilities in manufacturing and selling
    products).
    Although there are no California defendants in this case,
    there are plaintiffs who are California residents. And
    California has a legitimate interest in ensuring that its injured
    residents are compensated for injuries resulting from the
    design and manufacture of faulty products, as well as
    providing an easy way to prove liability. So, the interests
    served by California’s strict-products-liability laws are also
    relevant.
    We conclude, as did the district court, that there is a so-
    called “true conflict” here. California has an interest in
    holding manufacturers of defective products liable in tort to
    ensure compensation for its residents. Japan, on the other
    hand, has an interest in consistent application of its liability-
    limiting statute to businesses participating in its nuclear
    industry. We therefore move to the final step of the analysis.
    c. Japan’s Interests Would be More Impaired if Its
    Law Were Not Applied
    Once a true conflict is identified, we must consider the
    “comparative impairment” step of the analysis, which “seeks
    to determine which state’s interest would be more impaired
    COOPER V. TOKYO ELEC. POWER CO.                  25
    if its policy were subordinated to the policy of the other
    state.” Offshore 
    Rental, 583 P.2d at 726
    . The conflict
    “should be resolved by applying the law of the state whose
    interest would be the more impaired if its law were not
    applied.”
    Id. The purpose
    of this step is not for us to judge
    which law is “better” or “worthier” social policy; instead, we
    are “to decide—in light of the legal question at issue and the
    relevant state interests at stake—which jurisdiction should be
    allocated the predominating lawmaking power under the
    circumstances of the present case.” 
    McCann, 225 P.3d at 534
    .
    California’s courts have frequently applied foreign laws
    that serve to protect businesses by limiting liability, even
    when applying that law precludes recovery by injured
    California residents. For example, in Offshore Rental, a
    California corporation brought a claim against a Louisiana
    company for the loss of services of a “key” employee who
    was injured on the defendant company’s premises in
    
    Louisiana. 583 P.2d at 729
    . The California Supreme Court
    noted that Louisiana’s interest in applying its own law to the
    case was “to protect negligent resident tort-feasors acting
    within Louisiana’s borders from the financial hardships
    caused by the assessment of excessive legal liability or
    exaggerated claims resulting from the loss of services of a
    key employee.”
    Id. at 725.
    California had made a different
    choice in legal policies: California had “an interest in
    protecting California employers from economic harm because
    of negligent injury to a key employee inflicted by a third
    party.”
    Id. Weighing these
    competing interests, the court
    held that “[a]t the heart” of Louisiana’s liability-limiting law
    was “the vital interest in promoting freedom of investment
    and enterprise within Louisiana’s borders, among investors
    incorporated both in Louisiana and elsewhere.”
    Id. at 728.
    26            COOPER V. TOKYO ELEC. POWER CO.
    Imposing liability in this situation, when Louisiana had
    decided not to, would therefore “strike at the essence of a
    compelling Louisiana law.”
    Id. Particularly because
    the
    accident occurred in Louisiana, California’s interest in
    compensation for injured California companies could not
    overcome Louisiana’s greater interest in protecting
    businesses operating there.
    Id. at 728–29.
    Applying
    Louisiana law and finding no cause of action, the court
    affirmed dismissal of the suit.
    Similarly, in McCann, the California Supreme Court
    applied Oklahoma law that limited a defendant’s liability
    even though it precluded recovery for a California 
    plaintiff. 225 P.3d at 537
    . McCann was exposed to asbestos while
    working in an oil refinery in Oklahoma in the 1950s. Many
    years later he developed mesothelioma, and brought suit in
    his home state, California, against the manufacturer of a
    boiler installed in the refinery. The manufacturer was a
    resident of neither Oklahoma nor California, but of New
    York. Relying on Offshore Rental, the court applied
    Oklahoma’s 10-year statute of repose to the plaintiff’s claim
    instead of California’s statute of limitations.5
    Id. The court
    noted that Oklahoma had an interest in promoting commercial
    activity within the state by limiting businesses’ liability,
    while California had a general interest in ensuring
    compensation for its injured residents, and had a special
    interest in providing relief from “asbestos-related harm.”
    Id. at 532.
    Nevertheless, the court concluded that applying
    5
    Under Oklahoma’s statute of repose, the time for filing suit ran from
    the time the construction project was completed, whether McCann knew
    of his injury or not. Under California’s statute of limitations, McCann had
    one year from the time he learned of his mesothelioma. 
    McCann, 225 P.2d at 521
    & n.2, 523, 529.
    COOPER V. TOKYO ELEC. POWER CO.                         27
    California’s statute of limitations would “significantly
    undermine Oklahoma’s interest in establishing a reliable rule
    of law governing a business’s potential liability for conduct
    undertaken in Oklahoma.”
    Id. at 535.
    In contrast, failure to
    apply California law would have had a less significant impact
    on California’s interest.
    Id. While it
    precluded the plaintiff’s
    recovery, California courts take a “restrained view” of
    California’s interest in recovery for its residents for injuries
    that occur in another jurisdiction.
    Id. at 535–36
    (discussing
    Offshore 
    Rental, 583 P.2d at 728
    , and Castro, 
    65 Cal. Rptr. 3d
    at 443–44). Given Oklahoma’s strong interest in limiting
    liability for commercial activity within its borders, the
    California Supreme Court applied Oklahoma law and
    dismissed the McCann’s claim.
    Id. at 537.
    In light of this precedent, the district court correctly
    decided that Japanese law should apply to this case. Japan’s
    interests here are similar to those at issue in Offshore Rental,
    McCann, and other cases in which California courts (and
    federal courts applying California’s choice-of-law rules) have
    found that California’s interest in compensation for injured
    residents failed to overcome a foreign jurisdiction’s interest
    in limiting defendants’ substantive liability for injuries
    occurring within its borders.6 Japan’s interest here is in
    6
    See Arno v. Club Med Inc., 
    22 F.3d 1464
    , 1469 (9th Cir. 1994)
    (applying French law to plaintiff’s vicarious-liability claim because
    Guadeloupe’s interest in “encouraging local industry . . . and reliably
    defining the duties and scope of liability of an employer doing business
    within its borders” would be more impaired than California’s interest in
    “providing compensation to its residents” if its law was not applied);
    Castro, 
    65 Cal. Rptr. 3d
    at 443–44 (applying Alabama law because its
    interest in allocating liability would be more impaired by application of
    California’s more permissive statute than would California’s interest in
    compensation for injured residents if Alabama law was applied); Tucci v.
    28            COOPER V. TOKYO ELEC. POWER CO.
    limiting liability for defendants engaged in the nuclear-power
    industry in Japan. Japan made a conscious decision to
    encourage nuclear power in Japan. It balanced “providing
    protection for injured parties” with “contributing to the sound
    development of nuclear reactor operations.” Compensation
    Act, ch. I, art. 1. Under the Compensation Act, “the Nuclear
    Operator is liable to compensate for damages in connection
    with the Operation . . . of [the] Nuclear Reactor” and “no
    other persons shall be liable to compensate for damages other
    than the Nuclear Operator.”
    Id. at ch.
    II, arts. 3–4. In
    comparing Japan’s and California’s interests, we cannot judge
    which policy embodies “the better or worthier rule,” but
    instead must determine which jurisdiction’s interest would be
    most “significantly impair[ed]” if its law were not applied.
    
    McCann, 225 P.3d at 534
    .
    We have little difficulty concluding that failure to apply
    Japanese law in these circumstances would significantly
    impair Japan’s interests. Japan’s Compensation Act is
    directed specifically at accidents at a nuclear facility;
    California’s products liability rules are general in nature and
    presumably cover everything from toasters to airplanes. The
    release of radiation occurred at the FNPP on Japanese soil
    and the United States sent the U.S.S. Ronald Reagan to Japan
    in aid of the disaster.7
    Club Mediterranee, S.A., 
    107 Cal. Rptr. 2d 401
    , 408–12 (Ct. App. 2001)
    (applying Dominican law in light of the Dominican Republic’s superior
    interest in “assuring that businesses . . . face limited and predictable
    financial liability for work-related injuries”).
    7
    Before the district court and in their opening brief, the plaintiffs
    argued that the injury here did not occur in Japan because it happened in
    international waters. In reply and at oral argument, the plaintiffs amended
    that argument and now claim that the injury occurred on “U.S. soil”
    COOPER V. TOKYO ELEC. POWER CO.                         29
    Even if application of the Compensation Act would bar
    any relief for these plaintiffs, we would still be required to
    apply Japanese law. See 
    McCann, 225 P.3d at 537
    –38;
    Offshore 
    Rental, 583 P.2d at 729
    . But application of
    Japanese law does not entirely foreclose recovery for the
    plaintiffs here. Japanese law allows for compensation for the
    plaintiffs’ injuries—just not from GE.           This makes
    application of Japanese law less intrusive on California’s
    interests than in cases like McCann and Offshore Rental. For
    these reasons, Japanese law applies to the claims against GE.
    Because there is no dispute on appeal that application of
    Japanese law requires dismissal of all claims against GE, we
    affirm the dismissal of these claims with prejudice.8
    B. TEPCO
    We next address TEPCO’s motion to dismiss. The
    district court also engaged in a choice-of-law analysis for the
    claims against TEPCO. After determining that Japanese law
    should apply, the court dismissed the claims against TEPCO
    on international-comity grounds. We begin by addressing the
    choice-of-law analysis. We then consider international
    comity.
    1. Choice of Law
    The plaintiffs raise the same challenges to this choice-of-
    law analysis as they did to the analysis of the claims against
    because the sailors were injured on U.S. ships. This argument, presented
    for the first time in the reply, has been forfeited. See Rizk v. Holder,
    
    629 F.3d 1083
    , 1091 n.3 (9th Cir. 2011).
    8
    We therefore need not reach the alternative arguments in GE’s brief.
    30           COOPER V. TOKYO ELEC. POWER CO.
    GE. For the reasons previously stated, the choice-of-law
    analysis is not premature or inappropriate at this stage. As to
    the merits of the choice-of-law analysis, the district court
    correctly found that Japanese law also applies to the
    plaintiffs’ claims against TEPCO.
    a. The Laws Differ
    There is no disagreement that the laws of Japan and
    California differ in three ways with regard to the claims
    against TEPCO. First, under Japanese law, the Compensation
    Act is the exclusive means of redress, Saikô Saibansho [Sup.
    Ct.] May 14, 2009, 2066 HANREI JIHÔ [HANJI] 54,9 but under
    California law, there is no such exclusive remedy. Second,
    the Compensation Act requires a “high probability” of
    causation, Saikô Saibansho [Sup. Ct.] Oct. 24, 1975, 29 Saikô
    Saibansho Minji Hanreishû [Minshû] 1417, 1419–20, while
    California negligence principles require the plaintiffs to show
    only that their injuries were “more likely than not” caused by
    radiation exposure. Jones v. Ortho Pharm. Corp., 209 Cal.
    Rptr. 456, 460 (Ct. App. 1985). Finally, Japanese law has a
    broad definition of compensatory damages, including
    damages for proprietary and material losses, spiritual or
    mental suffering (“consolation money”), and income lost over
    a lifetime, but it does not recognize or allow for punitive
    damages, Saikô Saibansho [Sup. Ct.] July 11, 1997, 51(6)
    9
    The cases cited here were explained in a detailed declaration from
    Yasuhei Taniguchi, a retired professor of law who has taught at several
    Japanese universities. In addition to an LL.B. from Kyoto University,
    Professor Taniguchi holds an LL.M. from the University of California at
    Berkeley and a J.S.D. from Cornell University. His analysis was credited
    by the district court and is not disputed by the plaintiffs.
    COOPER V. TOKYO ELEC. POWER CO.                        31
    Saikô Saibansho Minji Hanreishû [Minshû] 2573, while
    California law does. Cal. Civ. Code § 3294.
    b. Both Jurisdictions Have Legitimate Interests
    The plaintiffs made no argument here or before the
    district court regarding the interests of either forum. We
    conclude that the same interests are implicated here as in the
    analysis of the claims against GE. California has an interest
    in ensuring compensation for its injured residents, while
    Japan has an interest in the consistent application of the
    Compensation Act to protect its nuclear industry. There is
    therefore a true conflict and we proceed to step three of the
    governmental-interest test.
    c. Japan’s Interests Would be More Impaired if Its
    Law Was Not Applied
    The analysis at this step is also much the same as for the
    claims against GE. As 
    noted supra
    , California courts have
    frequently applied foreign laws like the Compensation
    Act—which serve to limit liability for businesses—in these
    situations. Japan has an even greater interest in its law
    applying to the claims against TEPCO than it did with respect
    to GE. TEPCO is a Japanese corporation operating a nuclear
    reactor in Japan. It is not only subject to general principles of
    Japanese law but, as evidenced by the Compensation Act, to
    a series of special rules regarding its responsibility following
    a nuclear disaster, just as American nuclear plant operators
    are subject to special liability rules under the Price-Anderson
    Act.10 Furthermore, following the disaster and consistent
    10
    See 42 U.S.C. § 2210. See also Duke Power Co. v. Carolina Envtl.
    Study Grp., 
    438 U.S. 59
    , 63–67 (1978) (background on the Act).
    32            COOPER V. TOKYO ELEC. POWER CO.
    with the Compensation Act, the Japanese government has
    come forward to fund compensation for the victims of the
    FNPP meltdown. We were advised in TEPCO’s brief and at
    oral argument that the Japanese government has allocated, to
    date, more than $76 billion to compensate victims, and that
    more than 21,000 victims have received some form of
    compensation. Japan has a strong interest in the uniform
    application of the Compensation Act. Subjecting TEPCO to
    California’s negligence rules would seriously undermine the
    comprehensive scheme in the Compensation Act and impair
    Japan’s interests. In contrast, California has an interest in
    seeing the victims of a nuclear disaster compensated, but that
    interest would be equally served under Japanese law.11 The
    Compensation Act operates to compensate those injured by
    nuclear accidents and the plaintiffs offered no showing that
    they cannot be adequately compensated for their injuries
    under Japanese law.12
    Because Japan’s interests would be more impaired if
    California’s laws were applied than California’s would if
    Japanese law were applied, we conclude that Japanese law
    applies to the claims against TEPCO and affirm the district
    court’s holding on the choice-of-law issue. We now proceed
    to the question of whether, given that Japanese law must be
    applied in any proceedings in the Southern District of
    11
    Of course, as discussed, the plaintiffs would not be able to recover
    punitive damages under Japanese law. But punitive damages are not
    intended to compensate for a plaintiff’s losses, see State Farm Mut. Auto.
    Ins. Co. v. Campbell, 
    538 U.S. 408
    , 416 (2003), and TEPCO has stated
    that injured parties will be fully compensated for proven injuries.
    12
    At oral argument, TEPCO agreed to waive any statute of limitations
    defense provided the plaintiffs filed their claims in Japan within a
    reasonable amount of time.
    COOPER V. TOKYO ELEC. POWER CO.                   33
    California, the district court abused its discretion in
    dismissing the case on international-comity grounds.
    2. International Comity
    “International comity ‘is the recognition which one nation
    allows within its territory to the legislative, executive or
    judicial acts of another nation, having due regard both to
    international duty and convenience, and to the rights of its
    own citizens or of other persons who are under the protection
    of its laws.’” Mujica v. AirScan Inc., 
    771 F.3d 580
    , 597 (9th
    Cir. 2014) (quoting In re Simon, 
    153 F.3d 991
    , 998 (9th Cir.
    1998)) (other citations omitted). It is “a doctrine of
    prudential abstention, one that ‘counsels voluntary
    forbearance when a sovereign which has a legitimate claim to
    jurisdiction concludes that a second sovereign also has a
    legitimate claim to jurisdiction under principles of
    international law.’”
    Id. at 598
    (quoting United States v.
    Nippon Paper Indus. Co., 
    109 F.3d 1
    , 8 (1st Cir. 1997)).
    International comity embodies the policy of “good
    neighbourliness, common courtesy and mutual respect
    between those who labour in adjoining judicial vineyards.”
    Id. There are
    two kinds of international comity: prescriptive
    comity (addressing “the extraterritorial reach of federal
    statutes”) and adjudicative comity (a “discretionary act of
    deference by a national court to decline to exercise
    jurisdiction in a case properly adjudicated in a foreign state”).
    Id. at 598
    –99. This case deals only with adjudicative comity.
    In deciding whether to apply the doctrine of adjudicative
    comity, the courts weigh “several factors, including [1] the
    strength of the United States’ interest in using a foreign
    34          COOPER V. TOKYO ELEC. POWER CO.
    forum, [2] the strength of the foreign governments’ interests,
    and [3] the adequacy of the alternative forum.”
    Id. at 603
    (quoting Ungaro-Benages v. Dresdner Bank AG, 
    379 F.3d 1227
    , 1238 (11th Cir. 2004)) (brackets in original)). In
    Mujica, we expounded on how to assess the United States’
    and the foreign government’s interests in relying on a foreign
    forum:
    The (nonexclusive) factors we should
    consider when assessing [each country’s]
    interests include (1) the location of the
    conduct in question, (2) the nationality of the
    parties, (3) the character of the conduct in
    question, (4) the foreign policy interests of the
    [countries], and (5) any public policy
    interests.
    Id. at 604;
    see also
    id. at 607.
    These considerations need not
    be addressed mechanically.
    We review the district court’s international-comity
    determination for an abuse of discretion and reverse only if
    the district court applied an incorrect legal standard or if its
    “application of the correct legal standard was (1) ‘illogical,’
    (2) ‘implausible,’ or (3) without ‘support in the inferences
    that may be drawn from the facts in the record.’”
    Id. at 589
    (quoting United States v. Hinkson, 
    585 F.3d 1247
    , 1262 (9th
    Cir. 2009) (en banc)). The district court here “correctly laid
    out [the] legal standard,” so “the only question is whether the
    district court’s decision . . . to dismiss Plaintiffs’ claims was
    illogical, implausible, or unsupported by the record.”
    Cooper 
    III, 860 F.3d at 1205
    .
    COOPER V. TOKYO ELEC. POWER CO.                            35
    In our previous opinion, we concluded that the district
    court did not abuse its discretion when it decided against
    dismissing the claims against TEPCO on comity grounds.
    Id. at 1209.
    We recognized that this was a “difficult case” and
    that there were “strong reasons for dismissing the plaintiffs’
    claims in favor of a Japanese forum.”
    Id. We noted
    that
    “further developments in the district court may counsel in
    favor” of dismissal, particularly once the district court
    determined which jurisdiction’s law would apply.
    Id. at 1210
    n.12. On remand, the district court reconsidered its comity
    analysis based on new developments, finding that these
    developments tilted the scales towards dismissal. The
    location of the conduct in question, nationality of the parties,
    nature of the conduct, and the public-policy interests
    remained the same. But after considering the statements from
    the Japanese and United States governments—which the
    district court did not have before it when it first ruled on the
    issue—the district court found that the foreign-policy
    interests now favored dismissal.13 And because the choice-
    of-law analysis is relevant to comity decisions, the district
    court found that its conclusion that Japanese law applied also
    13
    The district court also left its previous decision on the adequacy-of-
    the-alternative-forum factor undisturbed, and the plaintiffs do not
    challenge this factor on appeal. They make passing reference to “the
    disparities between the American justice system and Japan’s,” but do so
    without citation to the record or law that supports the implication that
    Japan would be an inadequate forum. They claim that the defendants
    needed to present “clear and incontrovertible evidence” that Japan’s courts
    would not “deprive Plaintiffs of due process and equal protection of law
    to which they are entitled,” but our precedent imposes no such
    requirement. The plaintiffs raised similar unsubstantiated claims of bias
    in their previous appeal, but we found that there was “no doubt that Japan
    would provide an adequate alternative forum.” Cooper 
    III, 860 F.3d at 1209
    . The district court did not abuse its discretion by leaving its
    previous analysis of this factor in place.
    36          COOPER V. TOKYO ELEC. POWER CO.
    weighed in favor of dismissal on comity grounds. The
    plaintiffs contend that this was an abuse of discretion because
    “nothing has changed except for the court’s willingness to
    revisit the issue of international comity and decide to punt
    this case to Japan.”
    First, the conclusion that Japanese law applies to the case
    does affect the comity analysis. See Cooper 
    III, 860 F.3d at 1210
    n.12 (citing 
    Mujica, 771 F.3d at 602
    ; Ungaro-
    
    Benages, 379 F.3d at 1240
    ). It was not an abuse of discretion
    for the district court to take the applicability of Japanese law
    into consideration. If Japan’s interest in the applicability of
    its laws to this case was strong enough to overcome
    California’s interests in the choice-of-law analysis, it was not
    illogical or implausible for the district court to find that Japan
    had a similarly strong interest in being the place where the
    plaintiffs’ claims are litigated. We can take notice of the fact
    that if the suit proceeds in the Southern District of California,
    the district court will have to inform itself at every turn of the
    nuances of Japanese civil law. That would require
    understanding who bears the burden of proof, principles of
    causation, and what constitute compensable damages. Not
    only would the district court have to educate itself on the law,
    but it would need to understand how the Compensation Act
    has been administered in the thousands of cases resolved in
    Japan, lest the “change of courtrooms” mean a change in
    result. Van 
    Dusen, 376 U.S. at 639
    .
    The other “significant change” that the district court
    found affected its analysis was the amicus briefs filed in our
    court during the first appeal. Neither government had
    expressed its views on litigating in U.S. courts before
    Cooper III. After Japan filed an amicus brief in Cooper III
    expressing a strong interest in the case being litigated in
    COOPER V. TOKYO ELEC. POWER CO.                          37
    Japan, we invited the United States Department of State to
    give its views on whether the litigation should proceed in the
    United States.14 We considered both amicus briefs and found
    that they expressed “important, competing policy interests”
    that required “difficult judgment calls” from the district court.
    Cooper 
    III, 860 F.3d at 1209
    . On remand, the district court
    weighed the interests expressed in the governments’ amicus
    briefs and decided that, in light of the Japanese government’s
    strong objection to the case being litigated in the United
    States, the foreign-policy factor now weighed in favor of
    dismissal.
    This was a significant change from the first time the
    district court engaged in the comity analysis, despite the
    plaintiffs’ assertions to the contrary. The first time the
    district court considered the comity factors, neither Japan nor
    the United States had expressed an opinion to the district
    court about the appropriate venue for the litigation. It was not
    improper for the district court to reconsider its previous
    holding in light of those statements.
    14
    In Cooper III, TEPCO and GE (appearing as amicus) argued that
    dismissal on comity grounds would promote the Convention on
    Supplementary Compensation for Nuclear Damage (“CSC”). The United
    States argued that, in denying dismissal on comity grounds, the district
    court did not abuse its discretion. See Cooper 
    III, 860 F.3d at 1199
    –1200.
    The CSC guarantees that its contracting parties will have exclusive
    jurisdiction over litigation arising out of a nuclear incident within their
    borders. Because Japan was not a contracting party to the CSC at the time
    of the FNPP disaster, the United States objected to the courts relying on
    this argument from TEPCO and GE. Otherwise, the United States argued
    that it had “no specific foreign policy interest necessitating dismissal in
    this particular case.”
    Id. at 1208.
    38          COOPER V. TOKYO ELEC. POWER CO.
    The plaintiffs suggest that the district court effectively
    “overruled the Ninth Circuit,” by weighing the amicus briefs
    and coming to a different conclusion. But the district court
    did nothing of the kind. In Cooper III, we fairly invited the
    district court to revisit the comity analysis if and when
    circumstances changed.
    Id. at 1210
    & n.12. And the
    statement from both governments about where the litigation
    should proceed was a changed circumstance for the district
    court. The argument that the statements were not new
    because we had them before us in Cooper III misunderstands
    the scope of our review in that case. The question before us
    then was not whether we, with the benefit of the statements,
    would have dismissed the suit on comity grounds, but
    whether the district court had abused its discretion with the
    information that it had. We were careful in stating our
    standard of review in that case. Cooper 
    III, 860 F.3d at 1205
    ,
    1209. Once the district court had the positions of the United
    States and Japanese governments before it, it was entirely
    proper for the district court to revisit the comity analysis.
    In its amicus brief, Japan strongly objected to this case
    being litigated in the United States. Japan has committed a
    significant sum of money and resources to ensure fair and
    consistent compensation for accident victims. Japan pointed
    out that if injured parties could bring their claims anywhere
    in the world, foreign courts might apply different legal
    standards, resulting in different outcomes for similarly
    situated victims. See
    id. at 1207.
    This would seriously affect
    the integrity of the compensation system established by the
    Japanese government. And because the Japanese government
    is financing TEPCO’s compensation payments, which are
    administered through Japanese courts, that risk is particularly
    troublesome. See
    id. at 1209
    (“Japan has an undeniably
    strong interest in centralizing jurisdiction over FNPP-related
    COOPER V. TOKYO ELEC. POWER CO.                     39
    claims.”). If Japan cannot exercise some control over the
    compensation process, it may be less willing to compensate
    FNPP victims who seek remedies outside of a Japanese
    forum. That may complicate the victims’ ability to be
    compensated. See Cooper 
    III, 860 F.3d at 1207
    (“Judgments
    originating in American courts may well be inconsistent with
    the overall administration of Japan’s compensation fund.”)
    When the district court revisited the comity factors, it
    noted TEPCO’s argument that Japan’s interests “have only
    grown stronger” since its brief was filed. With the benefit of
    this position, the district court found that the foreign policy
    factor weighed in favor of dismissal, despite the United
    States’ “important, competing policy interest.”
    The plaintiffs contend that it was illogical for the district
    court not to consider the United States’ amicus brief in its
    analysis and that the district court owed deference to the State
    Department’s opinion about whether to exercise jurisdiction.
    But the district court acknowledged the United States’
    statement and its competing foreign-policy concerns in its
    order. And to the extent that the plaintiffs contend that the
    State Department’s brief is an affirmative statement from the
    government that was entitled to special deference, the
    plaintiffs overstate their case. The plaintiffs point to no
    principle that requires district courts to defer to statements of
    interest from the United States. We will give “serious weight
    to” such statements, Sosa v. Alvarez-Machain, 
    542 U.S. 692
    ,
    733 n.21 (2004), but “[a] statement of national interest alone
    . . . does not take the present litigation outside the competence
    of the judiciary,” 
    Ungaro-Benages, 379 F.3d at 1236
    .
    Moreover, the United States issued a careful, cautious
    statement. The United States first “applaud[ed] the
    40          COOPER V. TOKYO ELEC. POWER CO.
    Government of Japan’s impressive efforts to provide recovery
    for damages caused by the nuclear accident at the Fukushima-
    Daiichi power plant, including through the creation of an
    administrative compensation scheme that has paid over
    $58 billion in claims.” It did express an interest in Japan not
    retroactively receiving exclusive jurisdiction over suits under
    the CSC, which Japan had not signed at the time of the FNPP
    incident. And, for that reason, the U.S. urged us not to
    overturn the district court’s decision in Cooper III. See
    Cooper 
    III, 860 F.3d at 1207
    –09. But outside of that, the
    United States said only that although “a district court could
    choose to dismiss a case based on international comity for a
    claim arising overseas[,] . . . [t]he United States has no
    specific firm policy interest necessitating dismissal in this
    particular case.” The United States stopped well short of
    urging that California was the proper forum to exercise
    jurisdiction in this case. The United States thus voiced its
    concerns with the reasons for which the district court would
    grant dismissal on comity grounds, but expressed no
    objection that Japan be permitted to adjudicate these claims
    in its own courts.
    The United States’ measured response pales in
    comparison to Japan’s unequivocal objection to the exercise
    of jurisdiction in U.S. courts. Recognizing Japan’s interests
    under these circumstances was not illogical or implausible,
    particularly once the district court determined that Japanese
    law would apply to the claims.
    We acknowledge that the case is complicated. It
    implicates strong, important policy interests in both countries.
    But comity is a “fluid doctrine” that can “change in the
    course of the litigation.” Cooper 
    III, 860 F.3d at 1210
    . We
    invited the district court to reevaluate its decision in
    COOPER V. TOKYO ELEC. POWER CO.                     41
    appropriate circumstances. The district court did so, and
    carefully explained its reasons. Having decided that Japanese
    law applies to the case and considering Japan’s strong
    interests in the case being litigated in Japan, the district court
    did not abuse its discretion when it dismissed the claims
    against TEPCO on international-comity grounds.15
    III. CONCLUSION
    We affirm the district court’s granting of both GE’s and
    TEPCO’s motions to dismiss.
    AFFIRMED.
    15
    In light our disposition, we do not reach TEPCO’s personal-
    jurisdiction argument.