FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

MELISSIA HENSON; KEITH TURNER, No. 18-56071 Plaintiffs-Appellants, D.C. No. v. CV 14-1240

ODW

FIDELITY NATIONAL FINANCIAL,

INC.,

Defendant-Appellee. OPINION

Appeal from the United States District Court for the Central District of California Otis D. Wright II, District Judge, Presiding

Argued and Submitted June 10, 2019 Anchorage, Alaska

Filed November 15, 2019

Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.

Opinion by Judge Tashima 2 HENSON V. FIDELITY NAT’L FINANCIAL

SUMMARY*

Motion for Relief from Judgment

The panel reversed the district court’s order denying plaintiffs’ Fed. R. Civ. P. 60(b)(6) motion for relief from judgment in an action under the Real Estate Settlement Procedures Act.

Plaintiffs sought relief from judgment based on an intervening change in the law in Microsoft Corp. v. Baker, 137 S. Ct. 1702 (2017) (holding that plaintiffs in putative class actions cannot transform a tentative interlocutory order into a final appealable judgment simply by dismissing their claims with prejudice). The panel addressed the analysis that courts should employ to guide their discretion when evaluating the merits of a Rule 60(b)(6) motion on the ground of an intervening change in the law in a non-habeas corpus case. The panel held that many of the factors set out in Phelps v. Alameida, 569 F.3d 1120 (9th Cir. 2009), a habeas case, are relevant, but courts must consider all of the relevant circumstances surrounding a specific motion.

The panel examined the district court’s analysis of the six Phelps factors, including the nature of the change in the law, plaintiffs’ diligence in pursuing relief, the parties’ reliance interest in the finality of the case, the delay between the judgment and the Rule 60(b) motion, the relationship between the original judgment and the change in the law, and comity concerns. The panel also examined additional considerations,

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

HENSON V. FIDELITY NAT’L FINANCIAL 3

such as the importance of heeding the intent of the rulings of the federal appellate courts, how best to stay true to the Supreme Court’s reasoning in Microsoft, and the negotiated nature of the voluntary dismissal in this case. The panel concluded that the district court’s denial of plaintiffs’ Rule 60(b)(6) motion was an abuse of discretion because it rested upon an erroneous view of the law as to several significant factors, and granting relief was appropriate. The panel reversed and remanded with directions to grant the Rule 60(b)(6) motion and for further proceedings.

COUNSEL

Cyril V. Smith (argued), Zuckerman Spaeder LLP, Baltimore, Maryland; Taras Kick and Robert J. Dart, The Kick Law Firm APC, Los Angeles, California; for Plaintiffs-Appellants.

Michael J. Gleason (argued), Hahn Loeser & Parks LLP, San San Diego, California, for Defendant-Appellee. 4 HENSON V. FIDELITY NAT’L FINANCIAL

OPINION

TASHIMA, Circuit Judge:

Federal Rule of Civil Procedure 60(b)(6) is a grand reservoir of equitable power that allows courts to grant relief from a final judgment for “any” reason that “justifies relief.” Fed. R. Civ. P. 60(b)(6). In Phelps v. Alameida, 569 F.3d 1120, 1135–40 (9th Cir. 2009), we set out the analysis that courts should employ to guide their discretion when evaluating the merits of a Rule 60(b)(6) motion that seeks relief from the dismissal of a habeas corpus petition on the ground of an intervening change in the law. However, we explicitly left open the question of whether the same Rule 60(b)(6) factors we identified in Phelps are also applicable beyond the habeas corpus context. See id. at 1135 n.19. Confronted with the appeal of a district court’s denial of a Rule 60(b)(6) motion that was also predicated on an intervening change in the law, but in a non-habeas case that entails entirely different circumstances than did Phelps, we now begin to answer that question. While we conclude that many of the Phelps factors are relevant to the Rule 60(b)(6) analysis in the present context, we reemphasize that courts must consider all of the relevant circumstances surrounding the specific motion before the court in order to ensure that justice be done in light of all the facts. See id. at 1133.

I

On September 9, 2013, Plaintiffs-Appellants Melissia Henson and Keith Turner (collectively “Plaintiffs”), two individuals who participated in separate real estate transactions, filed this putative class action lawsuit against Defendant-Appellee Fidelity National Financial, Inc.

HENSON V. FIDELITY NAT’L FINANCIAL 5

(“Fidelity”). Plaintiffs claimed that Fidelity’s practice of receiving payments from three overnight delivery vendors, in exchange for referring document delivery business to those vendors in connection with the settlement of federally related mortgage loans, violated §§ 8(a) and 8(b) of the Real Estate Settlement Procedures Act (“RESPA”), Pub. L. No. 93-533, 88 Stat. 1724, 12 U.S.C. § 2607.

Fidelity moved to dismiss the complaint, and the district court granted the motion as to all claims except for Turner’s claim under RESPA § 8(a). The district court dismissed all of Henson’s claims on statute of limitations grounds, and also dismissed with prejudice both Henson’s and Turner’s claims under RESPA § 8(b) for failure to state a claim. After filing an answer, Fidelity moved for judgment on the pleadings with respect to Turner’s remaining claim under RESPA § 8(a), but the district court denied that motion.

Turner subsequently moved for class certification. A week later, Turner moved in the alternative to continue the hearing on class certification in order to allow discovery on class certification issues. The district court denied both the discovery and class certification motions, finding that class member-specific questions would predominate over any class-wide inquiries.

Plaintiffs then entered into a detailed, negotiated stipulation of dismissal with Fidelity. Relying on Ninth Circuit precedent that permitted a plaintiff to obtain appellate review of certain interlocutory orders, including an order denying class certification, by dismissing any active claims with prejudice, Plaintiffs agreed to voluntarily dismiss the case with prejudice so that they could appeal both the district court’s denial of class certification and the partial grant of the 6 HENSON V. FIDELITY NAT’L FINANCIAL

motion to dismiss. See Omstead v. Dell, Inc., 594 F.3d 1081, 1085 (9th Cir. 2010) (“[A] plaintiff that deems an interlocutory ruling to be so prejudicial as to deserve immediate review . . . has the alternative of dismissing the complaint voluntarily with prejudice.” ) (citation omitted); see also Berger v. Home Depot USA, Inc., 741 F.3d 1061, 1065 (9th Cir. 2014) (“[A] dismissal of an action with prejudice, even when such dismissal is the product of a stipulation, is a sufficiently adverse—and thus appealable—final decision.”). The stipulation provided:

Plaintiffs Keith Turner and Melissia Henson and Defendant Fidelity National Financial, Inc., through their undersigned counsel, hereby stipulate to request dismissal of the above-captioned case with prejudice pursuant to Federal Rule of Civil Procedure 41(a)(2), and further stipulate that:

The parties understand that plaintiffs Turner and Henson will file an appeal to the dismissal in order to appeal certain of the Court’s orders, including its order denying class certification and its order granting (in part) defendant’s motion to dismiss. See Omstead v. Dell, Inc., 594 F.3d 1081, 1085 (9th Cir. 2010). If the orders appealed are affirmed, plaintiffs will take nothing by way of their complaint. By stipulating to this dismissal, [Fidelity] does not agree that such a dismissal establishes appellate jurisdiction and does not waive its ability to challenge appellate jurisdiction.

HENSON V. FIDELITY NAT’L FINANCIAL 7

The parties stipulate that by Plaintiffs voluntarily dismissing their claims with prejudice, the Court’s April 29, 2014 order on Defendant’s Motion for Judgment on the Pleadings is not a final judgment and shall not provide a basis for collateral estoppel against Defendant in subsequent litigation.

Plaintiffs agree to reimburse [Fidelity] $837.96 for certain costs that it has incurred in this action through the date of the filing of this stipulation . . . .

It is further agreed that if Plaintiffs or Plaintiffs’ counsel should re-file the claims at issue in this lawsuit or file a different lawsuit based on the same claims and conduct on behalf of a different Plaintiff(s), any such claims will be filed in the District Court for the Central District of California.

The district court entered the parties’ proposed order and dismissed the complaint with prejudice on the terms provided in the stipulation. Henson and Turner then timely filed separate notices of appeal from the final dismissal order. Both notices of appeal specifically identified the order granting in part Fidelity’s motion to dismiss (as to Henson’s claims and the RESPA § 8(b) claims), and the order denying class certification, as orders included in the final judgment to be reviewed on appeal.

After Henson’s and Turner’s appeals were fully briefed, but before they were calendared for oral argument, the cases were stayed pending the United States Supreme Court’s 8 HENSON V. FIDELITY NAT’L FINANCIAL

decision in Microsoft Corp. v. Baker, because that case appeared poised to address some of the appellate jurisdiction questions raised by Fidelity in Henson’s and Turner’s appeals. In Microsoft, the district court had similarly granted a stipulated motion to dismiss with prejudice. The parties had entered into the stipulation so that the plaintiffs could seek immediate appeal of the district court’s earlier order striking consumers’ class allegations. See Microsoft Corp. v. Baker, 137 S. Ct. 1702, 1706–07, 1711–12 (2017). Under the stipulation, the plaintiffs had “reserved the right to revive their claims should the Court of Appeals reverse the District Court’s certification denial.” Id. at 1707. On appeal, however, Microsoft had challenged appellate jurisdiction, arguing that the voluntary dismissal could not support appellate review of the district court’s interlocutory order striking class allegations. Id. at 1711. The Ninth Circuit rejected Microsoft’s argument, holding that it had jurisdiction to entertain the appeal under 28 U.S.C. § 1291. Id. at 1711–12.

On June 12, 2017, nearly three years after Plaintiffs had filed their stipulated voluntary dismissal, the Supreme Court issued its decision in Microsoft. The Court held that “[p]laintiffs in putative class actions cannot transform a tentative interlocutory order into a final judgment within the meaning of § 1291 simply by dismissing their claims with prejudice—subject, no less, to the right to ‘revive’ those claims if the denial of class certification is reversed on appeal.” Id. at 1715 (citations omitted). The Court explained that the tactic of using a stipulated voluntary dismissal to seek immediate review of an order denying class certification “undermine[s] § 1291’s firm finality principle” and subverts Federal Rule of Civil Procedure 23(f), which gives appellate courts discretion to choose whether to permit immediate

HENSON V. FIDELITY NAT’L FINANCIAL 9

appeals of interlocutory class certification orders. See id. at 1707, 1712–15. In so holding, the Supreme Court reversed the Ninth Circuit’s rule—on which Henson and Turner had relied—that a stipulated voluntary dismissal of a class action was a sufficiently adverse final order, as long as the individual plaintiffs’ cases had not settled, to give rise to appellate jurisdiction to review a district court’s denial of class certification. See Baker v. Microsoft Corp., 797 F.3d 607, 612 (9th Cir. 2015); see also Berger, 741 F.3d at 1065.

In light of Microsoft, Fidelity moved to dismiss both Henson’s and Turner’s appeals. Fidelity argued that the Supreme Court’s rule in Microsoft applied directly and deprived the Ninth Circuit of jurisdiction over Turner’s appeal from the denial of class certification. And Fidelity suggested that the Ninth Circuit also lacked jurisdiction to hear Henson’s appeal from the district court’s dismissal order because the Supreme Court’s rule in Microsoft applied beyond the class certification context and meant that a voluntary dismissal could never constitute a final order under § 1291.

In response, Plaintiffs argued that their case was factually distinguishable from Microsoft because, before the stipulated voluntary dismissal of the single remaining claim, the district court had already involuntarily dismissed all of Henson’s claims and one of Turner’s claims. Plaintiffs maintained that the Ninth Circuit had jurisdiction over their appeals because Microsoft did not overrule cases from “[e]very circuit permit[ting] a plaintiff, in at least some circumstances, voluntarily to dismiss remaining claims or remaining parties from an action to conclude the whole case in the district court and ready it for appeal.” 10 HENSON V. FIDELITY NAT’L FINANCIAL

A Ninth Circuit motions panel subsequently entered identical orders in Henson’s and Turner’s appeals, denying Fidelity’s motions to dismiss and remanding to the district court. The orders stated:

Appellee’s motion to dismiss this appeal . . . is denied. In light of the Supreme Court of the United States’ decision in Microsoft Corp. v. Baker, the court sua sponte remands this case to allow the parties to seek appropriate relief in the district court in the first instance. See Microsoft Corp. v. Baker, 137 S. Ct. 1702, 1715 (2017).1

On remand, Fidelity refused to stipulate that the case should proceed before the district court on Turner’s RESPA § 8(a) claim, which had survived Fidelity’s motion to dismiss and motion for judgment on the pleadings, but which had been voluntarily dismissed pursuant to the stipulation. As a result, Plaintiffs moved to vacate the prior dismissal pursuant to Federal Rule of Civil Procedure 60(b)(6).

In their Rule 60(b) motion, Plaintiffs argued that their circumstances, including Microsoft’s intervening change in the law, entitled them to relief from judgment under the factors set out in Phelps, 569 F.3d at 1135–40, which addressed Rule 60(b)(6) relief in light of an intervening change in the law. In addition, Henson argued that even if the

1 The Supreme Court in Microsoft had also remanded the case “for further proceedings consistent with this opinion.” Microsoft, 137 S. Ct. at 1715; see also Baker v. Microsoft Corp., 884 F.3d 811, 812 (9th Cir. 2018) (remanding to the district court “for further proceedings consistent with the opinion of the Supreme Court”).

HENSON V. FIDELITY NAT’L FINANCIAL 11

district court declined to vacate the dismissal order and revive the entire case, it should nonetheless grant her more limited relief by entering a new final judgment from which she could appeal the court’s earlier involuntary dismissal of her claims, because the Ninth Circuit had not addressed the merits of her prior appeal and it would be manifestly unjust under the circumstances to nullify her right to appeal altogether.

The district court denied the Rule 60(b) motion. Applying the Phelps framework to the intervening change of law argument, the district court ruled that the six factors set out in Phelps did not warrant relief. The district court did not address Henson’s separate argument that denying her relief from judgment would have the effect of unfairly nullifying her right to appeal the district court’s earlier involuntary dismissal of her claims. Plaintiffs timely appealed.

II

We have jurisdiction under 28 U.S.C. § 1291, and we review the denial of a motion for relief from judgment under Rule 60(b) for an abuse of discretion. Phelps, 569 F.3d at 1131. “[T]he district court necessarily abused its discretion if its denial ‘rested upon an erroneous view of the law,’” id. (quoting Faile v. Upjohn Co., 988 F.2d 985, 986–87 (9th Cir. 1993)), or on a “clearly erroneous finding of material fact,” Casey v. Albertson’s, Inc., 362 F.3d 1254, 1257 (9th Cir. 2004). Otherwise, we may not reverse “absent a definite and firm conviction that the district court committed a clear error of judgment in the conclusion it reached upon a weighing of relevant factors.” Valdivia v. Schwarzenegger, 599 F.3d 984, 988 (9th Cir. 2010). 12 HENSON V. FIDELITY NAT’L FINANCIAL

III

Rule 60(b)(6) provides that, “[o]n motion and just terms, the court may relieve a party or its legal representative from a final judgment, order, or proceeding for . . . any . . . reason that justifies relief.”2 Fed. R. Civ. P. 60(b)(6). This clause “gives the district court power to vacate judgments ‘whenever such action is appropriate to accomplish justice.’” United States v. Sparks, 685 F.2d 1128, 1130 (9th Cir. 1982) (quoting Klapprott v. United States, 335 U.S. 601, 615 (1949)); see also Fleming v. Gulf Oil Corp., 547 F.2d 908, 912 (10th Cir. 1977) (explaining that Klapprott, 335 U.S. at 614, shows that, “[l]ike Rule 60(b) generally, clause (6) should be liberally applied to situations not covered by the preceding five clauses so that, giving due regard to the sound interest underlying the finality of judgments, the district court, nevertheless, has power to grant relief from a judgment whenever, under all the surrounding circumstances, such action is appropriate in the furtherance of justice” (quoting 7 Moore’s Federal Practice at 342–43 (2d ed. 1975))). However, “[a] movant seeking relief under Rule 60(b)(6) must show ‘“extraordinary circumstances” justifying the reopening of a final judgment.’” Jones v. Ryan, 733 F.3d 825, 833 (9th Cir. 2013) (quoting Gonzalez v. Crosby, 545 U.S. 524, 535 (2005)).

Plaintiffs argue that the effect on their case of Microsoft’s intervening change in the law presents extraordinary circumstances warranting relief. They point to the fact that

2 As relevant here, a voluntary dismissal qualifies as a “judgment, order, or proceeding from which Rule 60(b) relief can be granted.” Kalt v. Hunter (In re Hunter), 66 F.3d 1002, 1004 (9th Cir. 1995) (citations omitted).

HENSON V. FIDELITY NAT’L FINANCIAL 13

Microsoft’s intervening change in the law caused the Ninth Circuit to remand rather than address the merits of their appeals, and that denying relief from judgment would therefore convert the stipulation—entered into three years before Microsoft and “intended to be a conditional dismissal (to allow appellate review)”—into an “irrevocable” dismissal without any opportunity for appellate review on the merits.3 In other words, denying Rule 60(b) relief would treat the voluntary dismissal, which was intended to facilitate appellate review, as a final adjudication of the entire case, even though there had been and was to be no appellate review

3 Plaintiffs argued in the district court that these circumstances are all the more extraordinary because of their timing. They explained:

The appeal had been fully briefed before the Ninth Circuit months before the Supreme Court granted certiorari in Microsoft. The timing of the Supreme Court’s ruling in Microsoft deprived the Ninth Circuit of jurisdiction over Plaintiffs’ appeal at the absolute worst time. Had it been issued earlier, Plaintiffs would have never agreed to dismiss their claims. Later, and Plaintiffs’ appeal of the District Court’s denial of class certification would have been resolved on the merits. Instead, Plaintiffs were caught in a small window of time, with their appeal pending, where the abrogation of Omstead, Berger, and Baker threatens to deny them their right to appeal the Court’s rulings altogether. In other words, the intervening change in law, and the timing of that change, have worked together to threaten truly catastrophic damage to Plaintiffs’ claims—leaving Plaintiffs deprived of their right to appeal the class certification ruling altogether, as well as depriving Plaintiff Henson of her right to appeal the District Court’s order dismissing all of her claims based upon the statute of limitations. This is exactly the sort of “extraordinary circumstance” warranting relief under Rule 60(b)(6). 14 HENSON V. FIDELITY NAT’L FINANCIAL

of the issues the district court had decided, then or later. Doing so would end the case in a manner that deprives Plaintiffs of any opportunity to obtain appellate review of the district court’s earlier dismissal and class certification orders, because Plaintiffs’ time to appeal the original judgment has run, and an appeal from the denial of a Rule 60(b) motion brings up for review only the denial of that motion, but not the underlying judgment.4 See Fed. R. App. P. 4(a)(1)(A); Harman v. Harper, 7 F.3d 1455, 1458 (9th Cir. 1993). Plaintiffs therefore contend that Microsoft’s change in the law provides a basis under Rule 60(b)(6) for vacating the voluntary dismissal order.

We have previously recognized that a change in the controlling law can— but does not always—provide a sufficient basis for granting relief under Rule 60(b)(6). See Phelps, 569 F.3d at 1132–33. To assess a Rule 60(b)(6) motion “predicated on an intervening change in the law,” a district court must “evaluate the circumstances surrounding the specific motion before the court.” Id. at 1133. This “case-by-case inquiry . . . requires the trial court to intensively balance numerous factors, including the competing policies of the finality of judgments and the incessant command of the court’s conscience that justice be done in light of all the facts.” Id. (quoting Stokes v. Williams, 475 F.3d 732, 736 (6th Cir. 2007)).

In Phelps, in the context of a Rule 60(b) motion that sought relief from the dismissal of a habeas petition on the

4 The Federal Rules provide an exception to this rule by extending the time to appeal from the judgment if the Rule 60 motion is filed within 28 days of entry of judgment, but that exception does not apply here. See Fed. R. App. P. 4(a)(4)(A)(vi).

HENSON V. FIDELITY NAT’L FINANCIAL 15

basis of a change in the law, we identified six factors as being “well-reasoned principles that should guide courts in exercising their discretion under Rule 60(b)(6).” Id. at 1140. We cautioned, however, that our discussion of those factors was not meant to “impose a rigid or exhaustive checklist,” because “‘Rule 60(b)(6) is a grand reservoir of equitable power,’ and it affords courts the discretion and power ‘to vacate judgments whenever such action is appropriate to accomplish justice.’” Id. at 1135 (first quoting Harrell v. DCS Equip. Leasing Corp., 951 F.2d 1453, 1458 (5th Cir. 1992); then quoting Gonzalez, 545 U.S. at 542). Moreover, we noted that “while some of the factors we emphasize here may be useful in contexts other than the one before us, we express no opinion on their applicability vel non beyond the scope of habeas corpus.” Id. at 1135 n.19 (italics omitted); see also Jones, 733 F.3d at 839 (“[The Phelps] factors are particularly useful when, as here, we are asked to apply Rule 60(b)(6) to a rejected petition for habeas corpus.”).

The district court here nevertheless analyzed the six Phelps factors, and only those factors, in the instant non- habeas context to determine whether the circumstances of this case, including the change in the controlling law effected by Microsoft, were sufficiently extraordinary to warrant relief under Rule 60(b)(6). Weighing all those “relevant” factors together, the district court “conclude[d] that the circumstances d[id] not warrant relief pursuant to Rule 60(b)(6).”

Plaintiffs argue on appeal that the district court’s denial of Rule 60(b)(6) relief constituted an abuse of discretion. They contend that the district court erred in its analysis of the Phelps change-in-the-law factors by assuming that Plaintiffs were at fault for their predicament, and therefore by “[giving] 16 HENSON V. FIDELITY NAT’L FINANCIAL

no weight to, or even count[ing] against them, [Plaintiffs’] legitimate reliance on the settled law of this Circuit and [the Ninth Circuit’s] decision to remand rather than to dismiss the appeals.” According to Plaintiffs, the district court’s flawed assumption—contrary to the terms of the parties’ stipulation—that Fidelity was entitled to rely on the finality of the judgment, based on the expectation that Ninth Circuit precedent would be overruled, tainted the district court’s analysis of three of the six Phelps factors.5

To determine whether the district court abused its discretion and to demonstrate the case-by-case inquiry required by Rule 60(b)(6), we examine the district court’s analysis of each of the Phelps factors sereatim. In doing so, we address the question left open by Phelps: the extent to which, and in what manner, those factors may be relevant to assessing a Rule 60(b)(6) motion in a non-habeas context. Relatedly, we also highlight additional considerations, such as the import of Microsoft’s reasoning and the parties’ stipulation, that are not addressed by the Phelps factors, but that nonetheless are relevant to the Rule 60(b)(6) determination in this case. See Phelps, 569 F.3d at 1133 (directing that courts evaluate all the circumstances

5 Henson also argues that, as to her specifically, the district court abused its discretion by denying her relief from judgment without addressing her argument that denying relief would have the effect of stripping her of the right to appeal the court’s earlier involuntary dismissal of her claims, notwithstanding the fact that she had timely appealed to the Ninth Circuit and found herself in this unusual procedural posture through no fault of her own simply because her appeal of the dismissal order became entangled with Turner’s appeal of the class certification denial. This argument may well have merit, but because the change-in-law argument is dispositive as to both Henson and Turner, we need not address it.

HENSON V. FIDELITY NAT’L FINANCIAL 17

surrounding the specific Rule 60(b)(6) motion before the court). We note that particularly where, as here, a district court is faced with applying our previously-discussed Rule 60(b)(6) factors in an entirely new context, the court should assess how that different context might alter the calculus of the factors’ application, and whether those factors adequately capture all of the relevant circumstances. As we emphasized in Phelps, the factors identified in that case were not intended to be a rigid or exhaustive list; a court’s ultimate charge in evaluating a Rule 60(b)(6) motion remains to “intensively balance” all the relevant factors, “including the competing policies of the finality of judgments and the incessant command of the court’s conscience that justice be done in light of all the facts.” See id. at 1133, 1135.

1. The Change in the Law

The first Phelps factor considers the nature of the intervening change in the law. See id. at 1135–36; see also Jones, 733 F.3d at 839. For example, in Phelps we had to decide whether to grant relief from dismissal of a habeas petition where, fifteen months after the petitioner’s unsuccessful appeal of the dismissal became final, the Ninth Circuit favorably changed the law on which the petition’s dismissal had been predicated. 569 F.3d at 1126–27, 1131. We determined that the post-judgment “change in the law . . . did not upset or overturn a settled legal principle” because “the law in our circuit was decidedly unsettled at the time Phelps’ habeas petition was before the district court.” Id. at 1136 (italics omitted). We concluded that this circumstance favored granting relief, because it meant that the dismissal of Phelps’ petition had not been based on a “prevailing interpretation” of the relevant statute, and in fact the change in the law adopted the legal position that Phelps had 18 HENSON V. FIDELITY NAT’L FINANCIAL

unsuccessfully advocated all along in his habeas proceedings. See id.

We see no reason why such equitable considerations related to the nature of the change in the law would not also be relevant to assessing whether Rule 60(b)(6) relief is warranted in a non-habeas context. We note, however, that the circumstances of the instant case are nearly the opposite of those considered in Phelps. Rather than seeking the benefit of a favorable change in the law as Phelps did, Plaintiffs here seek relief from an unfavorable change in the law that Plaintiffs claim has “the disastrous and fundamentally unjust consequence” of leaving them with the burdens of the parties’ negotiated stipulation, but none of the benefits. Because of these contrasting circumstances, the nature of a change in the law that may equitably weigh in favor of granting Rule 60(b)(6) relief here is likely to be different than it was in Phelps. For example, while the fact that the relevant Ninth Circuit law was “decidedly unsettled” cut in favor of relief in Phelps, here the fact that the relevant Ninth Circuit law was relatively settled may cut in favor of relief because it makes Plaintiffs’ reliance on that law more reasonable.

Thus, while we conclude that the change in the law factor is also applicable in non-habeas cases, we note that courts considering this factor should not in rote fashion rely on the conclusion from a different context that any particular type of change in the law favors or disfavors relief. Instead, a district court should weigh whether the specific nature of the change in the law in the case before it makes granting relief more or less justified under all of the circumstances, and should support its conclusion with a reasoned explanation grounded in the equitable considerations raised by the case at bench.

HENSON V. FIDELITY NAT’L FINANCIAL 19

Here, the district court concluded that the nature of Microsoft’s change in the law—namely, overruling precedent, on which Plaintiffs had relied, that was settled in the Ninth Circuit but not nationwide—weighed “slightly against” granting Rule 60(b) relief, because Fidelity had warned that it would challenge appellate jurisdiction, meaning that Plaintiffs were apprised of the risk of an adverse decision, but still chose to voluntarily dismiss their case. The district court further pointed out that the circumstances of the change in the law were “hardly extraordinary” because the change involved a resolution of a circuit split by the Supreme Court. Plaintiffs challenge the district court’s weighing of this factor, arguing that they were entitled to rely on then-settled circuit precedent, and that the district court therefore abused its discretion by treating the voluntary dismissal as a reckless gamble for which Plaintiffs should suffer the consequences despite their reasonable reliance on settled circuit law.

We hold that the district court’s conclusion was an abuse of discretion, as this factor was more properly viewed as neutral or slightly favoring Plaintiffs. It was not unreasonable for the district court to consider the extent to which Plaintiffs took a knowing and calculated risk by relying on Circuit precedent to voluntarily dismiss their case with prejudice, but the district court overlooked important competing considerations in reaching its evaluation.

On the one hand, unlike cases such as Phelps in which the court must decide whether to reopen a case to grant the benefit of a favorable change in the law, here the court had to assess whether to grant relief in the context of an unfavorable change in the law that seemingly prejudiced Plaintiffs by altering the intended effect of their previously-negotiated 20 HENSON V. FIDELITY NAT’L FINANCIAL

stipulation.6 Because granting relief from judgment under Rule 60(b)(6) is largely an equitable decision, it makes sense to consider whether Plaintiffs were blindsided by the change in law and could not have done anything to avoid its prejudicial effect, or whether the predicament they face is the result of their knowingly taking a calculated risk for which they should be held responsible.

To illustrate, analyzing whether Plaintiffs, equitably, should suffer the consequences of the unfavorable change in the law might reasonably involve considering:

(1) the extent to which the precedent upon which Henson and Turner relied was settled in the Circuit, which would make the reliance more reasonable (favors Plaintiffs because there was clear Circuit precedent);

(2) the extent to which Henson and Turner should have known that the law might change in an unfavorable way (neutral or slightly favors Plaintiffs because the Supreme Court had not yet even granted certiorari in Microsoft, and Plaintiffs reasonably relied on well-established Ninth Circuit law. Compare Omstead, 594 F.3d at 1085; Berger, 741 F.3d at 1065, with Microsoft Corp. v. Baker, 135 S. Ct. 890 (2016) (granting petition for a writ of certiorari));

6 In their briefing, Plaintiffs assert that this case presents “a question of correcting the manifest injustice in enforcing the dismissal term of a stipulation when the application of a change in the law to a pending appeal has invalidated the essence of the agreement on which the judgment of dismissal is based.”

HENSON V. FIDELITY NAT’L FINANCIAL 21

(3) the nature of the risk Henson and Turner knowingly took (slightly favors Plaintiffs because even if Henson and Turner knew that they risked an adverse decision regarding their early appeal if Fidelity succeeded in challenging appellate jurisdiction, that does not mean they knowingly risked permanent finality. Fidelity did not state that its position would be that the case was entirely over if there was no appellate jurisdiction; the only understanding incorporated in the stipulation was that finality depended on losing the appeal on the merits;

(4) the extent to which Plaintiffs were forced to rely on the precedent and put themselves in a precarious situation for lack of other reasonable options that would have avoided the risk (favors Fidelity because Plaintiffs had a choice about which procedure to use: entering a voluntary dismissal, petitioning the district court to certify the interlocutory order for appeal pursuant to 28 U.S.C. § 1292(b), requesting permission to file an interlocutory appeal under Federal Rule of Civil Procedure 23(f), or proceeding on Turner’s individual claim and again requesting class certification at a later stage,7 see Microsoft, 137 S. Ct. at 1711); and

7 This array of available choices distinguishes this case from United States v. Foster, 783 F.2d 1082 (D.C. Cir. 1986), on which Plaintiffs rely for their argument that a litigant is justified in relying on existing circuit precedent “even when it is subjected to criticism and (in retrospect) destined to be overturned.” In Foster, the court decided to apply a change in a criminal procedure rule only prospectively, because the criminal defendant’s reliance on the previously settled circuit law was not only “reasonable,” but “the only responsible course,” such that it would be unfair to “change the ground rules after the trial.” Id. at 1086 (emphasis added). 22 HENSON V. FIDELITY NAT’L FINANCIAL

(5) the extent to which the circumstances surrounding the change in law were extraordinary (perhaps neutral or slightly favors Plaintiffs, because although the timing of Microsoft was somewhat extraordinary in its effect on this case, cf. Phelps, 560 F.3d at 1127 (noting that if the intervening change in the law had been made “a few months earlier,” the habeas petition at issue “would have been remanded to the district court for a decision on the merits”), Microsoft was an otherwise routine Supreme Court decision resolving an open circuit split).

In light of these relevant considerations, the district court properly determined that the nature of the change in the law—the abrogation of settled Ninth Circuit precedent by the Supreme Court’s resolution of a circuit split—meant that Plaintiffs knowingly accepted some risk of an unfavorable change in the law.

On the other hand, the district court failed to factor in the consideration that litigants are entitled to rely on well- established circuit law as to which, at the time of reliance, there was no case pending in the Supreme Court either on a petition for writ of certiorari or on the merits. Although there is always the possibility that a well-established rule of this court will be abrogated by the Supreme Court, it is not in this Court’s institutional interest, or that of the litigants before it, to fault lawyers who proceed on the basis that an established procedural rule will remain intact, absent some tangible indication (such as a pending Supreme Court case) that it may not. Additionally, the district court should have considered the specific risk that Plaintiffs knowingly undertook. There is a difference between risking an adverse decision on the issue of appellate jurisdiction and the opportunity for an early appeal, and knowingly risking permanent finality. Weighing

HENSON V. FIDELITY NAT’L FINANCIAL 23

the relevant considerations outlined above, the change-in-the- law factor is neutral or slightly favors granting Rule 60(b) relief. It was an abuse of discretion for the district court to conclude otherwise.

2. Plaintiffs’ Diligence In Pursuing Relief

The second Phelps factor is the petitioner’s exercise of diligence in pursuing his claim for relief. See Phelps, 569 F.3d at 1135–36; see also Jones, 733 F.3d at 839. In Phelps, we concluded that the petitioner was diligent because he had pursued “all possible avenues” in advocating for his legal position, including securing a certificate of appealability, filing petitions for rehearing en banc and certiorari, and filing motions for reconsideration. See Phelps, 569 F.3d at 1136–37. Here, the district court concluded that, although Plaintiffs were diligent in requesting relief from judgment on remand, the diligence factor was neutral because Plaintiffs did not initially seek reconsideration of the class certification denial, or request that the district court certify an interlocutory appeal, before they proceeded to voluntarily dismiss their case. In reaching this conclusion, the district court adopted Fidelity’s interpretation that the diligence factor pertained to “Plaintiffs’ diligence (or lack thereof) in pursuing relief over the entire course of this litigation.” Plaintiffs challenge this interpretation, maintaining that their earlier strategic decision to voluntarily dismiss “has nothing to do with the diligence factor,” and that the district court therefore again inappropriately penalized them for relying on settled circuit precedent when they opted to voluntarily dismiss their case.

Once again, the different circumstances in this case as compared to Phelps affect the application and relevance of 24 HENSON V. FIDELITY NAT’L FINANCIAL

the instant factor. In Phelps, because the petitioner sought the benefit of a favorable change in the law, the fact that the petitioner had been diligent in advancing the legal position that was ultimately adopted by that change in the law was relevant to the equitable considerations implicated by a Rule 60(b)(6) motion. See id.; see also Gonzalez, 545 U.S. at 537–38 (considering diligence in terms of petitioner’s efforts to pursue review of the legal issue on which he later requested relief). Here, however, Plaintiffs of course were not advocating for Microsoft’s unfavorable change in the law, and thus the factor does not quite have the same pertinence. Perhaps the corollary consideration in this context is Plaintiffs’ diligence in attempting to avoid or mitigate the risk of the unfavorable change in the law from which they now seek relief. From that perspective, it was not entirely unreasonable under the circumstances for the district court to consider not only Plaintiffs’ diligence in moving for relief from judgment, but also their failure to seek reconsideration of the class certification denial or request that the district court certify an interlocutory appeal before choosing to voluntarily dismiss their case.8 On the other hand, unlike in Phelps, a motion for reconsideration or for certification of an interlocutory appeal would not have directly addressed the area of law that was later changed by Microsoft. Furthermore, it may be the case that Plaintiffs did not pursue those other avenues because they did not believe they had a

8 We disagree, however, with the district court’s broad phrasing that Plaintiffs’ general diligence over the entire course of the litigation is necessarily relevant to the Rule 60(b)(6) determination. Although the Phelps court discussed a lengthy timeline of events in considering Phelps’ diligence, all of those events related to the legal basis on which Phelps later sought relief from judgment. See Phelps, 569 F.3d at 1136–37. Thus, the diligence to be considered should bear a relationship to the relief requested in the Rule 60(b)(6) motion.

HENSON V. FIDELITY NAT’L FINANCIAL 25

sufficient basis to file such motions; courts should be cautious to count against a party seeking Rule 60(b)(6) relief the failure to file certain motions, unless the court can determine that filing such a motion would not have been an utter exercise in futility.

Thus, while the district court did not abuse its discretion in assessing the diligence factor and concluding that it was neutral, we note that under the circumstances the factor appears to be less salient to the overall Rule 60(b)(6) equitable balancing than it was in Phelps.

3. Reliance Interest in the Finality of the Case

The third factor we considered in Phelps is whether granting the Rule 60(b) motion for relief from judgment would upset “the parties’ reliance interest in the finality of the case.” Phelps, 569 F.3d at 1137; see also Jones, 733 F.3d at 839–40. In this case, the district court concluded that the parties’ interest in finality weighed against granting relief because there could be “no question . . . of the parties’ intention regarding finality in entering into the stipulation to dismiss this case.” Plaintiffs challenge this conclusion, arguing that Fidelity had no reliance interest in the finality of the stipulation-based judgment because the stipulation explicitly contemplated the potential for subsequent litigation, and stated that Plaintiffs would take nothing only if both of the appealed district court orders were affirmed, which never happened.

We agree with Plaintiffs and hold that the district court misconstrued this factor and overvalued Fidelity’s reliance interest in the finality of the judgment. First, the district court only considered Fidelity’s abstract interest in finality, but not, 26 HENSON V. FIDELITY NAT’L FINANCIAL

as Phelps instructed, whether Fidelity had any reliance interest in the finality of the judgment. In Phelps, we explained that courts should consider whether “the final judgment being challenged has caused one or more of the parties to change his legal position in reliance on that judgment” such that granting the motion for relief would “undo the past, executed effects of the judgment.” Phelps, 569 F.3d at 1137–38 (citation omitted). For example, a party might have a reliance interest in the finality of a judgment “when a judgment conveys land from one party to another and the prevailing party ‘enter[s] upon the land and install[s] pipes and appurtenances,’” or “when a judgment affords a federal habeas petitioner a new trial that results in the eventual dismissal of the charges and his release from custody.” Id. (alterations in original) (italics omitted) (quoting Ritter v. Smith, 811 F.2d 1398, 1401 (11th Cir. 1987)). Here, however, there is no evidence that Fidelity “has relied in such a fashion on the finality of the district court’s dismissal.” Id. at 1138. As in Phelps, “[t]here are no ‘past effects’ of the judgment that would be disturbed if the case were reopened for consideration of the merits . . . . Instead, the parties would simply pick up where they left off.” Id. Thus, as in Phelps, this factor should “weigh[] heavily in [Plaintiffs’] favor.” Id.

Furthermore, the finality interest in this case is even weaker than that in Phelps because, here, Plaintiffs’ appeal had not been finally decided before Microsoft changed the law. See id. at 1126 (explaining that Phelps’ appeal had become final fifteen months before the change in the law that prompted his Rule 60(b)(6) motion). Because the present case was still pending on appeal when Microsoft was decided, and was then remanded to the district court for further proceedings, Fidelity could not reasonably have believed that

HENSON V. FIDELITY NAT’L FINANCIAL 27

the dismissal was immutably final. Moreover, the dismissal itself belies any reliance interest in finality, because the dismissal was explicitly linked to an attempt to expedite the appeal. The parties’ stipulation contemplated that Plaintiffs’ claims would be revived and that litigation would continue in the district court if the Ninth Circuit reversed either or both of the appealed orders.9 And, although Fidelity reserved the right to argue that the immediate appeal was improper, it could not have relied on prevailing on that challenge, as Ninth Circuit law at the time was squarely against its position.

It was this very opportunity for revival of claims and subsequent litigation that led the Court in Microsoft to conclude that this type of conditional voluntary dismissal is not, in fact, “final.” See Microsoft, 137 S. Ct. at 1715. Plaintiffs persuasively argue that, “[i]f class certification orders are not appealable because a voluntary dismissal intended to be subject to revival is not really a final judgment, then there is little finality interest in the dismissal and there is every reason to allow vacatur of such pre- Microsoft dismissals . . . .” Thus, even after Microsoft, there

9 Not only did the stipulation contemplate continued litigation if Plaintiffs succeeded in their appeals, but Plaintiffs argue that, even if Fidelity prevailed on its jurisdictional challenge and succeeded in having the appeals dismissed, Fidelity could not have expected that the stipulated dismissal order would become final, because the stipulation provided only that Plaintiffs would take nothing if both of the district court’s orders were affirmed on appeal. The text of the stipulation further belies a finality interest because it explicitly states that, “by Plaintiffs voluntarily dismissing their claims with prejudice, the Court’s April 29, 2014 order on Defendant’s Motion for Judgment on the Pleadings is not a final judgment and shall not provide a basis for collateral estoppel against Defendant in subsequent litigation.” 28 HENSON V. FIDELITY NAT’L FINANCIAL

was no reason to anticipate that the voluntary dismissal would be treated as final, without either an opportunity to revive Turner’s voluntarily dismissed claim in the district court, or appellate review of the involuntary dismissal of Henson’s claims and the class certification order.

The district court’s contrary analysis of Fidelity’s reliance interest in the finality of the dismissal therefore ignored key considerations, failed to follow our guidance in Phelps, and was unsupported by the facts. The district court’s conclusion that this factor “weighs against granting Rule 60(b) relief” was an abuse of discretion. Fidelity has not shown any reliance interest in the finality of the judgment, and the appeal of the judgment was still pending when Microsoft was decided; this was not a case in which a party relied on the finality of a judgment after an appeal had become final and litigation had ended months or even years before a change in the law prompted a Rule 60(b) motion. As a result, whether relief from the judgment would upset “the parties’ reliance interest in the finality of the case” weighs heavily in favor of granting relief. See Phelps, 569 F.3d at 1137.

4. Delay Between the Judgment and the Rule 60(b) Motion

The fourth Phelps factor “examines the ‘delay between the finality of the judgment and the motion for Rule 60(b)(6) relief.’” Phelps, 569 F.3d at 1138 (quoting Ritter, 811 F.2d at 1402). “This factor stands for the ‘principle that a change in the law should not indefinitely render preexisting judgments subject to potential challenge.’” Jones, 733 F.3d at 840 (quoting Phelps, 569 F.3d at 1138). The district court found that the delay factor was neutral because, although the Rule 60(b) motion was filed 3.5 years after the voluntary

HENSON V. FIDELITY NAT’L FINANCIAL 29

dismissal, “the parties were actively involved in the appellate proceedings and were seeking resolution before the Ninth Circuit in [the] time immediately after the Court dismissed the case.” Plaintiffs do not challenge this determination, but we make one brief correction. The district court’s use of the date on which the voluntary dismissal was entered in the district court, rather than when any appeal of that dismissal became final, does not follow the guidance we set forth in Phelps. See Phelps, 569 F.3d at 1138 n.21 (“[W]e consider the length of time between when the original judgment . . . became final after appeal, and the time at which [a party] filed [its] first motion [for Rule 60(b)(6) relief].” (emphasis added) (citing Fed. R. App. P. 41)). In this case, the judgment never truly became final on appeal, and in any case, Plaintiffs moved for relief shortly after the Ninth Circuit remanded the case to the district court. As a result, when properly applied, the delay factor weighs in favor of granting Rule 60(b)(6) relief. See id. at 1138.

5. Relationship Between the Original Judgment and the Change in the Law

The fifth Phelps factor “looks to the closeness of the relationship between the decision resulting in the original judgment and the subsequent decision that represents a change in the law.” Jones, 733 F.3d at 840 (citing Phelps, 569 F.3d at 1138–39). As we explained in Phelps, this factor recognizes the ever-changing nature of the law and seeks to measure whether any particular change is extraordinary under the circumstances as a result of a direct relationship to the original judgment:

[T]his factor is designed to recognize that the law is regularly evolving. The foundation of 30 HENSON V. FIDELITY NAT’L FINANCIAL

the American judicial system that sets it apart from many regimes across the world is its common law heritage, which is immanent in judicial interpretations of legal texts ranging from statutes to judicial opinions to the Constitution itself. Given this tradition, legal rules and principles inevitably shift and evolve over time, but the mere fact that they do so cannot upset all final judgments that have predated any specific change in the law. Rather, the nature of that change is important. Accordingly, [this] factor directs courts to examine closely the original and intervening decisions at issue in a particular motion for [Rule 60(b)] relief predicated on an intervening change in the law: if there is “a close connection between the two cases, the court [will be more likely to] f[i]nd the circumstances sufficiently extraordinary to justify disturbing the finality of the [original] judgment.”

Phelps, 569 F.3d at 1139 (footnote omitted) (alterations in quote in original) (quoting Ritter, 811 F.2d at 1402). The district court found that the “close connection between Plaintiffs’ underlying case and [Microsoft]” weighed in favor of granting relief, and Plaintiffs do not challenge this determination. We agree with the district court both that there is a close connection between the original judgment and Microsoft, because the voluntary dismissal was explicitly predicated on the law that Microsoft changed, and that this factor therefore weighs in favor of Rule 60(b)(6) relief.

HENSON V. FIDELITY NAT’L FINANCIAL 31

6. Concerns of Comity

The sixth Phelps factor considers concerns of comity. Jones, 733 F.3d at 840 (citing Phelps, 569 F.3d at 1139). In Phelps, we ruled that this factor cut in favor of relief because the dismissal of Phelps’ habeas petition had been on procedural grounds, which meant that granting relief from the dismissal would not upend the comity principle. See Phelps, 569 F.3d at 1139 (“[I]n the context of Rule 60(b)(6), we need not be concerned about upsetting the comity principle when a petitioner seeks reconsideration not of a judgment on the merits of his habeas petition, but rather of an erroneous judgment that prevented the court from ever reaching the merits of that petition.”).10

The district court found that the comity factor weighed against granting Rule 60(b) relief because the predicament in which Plaintiffs found themselves was the result of their free, calculated, and deliberate trial strategy. Plaintiffs challenge this analysis, claiming that the district court again faulted them for relying on Circuit law.

We agree that the district court erred in its analysis of the comity factor; the district court’s decision that this factor weighed against relief rested upon an erroneous view of what considerations are relevant to the principle of comity. In particular, it is unclear that there is any relationship between the fact that Plaintiffs may have made a calculated decision

10 Furthermore, we noted that the habeas statute was aimed at ensuring that federal courts could protect constitutional rights, and that “[t]he delicate principles of comity governing the interaction between coordinate sovereign judicial systems do not require federal courts to abdicate their role as vigilant protectors of federal rights.” Id. at 1139. 32 HENSON V. FIDELITY NAT’L FINANCIAL

to voluntarily dismiss their case despite the potential consequences and principles of comity. The district court cited no Ninth Circuit precedent suggesting that this is an appropriate consideration under the comity factor, nor do we see why it is relevant.11 In fact, considerations of comity “between the independently sovereign state and federal judiciaries” that we discussed in Phelps do not apply here at all, because this case does not involve a federal habeas petition that challenges a state conviction. See Phelps, 569 F.3d at 1139 (“[I]n applying Rule 60(b)(6) to cases involving petitions for habeas corpus, judges must bear in mind that [a] federal court’s grant of a writ of habeas corpus . . . is always a serious matter implicating considerations of comity.” (internal quotation marks and italics omitted)). As a result, the district court’s finding that the comity factor weighed against relief was an abuse of discretion. The district court instead should have skipped the Phelps comity factor as inapplicable.

7. Additional Considerations

Instead of comity between the federal and state courts, the district court could have considered a somewhat analogous issue that is more relevant here, namely, the importance of heeding the intent of the rulings of federal appellate courts. Cf. id. at 1133, 1135 (explaining that the Phelps factors do not “impose a rigid or exhaustive checklist”—a district court must “evaluate the circumstances surrounding the specific motion before the court” to ensure that “justice [is] done in

11 As discussed above, however, the extent to which Plaintiffs are at fault for their own predicament as a result of taking a calculated risk may be an appropriate consideration in this case under the factor that considers the nature of the change in the law.

HENSON V. FIDELITY NAT’L FINANCIAL 33

light of all the facts”). For example, the district court could have taken account of the fact that, when the Ninth Circuit denied the motions to dismiss Plaintiffs’ appeals and instead remanded to the district court for appropriate relief in the first instance, it is unlikely that the Ninth Circuit intended for the district court to deny all relief and also refuse to re-enter final judgment, thereby denying Henson and Turner any opportunity to have their previously remanded appeals heard on the merits.

Similarly, it would have been relevant for the district court to consider how best to stay true to the Supreme Court’s reasoning in Microsoft. Nothing in Microsoft suggests that a litigant who agreed to a conditional dismissal to obtain appellate review of a class certification ruling should have her claims unconditionally dismissed, rather than being restored to the status quo prior to entering the stipulation.12 In fact, it is somewhat paradoxical that, although the Court in Microsoft ruled that a conditional voluntary dismissal like the one in this case is not a final judgment under § 1291, see Microsoft, 137 S. Ct. at 1715, the district court effectively treated the voluntary dismissal as a final, irrevocable judgment. This leaves Plaintiffs in a catch twenty-two because, under Microsoft, the dismissal was not a final judgment from which

12 The Supreme Court in Microsoft did not give any explicit indication of how the case should proceed, if at all, following the Court’s decision that there was no appellate jurisdiction to review the class certification order. See Microsoft, 137 S. Ct. at 1715. That is, the Supreme Court did not comment about whether the stipulated dismissal would remain in force, and thus paradoxically become “final,” or whether the district court should pick up where the case left off before the stipulated dismissal was entered. However, the Supreme Court did remand the case “for further proceedings consistent with this opinion,” suggesting that it did not view the voluntary conditional dismissal as being final and irrevocable. See id. 34 HENSON V. FIDELITY NAT’L FINANCIAL

Turner could appeal the denial of class certification, but in the district court, the voluntary dismissal was treated as having finally ended the case. The interest in avoiding such a contradiction with the Court’s reasoning in Microsoft weighs in favor of granting relief under Rule 60(b)(6).

Finally, the negotiated nature of the voluntary dismissal presents a circumstance unique to this case that is relevant to the determination of whether Rule 60(b)(6) relief from the judgment that was predicated on that stipulation is warranted. The district court did not address directly the explicit text of the stipulation to determine whether the terms of the agreement negotiated by the parties contemplated that the dismissal would become final even if the appellate court found that it lacked jurisdiction and therefore did not reach the merits of Plaintiffs’ appeals. In particular, while the stipulation does not set forth any specific agreement about how the dismissal would be treated in the event of a successful challenge to appellate jurisdiction, the stipulation does provide that “plaintiffs will take nothing by way of their complaint” if “the orders appealed are affirmed.” (Emphasis added.) Because Plaintiffs appeals were remanded without a decision on the merits, the district court’s orders were never affirmed. Thus, it would have been relevant to the equities to consider whether, under the strict text of the stipulation, Plaintiffs agreed to dismiss their case with prejudice in the current scenario, or whether giving effect to the terms of the negotiated agreement supported granting relief.13

13 We do not wade into the fact-finding required to answer this question, as it is not necessary to decide this case. We only point out the relevance of this consideration to guide future courts faced with Rule 60(b)(6) motions.

HENSON V. FIDELITY NAT’L FINANCIAL 35

8. Weighing All Relevant Considerations

“Weighing all the relevant factors together and evaluating the circumstances of this case,” the district court ultimately concluded that the circumstances did not warrant relief under Rule 60(b)(6). As detailed above, of the six Phelps factors, the district court found that the reliance interest and comity factors weighed against granting relief, the nature of the change in the law weighed slightly against granting relief, the diligence and delay factors were neutral, and the relative closeness between the judgment and the change in the law weighed in favor of relief.

However, as discussed above, the district court’s analysis of the change-in-the-law, reliance interest, and comity factors—those which it found weighed against relief—rested upon an incorrect view of the law. The change-in-the-law factor is neutral or slightly favors relief because Plaintiffs relied on well-established circuit law and did not knowingly risk permanent finality, even though they knew Fidelity would challenge appellate jurisdiction and they could have taken a different course of action. Properly interpreted and applied, the factor considering the parties’ reliance interest in the finality of the judgment weighs heavily in favor of Rule 60(b)(6) relief, because the change in the law occurred while the appeal was pending, before the judgment became final and before Fidelity developed any reliance interest in the finality of the judgment. In addition, because the instant case does not implicate principles of comity between state and federal courts, the comity factor is inapplicable and does not, as the district court found, weigh against granting relief.

Weighing all the Phelps factors, correctly analyzed, it is clear that all the relevant circumstances in this case heavily 36 HENSON V. FIDELITY NAT’L FINANCIAL

tip the scale in favor of granting Rule 60(b)(6) relief. Although by the district court’s reasonable calculus the nature in the change of the law factor weighed “slightly against” granting relief, all the other factors are neutral or weigh in favor of granting relief. Together, the circumstances of this case are therefore sufficiently “extraordinary” that granting relief from judgment here “is appropriate to accomplish justice.” Phelps, 569 F.3d at 1135 (quoting Gonzalez, 545 U.S. at 542).

IV

Because the district court’s denial of Plaintiffs’ Rule 60(b)(6) motion rested upon an erroneous view of the law as to several significant factors, the district court’s ruling was an abuse of discretion. See Phelps, 569 F.3d at 1131. Moreover, in light of all the circumstances of this case, we are left with “a definite and firm conviction that the district court committed a clear error of judgment in the conclusion it reached upon a weighing of relevant factors.” See Valdivia, 599 F.3d at 988.

We therefore reverse the district court’s denial of Plaintiffs’ Rule 60(b)(6) motion for relief from judgment, and remand with directions to grant the Rule 60(b)(6) motion and for further proceedings.

REVERSED and REMANDED with directions.