Tamarin Lindenberg v. Jackson Nat'l Life Ins. Co.

RECOMMENDED FOR FULL-TEXT PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 19a0056p.06 UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT TAMARIN LINDENBERG, individually and as natural ┐ guardian of her minor children ZTL and SML, │ Plaintiff-Appellee/Cross-Appellant, │ │ v. │ > Nos. 17-6034/6079 │ JACKSON NATIONAL LIFE INSURANCE COMPANY, │ Defendant-Appellant/Cross-Appellee, │ │ STATE OF TENNESSEE, │ Intervenor-Appellee. │ │ ┘ Appeal from the United States District Court for the Western District of Tennessee at Memphis. No. 2:13-cv-02657—Jon Phipps McCalla, District Judge. Decided and Filed: March 28, 2019 Before: CLAY, STRANCH, and LARSEN, Circuit Judges. _________________ COUNSEL ON PETITIONS FOR REHEARING EN BANC: Joseph Ahillen, OFFICE OF THE ATTORNEY GENERAL OF TENNESSEE, Nashville, Tennessee, for Intervenor Appellee. Daniel W. Van Horn, Gadson W. Perry, BUTLER SNOW LLP, Memphis, Tennessee, for Appellant/Cross-Appellee. ON RESPONSE IN OPPOSITION: Molly Glover, Charles S. Higgins, BURCH, PORTER & JOHNSON, PLLC, Memphis, Tennessee, for Appellee/Cross- Appellant. ON BRIEF: Cary Silverman, SHOOK, HARDY & BACON L.L.P., Washington, D.C., for Amicus Curiae. CLAY, J. (pp. 3–5), delivered a separate opinion concurring in the denial of rehearing en banc in which STRANCH, J., joined. BUSH, J. (pp. 6–15), delivered a separate opinion dissenting from the denial of rehearing en banc. NALBANDIAN, J. (pp. 16–18), delivered a separate statement regarding the denial of rehearing en banc in which THAPAR, BUSH, and LARSEN, JJ., joined. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 2 _________________ ORDER _________________ The court received petitions for rehearing en banc. The original panel has reviewed the petitions for rehearing and concludes that the issues raised in the petitions were fully considered upon the original submission and decision. The petitions then were circulated to the full court. Less than a majority of the judges voted in favor of rehearing en banc. Therefore, the petitions are denied. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 3 _________________ CONCURRENCE _________________ CLAY, Circuit Judge, concurring in the denial of rehearing en banc. It is incredulous that some of my colleagues would have this Court establish rigid, mechanical, and unflinching criteria for certification to state courts in lieu of our established practice of trusting panels to exercise their experience, discretion, and best judgment to determine when certification is appropriate. The Supreme Court has recognized that the decision of whether to certify “rests in the sound discretion of the federal court.” Lehman Bros. v. Schein, 416 U.S. 386, 391 (1974). This approach recognizes that federal courts weigh numerous competing considerations when determining whether to certify. Of course, certification “is most appropriate when the question is new and state law is unsettled.” Transamerica Ins. Co. v. Duro Bag Mfg. Co., 50 F.3d 370, 372 (6th Cir. 1995). But federal courts may also consider factors such as comity, cooperative federalism, and judicial economy. See Rutherford v. Columbia Gas, 575 F.3d 616, 628 (6th Cir. 2009) (Clay, J., dissenting). These multifarious considerations cannot be reduced to a checklist or simple mathematical formula, as my colleagues would have us believe. Certainly, the decision concerning whether to certify is not always straightforward. Resolving requests for certification often entails a difficult analysis of several competing considerations. But the mere fact that ceding our discretion would be easier, and perhaps even more expedient, is not an adequate reason for us to shirk from our judicial obligations. Rather than adopt a rigid formula that answers the question for us of when to certify, we should trust ourselves and our own judgment, and that of our capable colleagues on this Court, to exercise our discretion wisely after considering the unique circumstances and considerations that may be present in a given case. On the surface, my colleagues purport to take issue with this Court’s procedure for certification. But, on a more fundamental level, they appear to challenge this Court’s very jurisdiction to decide matters of state law in diversity cases, a power that emanates from Article Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 4 III and which Congress has codified in 28 U.S.C. § 1332. It is an “undisputed constitutional principle that Congress, and not the Judiciary, defines the scope of federal jurisdiction within the constitutionally permissible bounds.” New Orleans Pub. Serv., Inc. v. Council of City of New Orleans, 491 U.S. 350, 359 (1989). Federal courts have a “virtually unflagging obligation . . . to exercise the jurisdiction given them,” Colo. River Water Conservation Dist. v. United States, 424 U.S. 800, 817 (1976), and “have no more right to decline the exercise of jurisdiction which is given, than to usurp that which is not given,” Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 404 (1821). Thus, when diversity jurisdiction is properly invoked, federal courts have a “duty . . . to decide questions of state law whenever necessary to the rendition of a judgment.” Meredith v. City of Winter Haven, 320 U.S. 228, 234 (1943); see Burgess v. Seligman, 107 U.S. 20, 33 (1883) (explaining that “[t]he federal courts have an independent jurisdiction in the administration of state laws, co-ordinate with, and not subordinate to, that of the state courts”). And “it is still the duty of the federal courts, where the state law supplies the rule of decision, to ascertain and apply that law even though it has not been expounded by the highest court of the State.” Fid. Union Tr. Co. v. Field, 311 U.S. 169, 177 (1940) (call number omitted). To the extent that my colleagues wish to circumvent Congress’s directive that we decide state law issues in diversity cases, they ignore their constitutional obligation to exercise the jurisdiction conferred by Congress. To the extent that they would create new rules to infringe upon jurisdictional prerequisites for referral of cases to state courts, they engage in judicial activism in contravention of Congress’s prerogative to define the jurisdiction of federal courts. Even if they doubt the wisdom of the scope of federal court jurisdiction as it currently stands, that does not justify their oblique attempt to circumscribe federal jurisdiction by impeding or eliminating our discretion to decide when certification is appropriate. Moreover, my colleagues’ concerns are unfounded. When this Court sits in diversity, we apply state law, see Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938), and therefore act as “only another court of the State.” Guar. Tr. Co. v. York, 326 U.S. 99, 108 (1945). When required to do so, we predict state law, but we do not devise it. In many instances, federal courts are more than capable of correctly deciding state law issues without certifying them to the state’s highest court. In those cases, certification would serve little purpose other than to needlessly delay resolution of Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 5 the ultimate issues in the case. Some state courts frequently take an extended period of time to decide whether to address certified questions, only to ultimately reject the certification request and refuse to answer the questions for which we have sought guidance. I am personally aware of multiple instances in which state courts in our circuit have sat on certification requests for up to a year or more, only to deny the requests without taking any action. Of course, certification may be warranted in some cases. But we should not create a mechanical rule that would require us to certify issues in circumstances where our sound discretion and judicial experience would not direct us to seek certification. Finally, in arguing for certification here, my colleagues have taken my statements from Rutherford, 575 F.3d 616, out of context. In Rutherford, this Court faced the question of whether the equitable doctrines of estoppel, laches, and waiver applied to an express easement under Ohio law. Id. at 618. The majority declined to certify the question, holding that the outcome was “largely controlled,” id., by our recent decision in Andrews v. Columbia Gas Transmission Corp., 544 F.3d 618 (6th Cir. 2008). I dissented, arguing that stare decisis did not preclude certification because Andrews “relied almost exclusively” on a single intermediate court case that was likely wrongly decided, and because Andrews failed to discuss, much less distinguish, several cases from the Ohio Supreme Court that indicated that body would likely reach the opposite conclusion as the Andrews panel. Rutherford, 575 F.3d at 620–21. (Clay, J., dissenting). Rutherford is the inverse of this case. There, the panel privileged federal precedent over state decisions. Here, the panel stands accused of doing the opposite. Ultimately, this panel properly considered the circumstances of the case. A jury found in Plaintiff’s favor in December 2014. Three and a half years later, when this appeal was briefed and argued, neither party moved for certification. The State as intervenor did so only in a footnote, and only with regard to “the constitutional questions.” But of course, the district court had already certified the constitutional questions to the Tennessee Supreme Court. That body, after waiting approximately seven months, declined to answer. Under these circumstances, it was well within our discretion to elect against a second certification attempt. Accordingly, I concur in the denial of rehearing en banc. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 6 _________________ DISSENT _________________ JOHN K. BUSH, Circuit Judge, dissenting from the denial of rehearing en banc. This case presents an unusually strong set of reasons for certification to the Tennessee Supreme Court of state-law questions. It also highlights the need for our circuit to clarify and define certification standards to address the constitutional federalism considerations that underlie Erie Railroad Co. v. Tompkins, 304 U.S. 64 (1938). I therefore respectfully dissent from the denial of rehearing. To explain the reasons for my dissent, some history first is in order. The “judicial Power” of Article III extends to, among other categories, “Controversies . . . between Citizens of different States” and between state citizens and foreign citizens or subjects. U.S. Const. art. III, § 2. A common Antifederalist criticism of the United States Constitution was that it granted too much power to federal courts at the expense of states generally and state judiciaries in particular.1 Responding to Antifederalist criticism, Federalists defended federal-court authority to hear such cases—what would be called diversity jurisdiction—as a way to give out-of-state or foreign litigants a fair shake in court. Federal courts were thought to have less bias than state courts in favor of in-state parties, and diversity jurisdiction was designed to address the perceived unfairness of state courts.2 Diversity jurisdiction did not violate federalism principles 1 See, e.g., Address by a Plebian (1788), reprinted in The Essential Antifederalist 63, 71 (W.B. Allen & Gordon Lloyd eds., 2d ed. 2002) (“The opposers to the constitution have said that it is dangerous because the judicial power may extend to many cases which ought to be reserved to the decision of the State courts . . . .”); George Mason, Objections (1787), reprinted in The Essential Antifederalist, supra, at 16, 17 (“The judiciary of the United States is so constructed and extended as to absorb and destroy the judiciaries of the several states . . . .”); Centinel Letter 1 (1787), reprinted in The Essential Antifederalist, supra, at 96, 101 (expressing concern that “it is more than probable that the state judicatories would be wholly superseded”); Brutus, Essay XI, reprinted in The Essential Antifederalist, supra, at 185, 188 (“The judicial power will operate to effect in the most certain, but yet silent and imperceptible manner, what is evidently the tendency of the constitution: an entire subversion of the legislative, executive and judicial powers of the individual states.”); Brutus, Essay XV, reprinted in The Essential Antifederalist, supra, at 196, 199 (“Perhaps nothing could have been better conceived to facilitate the abolition of the state governments than the constitution of the judicial.”). 2 See, e.g., Guar. Tr. Co. v. York, 326 U.S. 99, 111 (1945) (“The Framers of the Constitution, according to [Chief Justice John] Marshall, entertained ‘apprehensions’ lest distant suitors be subjected to local bias in State courts, or, at least, viewed with ‘indulgence the possible fears and apprehensions’ of such suitors.” (quoting Bank of the U.S. v. Deveaux, 9 U.S. 61, 87 (1809))); Erie, 304 U.S. at 74 (“Diversity of citizenship jurisdiction was Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 7 because it did not deputize federal courts to apply a different law than would have applied in the case had it been decided in state court.3 This understanding underlay Section 34 of the Judiciary Act of 1789, enacted by the First Congress, which provided that “the laws of the several states, except where the [C]onstitution, treaties or statutes of the United States shall otherwise require or provide, shall be regarded as rules of decision in trials at common law in the courts of the United States, in cases where they apply.” Judiciary Act of 1789, ch. 20, § 34, 1 Stat. 73, 92 (emphasis added). In the Erie decision, the Supreme Court confirmed that “laws of the several states” includes the decisions of the state courts as well as enacted statutes and other sources of state law. See Erie, 304 U.S. at 78. This holding is derived from constitutional principles of federalism. See id. at 77–78; Boyle v. United Techs. Corp., 487 U.S. 500, 517 (1988) (“Erie was deeply rooted in notions of federalism.”). Therefore, under Erie, federal courts sitting in diversity must make an informed assessment of what state law is by looking to state courts’ decisions as well as to state statutes and state constitutions.4 However, a federal judge’s assessment of state law “cannot escape being a forecast rather than a determination” if the state courts have not yet definitively resolved an issue. R.R. conferred in order to prevent apprehended discrimination in state courts against those not citizens of the state.”); Anthony J. Bellia Jr. & Bradford R. Clark, The Law of Nations and the United States Constitution 25–26 (2017) (noting that “James Madison argued strongly in favor of diversity jurisdiction at the Virginia ratifying convention on the ground that ‘foreigners cannot get justice done them in [state] courts, and this has prevented many wealthy gentlemen from trading or residing among us’” (quoting 3 The Debates in the Several State Conventions on the Adoption of the Federal Constitution 583 (Jonathan Elliot ed., 2d ed. 1901))). 3 See William A. Fletcher, The General Common Law and Section 34 of the Judiciary Act of 1789: The Example of Marine Insurance, 97 Harv. L. Rev. 1513, 1514 (1984); see generally Bellia & Clark, supra note 2, at 28 (noting that in early cases involving general commercial law, called the “law merchant,” “both federal and state courts deciding commercial cases ‘considered themselves to be deciding questions under a general law merchant that was neither distinctively state nor federal’” (quoting Fletcher, supra, at 1554)). 4 See, e.g., Carnation Co. v. T.U. Parks Constr. Co., 816 F.2d 1099, 1100 (6th Cir. 1987) (“In Erie Railroad, the Court held that the federal district court was required, under a proper interpretation of the Judiciary Act of 1789 and, indeed, by the Constitution, to apply the law of the state in which it sits in resolving questions of substantive law.”); Bradford R. Clark, Ascertaining the Laws of the Several States: Positivism and Judicial Federalism After Erie, 145 U. Pa. L. Rev. 1459, 1551 (1997) (noting that “the principles of judicial federalism recognized in [the Erie decision] preclude federal courts from ‘declar[ing] substantive rules of common law applicable in a State’” and that diversity jurisdiction under Erie is not intended “to provide an alternative source of law” (quoting Erie, 304 U.S. at 78)). Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 8 Comm’n v. Pullman Co., 312 U.S. 496, 499 (1941). A federal court might make an inaccurate forecast and later be proved wrong if the state supreme court decides the issue the other way. Probably in response to the problem of inaccurate federal-court guesses, Florida in 1945 was the first state to enact a certification procedure, whereby the state high court could accept and decide questions of state law necessary to the decision of lawsuits pending in federal courts of appeal. See Clark, supra note 4, at 1545. The Supreme Court recognized the procedure for the first time in Clay v. Sun Insurance Office Ltd., 363 U.S. 207, 212 (1960). Today, all of the states except North Carolina have certification procedures.5 As certification became mainstream, the Supreme Court repeatedly commented favorably on the procedure and sometimes instructed lower courts to consider certification on remand. See, e.g., Arizonans for Official English v. Arizona, 520 U.S. 43, 79 (1997); Lehman Bros. v. Schein, 416 U.S. 386, 391–92 (1974); see also Minn. Voters All. v. Mansky, 138 S. Ct. 1876, 1893 (2018) (Sotomayor, J., dissenting). Because the Supreme Court has not announced concrete rules to govern lower federal courts in deciding whether to certify questions, those lower federal courts have had to make their own guidelines. Our circuit standards do nothing to narrow the discretion left to each district judge and Sixth Circuit panel. See Transamerica Ins. Co. v. Duro Bag Mfg. Co., 50 F.3d 370, 372 (6th Cir. 1995) (stating only that certification may be appropriate where a question of state law is “new” and “unsettled”). This lack of direction creates the potential for intra-circuit conflict as to when certification is appropriate and reduces predictability. The lack of predictability convinces me that this circuit should have more concrete standards to guide its decisionmaking in these recurring situations; what is more, this was the ideal case in which to begin delineating those standards. Specifically, we should seriously consider establishing a presumption in favor of certification where, as here, the state supreme court has not settled the issue; a prior published panel decision has addressed the issue but the current panel is inclined to disagree with the prior decision; and neither party objects to certification. Sixth Circuit case law states that certification is appropriate if the question of state law is “new” and “unsettled,” but that case law unfortunately fails to provide guidance in a recurring set 5 See Gregory L. Acquaviva, The Certification of Unsettled Questions of State Law to State High Courts: The Third Circuit’s Experience, 115 Penn St. L. Rev. 377, 384–85 (2010). Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 9 of cases. Transamerica, 50 F.3d at 372. Those are the cases in which the question may not be new in the sense that no court has addressed it, but a decision from a federal court has the foreseeable potential to create a different state-law rule than what the state supreme court would have produced. This is such a case. A previous decision of this circuit held that punitive damages were unavailable on a claim for bad-faith breach of an insurance contract. Heil Co. v. Evanston Ins. Co., 690 F.3d 722, 728 (6th Cir. 2012). A later Tennessee Court of Appeals decision, by contrast, held that punitive damages were available. See Riad v. Erie Ins. Exch., 436 S.W.3d 256, 275–76 (Tenn. Ct. App. 2013). Finding that Riad had discredited Heil, the panel majority here departed from circuit precedent. Lindenberg v. Jackson Nat’l Life Ins. Co., 912 F.3d 348, 357–59 (6th Cir. 2018). As a result, the panel majority reached a decision that is not opposed by any controlling Tennessee authority but that nonetheless presents a significant danger of being wrong, for reasons discussed thoroughly by the dissent. See id. at 372–76 (Larsen, J., dissenting in part). If and when the Tennessee Supreme Court reaches the issue, it may well hold that punitive damages are not available on a bad-faith claim. As to the state constitutional question, there is also substantial reason to doubt that the Tennessee Supreme Court will invalidate the punitive-damages cap under the Tennessee constitution. See id. at 379–86 (Larsen, J., dissenting in part); see generally Br. Amici Curiae Chamber of Commerce of the United States of America, American Tort Reform Association, National Association of Manufacturers, National Federation of Independent Business, Small Business Legal Center, and American Property Casualty Insurance Association. In the meantime, plaintiffs who want punitive damages but seek to avoid the cap will be likely to file in federal district court. This is exactly the sort of forum-shopping that the Erie decision was meant to reduce. See Erie, 304 U.S. at 74–75 (stating that federal courts’ application of a general common law “made rights enjoyed under the unwritten ‘general law’ vary according to whether enforcement was sought in the state or in the federal court” and “rendered impossible equal protection of the law”); see also McCarthy v. Olin Corp., 119 F.3d 148, 157 (2d Cir. 1997) (Calabresi, J., dissenting); Todd v. Societe BIC, S.A., 9 F.3d 1216, 1222 (7th Cir. 1993) (en banc) (certifying Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 10 two questions of state tort law where “the panel’s analysis had substantial propensity to attract all future cases of this kind into federal court”). Indeed, dissenting from a Second Circuit panel’s decision to decide a state tort-law issue instead of certifying, Judge Guido Calabresi wrote: [F]ederal courts . . . have tended to be far too reluctant to certify questions to the state courts . . . . Specifically, federal courts have all too often refused to certify when they can rely on state lower court opinions to define state law. I view this reluctance as both wrong and unjust. Reluctance to certify is wrong because it leads to precisely the kind of forum shopping that Erie . . . was intended to prevent. See Hanna v. Plumer, 380 U.S. 460, 468 . . . (1965) (noting that one of the aims of the Erie decision was “discouragement of forum-shopping”). This is especially so in situations where there is some law in the intermediate state courts, but no definitive holding by the state’s highest tribunal. In such cases, and in the absence of certification, the party that is favored by the lower court decisions will almost invariably seek federal jurisdiction. It will do this in order to prevent the state’s highest court from reaching the issue, in the expectation that the federal court—unlike the state’s highest court—will feel virtually bound to follow the decisions of the intermediate state courts . . . . When federal courts, in effect, prevent state courts from deciding unsettled issues of state law, they violate fundamental principles of federalism and comity . . . . Federal courts that refuse to certify end up “mak[ing] important state policy, in contravention of basic federalism principles.” Hakimoglu v. Trump Taj Mahal Assocs., 70 F.3d 291, 302 (3d Cir. 1995) (Becker, J., dissenting) . . . . Reluctance to certify is unjust because, as has happened with some frequency, the federal court, having refused to certify, may decide an issue of state law one way, only to discover that the state’s highest court, when presented with the issue in a later case, reaches the opposite result . . . . McCarthy, 119 F.3d at 157–59 (Calabresi, J., dissenting) (emphasis added) (footnotes and some citations omitted). Judge Frank Easterbrook of the Seventh Circuit sounded the same theme in his opinion for the en banc court in a tort case that raised unsettled state-law questions: In a case such as [this tort case] . . . . any substantial divergence between the federal court’s estimate of state law and the state’s view of its own law will funnel all similar litigation to federal court . . . . If the federal court treats the plaintiff more favorably than the state tribunal would, then the plaintiff always files in Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 11 federal court; similarly[,] any departure in the [defendant’s] favor leads the defendant to remove any suit filed in state court. In either case, the state loses the ability to develop or restate the principles that it believes should govern the category of cases. Certification then ensures that the law we apply is genuinely state law. Todd, 9 F.3d at 1222 (citation omitted). If Calabresi and Easterbrook—two prominent federal judges of sometimes differing perspectives—have voiced identical worries about incentivizing forum-shopping through reluctance to certify, we in the Sixth Circuit should consider taking a definite step toward remediating those worries. Despite these forum-shopping concerns, one objection to certification is that state courts, and the Tennessee Supreme Court in particular, often decline to answer certified questions. Although undoubtedly a true statement, that objection did not counsel against certification here for three reasons. First, even if the state court declined to answer, this court would still have done the Tennessee Supreme Court the courtesy—requested by the Tennessee attorney general—of giving it the opportunity to speak authoritatively on its own law. See En Banc Pet. of State of Tenn. as Intervenor-Def./Cross-Appellee at 6. If that court declined to do so, then responsibility for any “friction-generating error” produced by a decision of this court would not lie at our door alone. Arizonans, 520 U.S. at 79. Second, the Tennessee Supreme Court has made clear that it views certification as a valuable mechanism for preserving the sovereignty of state courts: More importantly, the certification procedure protects states’ sovereignty. To the extent that a federal court applies different legal rules than the state court would have, the state’s sovereignty is diminished [because] the federal court has made state law. Such an impact on state sovereignty is no small matter, especially since a federal court’s error may perpetuate itself in state courts until the state’s highest court corrects it. Haley v. Univ. of Tenn.-Knoxville, 188 S.W.3d 518, 521 (Tenn. 2006) (citations and internal quotation marks omitted). Thus, the Tennessee judiciary has a favorable view of certification as a general matter, although, of course, the Tennessee Supreme Court is under no obligation to answer certified questions. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 12 Third, in this case specifically, the Tennessee Supreme Court expressly stated in its denial of the district court’s certified constitutional questions that it was making no comment on what it might do if the Sixth Circuit later certified the question of constitutionality and that question had become determinative because the predicate question of punitive damages’ availability was also presented and certified. See Lindenberg, 912 F.3d at 372 (Larsen, J., dissenting in part). So the danger that the court would decline to answer certified questions was no greater than usual here and quite possibly less than usual, given the importance of the state constitutional question. Speaking of the constitutional question, it is unusual for the panel to have invalidated a state statute on state constitutional grounds. This decision is in tension, in two respects, with the approach that the Supreme Court of the United States and our court have counseled in similar cases. First, the Supreme Court and our court have indicated that abstention or certification is appropriate where a decision on state law may allow the federal court to avoid a federal constitutional question. See, e.g., Bellotti v. Baird, 428 U.S. 132, 146–47 (1976); Planned Parenthood Cincinnati Region v. Strickland, 531 F.3d 406, 410–11 (6th Cir. 2008). Because federalism concerns as well as avoidance concerns appear in a case like this one, where a state constitutional question lurks behind a predicate state-law question, certification seems doubly wise. Indeed, the Eleventh Circuit has noted that “[c]ertification . . . is especially appropriate in a case . . . where the decisional task involves interpreting the state constitution.” LeFrere v. Quezada, 582 F.3d 1260, 1268 (11th Cir. 2009) (citation omitted).6 Second, and relatedly, the Supreme Court has indicated that the possibility of making an Erie guess that results in invalidating a state law should be avoided where certification makes avoidance possible. See Arizonans, 520 U.S. at 79. 6 See also Estate of McCall ex rel. McCall v. United States, 642 F.3d 944, 946 (11th Cir. 2011) (certifying question whether cap on noneconomic medical malpractice damages violated provisions of Florida constitution on which there was no Florida Supreme Court guidance); Int’l Soc’y for Krishna Consciousness of Cal. Inc. v. City of L.A., 530 F.3d 768, 773–76 (9th Cir. 2008) (certifying question whether airport was “public forum” under California constitution); Parcell v. Governmental Ethics Comm’n, 626 F.2d 160, 161 (10th Cir. 1980) (certifying question whether method of appointing Ethics Commission members violated “the doctrine of separation of powers as the same is recognized as a part of the Kansas State Constitution”). Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 13 Also important to note is that, although neither Ms. Lindenberg nor Jackson National moved our court to certify the particular questions at issue here, our court can and does certify questions sua sponte. See, e.g., Am. Booksellers Found. for Free Expression v. Strickland, 560 F.3d 443, 444 (6th Cir. 2009) (order). This would not be quite a sua sponte certification, anyway: Ms. Lindenberg already moved for certification of the state constitutional question in the district court, at which point the Tennessee attorney general intervened in the action. See Lindenberg, 912 F.3d at 355. The district court did certify the constitutional question (and one other constitutional question, not at issue here), but the Tennessee Supreme Court declined to answer it because it would not have been determinative of the litigation. Id.; see id. at 371–72 (Larsen, J., dissenting in part). On appeal, the Tennessee attorney general asked our court to certify the state constitutional questions in the event we found punitive damages were available. Br. of State of Tenn. as Intervenor-Def./Cross-Appellee at 9 n.2. Then, when asked at oral argument before our court if they had any objection to certification of the now determinative questions regarding the availability of punitive damages and the constitutionality of the punitive- damages cap, both sides agreed that certification would be appropriate. Thus, all factors seem to point toward certification here. But because our circuit has no guidelines for certification beyond suggesting that it is appropriate for novel and unsettled questions of state law, the panel could disregard the availability of the certification procedure. To clarify our standards is the primary reason we should have granted rehearing in this case. The case is appropriate for en banc reconsideration in other ways as well, however. The Federal Rules of Appellate Procedure state that one ground for en banc rehearing is a split in circuit precedent. See Fed. R. App. P. 35(a)(1). We have that here. Furthermore, our circuit rules state that a panel decision may not be overruled except by the en banc court.7 6th Cir. R. 32.1(b). The panel decision here not only departs from precedent but also creates a major risk of 7 The Lindenberg majority noted that “a single decision of a state court of appeals may abrogate this Court’s interpretation of state law, at least in circumstances where (1) state law treats an appellate court decision as controlling in the absence of a ruling from the state supreme court; (2) there is no indication from the state supreme court that it would reach a different outcome; and (3) the state appellate court’s decision is irreconcilable with our own ruling.” 912 F.3d at 357 (citations omitted). However, here, as the dissent persuasively argued, there is reason to believe the Tennessee Supreme Court would reach a different outcome. See id. at 370, 373, 375 (Larsen, J., dissenting in part). Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 14 horizontal forum-shopping, in contravention of fundamental federalism principles. Thus, it involves an issue of great importance. See Fed. R. App. P. 35(a)(2) (providing for en banc rehearing in cases of “exceptional importance”). In addition, to the extent our internal operating procedures counsel against rehearing solely state-law issues, see 6th Cir. I.O.P. 35(a), we would not be reconsidering such issues en banc here. As for the state-law issues, we would not be deciding them: we would simply be asking the state court to do that. And this would not be the first time that we have, while sitting en banc, certified issues to a state supreme court. See Duffy v. Foltz, 804 F.2d 50, 53–54 (6th Cir. 1986) (applying state court’s answer to certified question after en banc court certified the question); cf. Todd, 9 F.3d at 1222 (Seventh Circuit certifying two questions of state tort law, which had become relevant as a result of the en banc court’s vacating the panel decision, and observing that “[l]ittle would be served by substituting the guess of eleven judges for that of three; far better to pose the questions to the only judges who can give definitive answers”). But, in any event, if we had reheard this case en banc, we could have considered a very important federal question: what certification standard should apply in our circuit to implement the constitutional federalism principles articulated by Erie and its progeny. In other words, we should have used this case to articulate more meaningful standards to guide certification decisions. At the very least, there should be a presumption in favor of certification where, as here, a state supreme court has not decided an issue; neither party objects to certification; and a prior precedential panel decision of this court stands between the current panel and the decision it wishes to reach on state law. See Clark, supra note 4, at 1553–54 (arguing for “a presumption in favor of certification in cases presenting” unsettled questions of state law whose “resolution entail[s] the exercise of significant policymaking discretion”); cf. McCarthy, 119 F.3d at 161 (Calabresi, J., dissenting) (stating that certification is “appropriate” in “virtually any case in which 1) a significant and dispositive issue of state law is in genuine doubt . . . and 2) certification is specifically requested by the party that did not invoke federal court jurisdiction”). There should likewise be a presumption in favor of certification where the panel is facing an unclear issue of state constitutional law. See LeFrere, 582 F.3d at 1268. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 15 Such a presumption would not upend the way we currently decide cases in the Sixth Circuit. I am not advocating for certifying questions in a vast set of new situations or for requiring every panel to certify if a certain group of boxes is checked. However, I do think we should make it easier for litigants to predict when this court will certify questions and easier for the en banc court to determine whether a panel has made a grave error in deciding a question of unsettled state law itself instead of certifying. In sum, we have missed an opportunity to address a significant issue that is likely to recur. Assuming the Supreme Court provides no further guidance (but perhaps it will, which I would welcome), the burden falls on each circuit to define standards for certifying questions, and at some point we should examine our standards more carefully. Otherwise, we risk validating the prediction of the Antifederalists: that an encroaching federal judiciary would use federal judicial power to diminish the power of state judiciaries. To minimize the risk of unnecessary interference with the autonomy and independence of the states, we should more frequently accept state courts’ open invitations to pose to them certified questions regarding their own law. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 16 _________________ STATEMENT _________________ NALBANDIAN, Circuit Judge. Today’s decision marks a missed opportunity for our court to more firmly establish its commitment to a “cooperative judicial federalism.” Arizonans for Official English v. Arizona, 520 U.S. 43, 77 (1997). I would have granted rehearing en banc to certify the state-law questions to the Tennessee Supreme Court, rather than risk the kind of “friction-generating error” that arises when federal courts invalidate state statutes. Id. at 79. But the panel’s decision not to certify the questions is not the last word on the matter. Nothing prevents future courts—whether another panel from this circuit or one of our district court colleagues—from certifying the same questions to the Tennessee Supreme Court should they arise again. At first blush, it may seem inconsistent with stare decisis for a district court or a later panel of this court to certify a question after one panel has already made an Erie prediction about state law. But we have endorsed using the certification process to clarify state law in this exact situation. See Geib v. Amoco Oil Co., 29 F.3d 1050, 1060–61 (6th Cir. 1994). Rather than allow federal courts to “authoritatively determin[e] unresolved state law,” the better practice is to send those questions to the state judiciary for resolution. Id. at 1061 (quoting Clay v. Sun Ins. Office Ltd., 363 U.S. 207, 212 (1960)). The state court, of course, can turn us down. See, e.g., Geib v. Amoco Oil Co., 163 F.3d 329, 330 (6th Cir. 1995). And in that case we must keep following the dictates of stare decisis. Id. But until the state judiciary speaks on an unsettled issue of state law, no amount of decisions from this court prevents the next court from certifying the question. Years after Geib, another panel in our circuit faced this same dilemma. The majority opted not to exercise its discretion to certify because, among other reasons, “it would arguably be inconsistent with” the court’s precedent. Rutherford v. Columbia Gas, 575 F.3d 616, 619 (6th Cir. 2009). But in a thoughtful dissent, Judge Clay cast serious doubt on that proposition. “[C]ertifying a question to a state court does not implicate, much less contradict, our obligations under stare decisis.” Id. at 623 (Clay, J., dissenting). That’s because asking the state court to Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 17 weigh in does not modify or overturn prior precedent. Id. It’s an exercise of deference to the judicial body that actually holds the power to resolve unsettled questions of state law. And, more importantly, the majority’s decision in Rutherford was not inconsistent with what we said in Geib. The majority in Rutherford only exercised its discretion not to certify that particular question, did not cite the prior panel decision in Geib, and did not discuss the question of whether certification would have been inconsistent with stare decisis. To be sure, not every circuit agrees. The Fifth Circuit, for example, adopts a more restrictive view of stare decisis. Courts there must continue applying suspect precedent rather than certify the issue to the state court. See Hughes v. Tobacco Inst., Inc., 278 F.3d 417, 425 (5th Cir. 2001); Lee v. Frozen Food Express, Inc., 592 F.2d 271, 272 (5th Cir. 1979). But the Fifth Circuit also permits rehearing en banc to correct panel decisions that misapply state law. See I.O.P. following 5th Cir. R. 35; see also Sturgeon v. Strachan Shipping Co., 731 F.2d 255, 256 (5th Cir. 1984); Hudson v. R. J. Reynolds Tobacco Co., 427 F.2d 541, 542 (5th Cir. 1970) (“We are bound by the [precedent] on this issue . . . until, if ever, the Court en banc redecides the question or the Louisiana courts hold differently.”). We, on the other hand, have no such luxury. Our internal rules preclude rehearing en banc for alleged errors of state law. See 6 Cir. I.O.P. 35(a). Adopting the Fifth Circuit rule while maintaining our own rules for rehearing en banc would turn a randomly selected, three-judge panel into the court of last resort for many state-law issues. See Geib, 29 F.3d at 1060 (explaining the “very real danger that Michigan’s courts will be denied any meaningful participation in the interpretation of” their own law for issues that almost always involve diverse parties). But we have avoided that predicament by establishing a narrower view of stare decisis. Federal courts have a duty to properly decide questions of state law. It’s a duty “from which we may not shrink.” Rutherford, 575 F.3d at 624 (Clay, J., dissenting). Certification is one tool to assist us in this endeavor. See Lehman Bros. v. Schein, 416 U.S. 386, 390–91 (1974). And it makes no difference whether one panel has already spoken on the issue. See Geib, 29 F.3d at 1060–61. See also Eads v. Morgan, 298 F. Supp. 2d 698, 707 (N.D. Ohio 2003). We are, after all, merely predictors of state law. See Stryker Corp. v. XL Ins. Am., 735 F.3d 349, 358 (6th Cir. 2012). We speculate about how the state judiciary might answer these unsettled questions. Nos. 17-6034/6079 Lindenberg v. Jackson Nat’l Life Ins. Co., et al. Page 18 But stare decisis does not turn unsettled questions of state law into settled ones. And federal courts must always be free to seek answers from the only judicial body capable of providing them. ENTERED BY ORDER OF THE COURT ____________________________________ Deborah S. Hunt, Clerk