575 F.3d 616
6th Cir.
2009

Lead Opinion

OPINION

ROGERS, Circuit Judge.

Like cases should end in like judgments. Once this court decides questions of law presented in a dispute, a nearly identical dispute ought to yield a similar outcome. Emily Rutherford challenges the Columbia Gas Transmission Corporation’s ability to maintain its pipeline easement by clearing certain trees from her land, but her appeal comes to us soon after we affirmed the same magistrate’s rejection of another landowner’s similar claims concerning a similar easement relating to the same gas pipelines. The facts and legal arguments of the two appeals are nearly indistinguishable, so our prior decision requires affirmance of the judgment below.

While the application of stare decisis to this case is straightforward, the procedural posture creates problems of appellate jurisdiction. This suit commenced when Rutherford (a citizen of Ohio) filed an Ohio court suit to keep Columbia from cutting down seven trees that grew on Columbia’s easement on her land. She sought a declaratory judgment protecting the trees, a declaratory judgment defining the easement, damages, and costs. Columbia (incorporated in Delaware; principal place of business in West Virginia) removed based on diversity jurisdiction and filed various counterclaims. Columbia’s counterclaims included a request for a declaratory judgment defining its easement, permitting it to remove the trees, and finding Rutherford liable for various damages; a claim for injunctive relief; a claim for damages for breach of contract and property rights; a claim for punitive damages; and a claim for attorney fees and costs.

*618The magistrate made various findings of fact and conclusions of law that explicitly rejected each of Rutherford’s claims. However, the magistrate did not address Columbia’s then-still-pending counterclaims. Nevertheless, the magistrate directed the clerk to enter judgment for Columbia based on the findings, and Rutherford filed a notice of appeal from the judgment, even though it was not final because it did not resolve all claims between the parties. See Fed.R.Civ.P. 54(b).

After we directed the parties to supplement their inadequate jurisdictional briefing and address this problem, the parties evidently asked the district court to amend the judgment. The magistrate entered an order entitled “Nunc Pro Tunc Order Entry of Judgment.” The order reiterates the previously entered judgment, without explanation adds a declaratory judgment finding a right of way of 25 feet on each side of the pipelines, and dismisses Columbia’s other counterclaims.

The magistrate did not seek leave under Fed.R.Civ.P. 60(a) to correct the judgment while an appeal was docketed in this court, and it is hardly clear that an order that adds new relief to a judgment could be a correction within the scope of Rule 60(a). But resolution of that question is not necessary because at oral argument Columbia explicitly stated that it was willing to relinquish all of its claims for relief in order to ensure appellate jurisdiction. While we do not encourage this procedure, it is enough to permit the exercise of appellate jurisdiction in this case. See Scarbrough v. Perez, 870 F.2d 1079, 1081-82 (6th Cir.1989); G.G. Marck & Assocs., Inc. v. Peng, 309 Fed.Appx. 928, 931-32 (6th Cir.2009); but see Smoot v. Mazda Motors of Am., Inc., 469 F.3d 675, 676-78 (7th Cir.2006) (both parties ordered to show cause why their counsel should not be sanctioned for deficient jurisdictional statements). All of Columbia’s counterclaims, including its request for declaratory relief, are therefore considered dismissed with prejudice. We thus have jurisdiction to review the magistrate’s rejection of Rutherford’s claims for relief.

These claims are largely controlled by our recent decision in Andrews v. Columbia Gas Transmission Corp., 544 F.3d 618 (6th Cir.2008). There we upheld the rejection of similar claims against Columbia for removing trees from its easement across Andrews’s land. See id. at 621-22. Like Rutherford, Andrews had requested declaratory and injunctive relief to protect trees on the easement. Id. at 622. As in Rutherford’s case, the magistrate rejected Andrews’s Ohio law claims of laches, estoppel, and waiver. Id. at 622-23. And as he did in Rutherford’s case, the magistrate found that Columbia’s planned tree removal was consistent with the terms of the easement. Id. Rutherford offers little to distinguish her case from the judgment affirmed in Andrews.

In her reply brief, Rutherford argues that the court should distinguish Andrews because Rutherford’s trees likely had been planted in the mid-1950s, around the time Columbia’s predecessor obtained the last of the easements at issue. The Andrews trees did not exist at the time of the creation of the easements at issue in that case. Id. at 625. Rutherford argues that the fact that the gas company allowed small trees to remain on the property when it installed the second gas line shows that the parties contemplated allowing large trees to grow on the easement. But in construing an express easement with unclear dimensions, the court considers not only “circumstances surrounding” the creation of the easement, but also “what is reasonably necessary and convenient to serve the purposes for which the easement was granted.” See id. at 624 and cases *619cited therein. Thus even if the circumstances surrounding the creation of the easements are marginally different in the two cases, Rutherford has no developed argument as to how the magistrate’s finding that a cleared right of way is reasonably necessary is distinguishable from the similar finding upheld in Andrews. See id. at 626-30.

Because in Andrews we upheld an indistinguishable factual finding that a cleared right of way is reasonably necessary to serve the purpose of the easement, we must uphold that magistrate’s finding in this case that Columbia may clear trees from Rutherford’s easement. And clearing even a 15-foot right of way on each side of the pipelines would require the removal of the seven trees at issue here. Thus, Rutherford is not entitled to relief on any of her claims.

Rutherford also contests the magistrate’s holding that various equitable doctrines do not apply to express easements as a matter of Ohio law. But when pressed on this point at oral argument, her counsel conceded that Andrews’s identical holding controls, and counsel merely invited us to “reconsider” Andrews. A published prior panel decision “remains controlling authority unless an inconsistent decision of the United States Supreme Court requires modification of the decision or this Court sitting en banc overrules the prior decision.” Salmi v. Sec’y of Health & Human Servs., 774 F.2d 685, 689 (6th Cir.1985). Without taking a case en banc, a “panel cannot” reconsider a prior published case that interpreted state law, “absent an indication by the [state] courts that they would have decided [the prior case] differently.” Blaine Constr. Corp. v. Ins. Co. of N. Am., 171 F.3d 343, 350 (6th Cir.1999). Or, as we recently stated in a diversity case, “[w]e are bound by [a prior published case that interpreted Ohio law] unless Ohio law has measurably changed in the meantime.” Big Lots Stores, Inc. v. Luv N’ Care, Ltd., 302 Fed.Appx. 423, 427 (6th Cir.2008). The Tenth Circuit has explained:

Where no controlling state decision exists, the federal court must attempt to predict what the state’s highest court would do. In performing this ventriloquial function, however, the federal court is bound by ordinary principles of stare decisis. Thus, when a panel of this Court has rendered a decision interpreting state law, that interpretation is binding on district courts in this circuit, and on subsequent panels of this Court, unless an intervening decision of the state’s highest court has resolved the issue.

Wankier v. Crown Equip. Corp., 353 F.3d 862, 866 (10th Cir.2003). As no Ohio court has suggested that Andrews misapplied Ohio law or reached a contrary holding, Andrews’s holding that laches, estoppel, and waiver do not apply to expressly granted easements controls this case. See Andrews, 544 F.3d at 630-31.

That conclusion is sufficient to resolve this case. Moreover, we decline to exercise our discretion to certify questions of state law to the Ohio Supreme Court. First, no party has requested certification at any stage of this litigation. Second, certification is not warranted because it would arguably be inconsistent with the Andrews court’s determination not to seek such a certification sua sponte. Third, available evidence does not suggest that the Ohio Supreme Court is likely to disagree with Andrews’s interpretation of Ohio law.

At the parties’ request, we consider the district court’s purported nunc pro tunc order as a confirmation that Columbia has agreed to the dismissal of its counterclaims. Because the order appealed from *620properly denied Rutherford relief on her claims, we affirm.





Concurrence Opinion

CLAY, Circuit Judge,

concurring in part and dissenting in part.

Although I agree with much of the majority’s reasoning, I disagree with the notion that, at least with respect to Rutherford’s claims regarding the equitable doctrines of waiver, laches, and estoppel, stare decisis requires us to blindly defer to our decision in Andrews v. Columbia Gas Transmission Corp., 544 F.3d 618 (6th Cir.2008). Because Andrews, in determining Ohio law with respect to the application of these equitable doctrines, relied on an intermediate state appellate court decision to the exclusion of significant evidence that the Supreme Court of Ohio would reach a different conclusion, our obligation to properly determine Ohio state law conflicts with our obligation to respect and follow Andrews. Under these unique circumstances, the balance of interests strongly suggests that we should certify a question to the Supreme Court of Ohio seeking its input on the proper scope of these doctrines under Ohio law. Therefore, I respectfully dissent as to the resolution of Rutherford’s equitable claims.

I.

Our decision in Andrews, like the magistrate judge’s decision in this case, relied almost exclusively on the Ohio Court of Appeals’ decision in Lone Star Steakhouse & Saloon of Ohio, Inc. v. Ryska, No.2003L-192, 2005 WL 1538259, *7, 2005 Ohio-App. LEXIS 3146, at *17 (Ohio Ct.App. June 30, 2005), to conclude that, unlike an easement obtained by prescription, an easement created by an express grant is not subject to the equitable doctrines of waiver, laches, and estoppel. There is no doubt that the Ohio Court of Appeals held in Lone Star that “equity does not acknowledge the extinguishment of such an [expressly granted] easement by recourse to estoppel or laches.” Id. at *7, 2005 Ohio App. LEXIS 3146, at *19. However, where a state’s highest court has not conclusively ruled on an issue, the decisions of intermediate state courts, although entitled to “ ‘some weight,’ ” are “ ‘not controlling.’ ” Comm’r of Internal Revenue v. Estate of Bosch, 387 U.S. 456, 465, 87 S.Ct. 1776, 18 L.Ed.2d 886 (1967) (quoting King v. Order of Travelers, 333 U.S. 153, 160-61, 68 S.Ct. 488, 92 L.Ed. 608 (1948)). Instead, federal courts sitting in diversity are obliged to “predict” how the state’s highest court would rule “by looking to ‘all available data.’ ” Prestige Cas. Co. v. Mich. Mut. Ins. Co., 99 F.3d 1340, 1348 (6th Cir.1996) (quoting Kingsley Assoc, v. Moll PlastiCrafters, Inc., 65 F.3d 498, 507 (6th Cir.1995)). Accordingly, while the holding of Lone Star provides presumptively persuasive “datum for ascertaining state law,” we may disregard that holding if we are “convinced by other persuasive data that the highest court of the state would decide otherwise.” West v. Am. Tel. & Tel. Co., 311 U.S. 223, 227, 61 S.Ct. 179, 85 L.Ed. 139 (1940).

A review of Ohio law reveals very persuasive “data” that the Supreme Court of Ohio would not accept the rule of Lone Star. The Supreme Court of Ohio has applied these equitable doctrines in the context of other written contracts, see, e.g., Pedler v. Aetna Life Ins. Co., 23 Ohio St.3d 7, 490 N.E.2d 605, 607 (1986)1 (ap*621plying the “principles of equitable estoppel” to written insurance contract, and holding that “an insurer is estopped from denying the full value of coverage stated on the insurance certificate, based upon limitations or exclusions contained in the master policy, where the insured bargained and paid for such coverage, unless the insured knew or should have known of his ineligibility”), and has interpreted right-of-way agreements according to principles of contract law, see Alexander v. Buckeye Pipe Line Co., 53 Ohio St.2d 241, 374 N.E.2d 146, 150 (1978). Moreover, in considering claims regarding the alleged interference with a right of way created by an express agreement between adjacent landowners, the Supreme Court of Ohio has expressly rejected the notion that “a plaintiffs remedy should be at law and not equity,” recognizing that equitable remedies may provide “the proper mode of enforcing the agreement.” Goldberger v. Bexley Properties, 5 Ohio St.3d 82, 448 N.E.2d 1380, 1383 (1983). Although not speaking to the applicability of the doctrine of laches and estoppel specifically, Goldberger recognized that other equitable doctrines, such as unclean hands and adverse possession, apply with equal force in the context of express easements. Id. Our decision in Andrews did not consider any of these decisions in determining that the Supreme Court of Ohio would accept the rule announced in Lone Star.

Furthermore, the rule announced in Lone Star seems to be predicated on a fundamental misreading of Zimmerman v. Cindle, 48 Ohio App.3d 164, 548 N.E.2d 1315 (1988). Although the syllabus to Zimmerman states that “an easement obtained by prescription” is subject to the doctrine of laches and estoppel, nothing in the Zimmerman opinion suggests that this is the only context in which those doctrines apply. If anything, Zimmerman seems to imply that these equitable doctrines apply to all easements. Id. at 1317 (discussing equitable doctrine of estoppel in the context of “an easement,” and apparently referring to the original easement at issue in the case which was “an express easement created by an instrument ... and duly recorded”). Lone StaPs narrow reading of Zimmerman also is belied by the fact that subsequent decisions by other Ohio Courts of Appeals have held that express easements are subject to the doctrine of abandonment. See, e.g., 1st Nat’l Bank v. Mt. Agency, LLC, 2009 WL 1278429, at *6, 2009 Ohio App. LEXIS 1958, at *13 (Ohio Ct.App. May 11, 2009).

To the extent that there is a conflict between Zimmerman and Lone Star, we would be obliged to apply the earlier rule of Zimmerman. See Sowards v. Loudon County, Tenn., 203 F.3d 426, 431 n. 1 (6th Cir.2000). That is especially true here given that Lone Star purports to apply the rule set forth in Zimmerman, and also characterizes Zimmerman as “the leading authority in Ohio on the extinguishment of easements via the doctrines of estoppel and laches.” See Lone Star, 2005 WL 1538259, *7, 2005 Ohio App. LEXIS 3146, at *18-19.

In addition, although the law of other states does not dictate the result here, it is relevant to our inquiry. See, e.g., Herman Miller, Inc. v. Palazzetti Imps. & Exps., Inc., 270 F.3d 298, 326 (6th Cir.2001) (“In *622reaching our decision [regarding Michigan law], we are also influenced by the number of states that have recognized a post-mortem right of publicity.”). To that limited extent, then, it is important to note that other states have recognized that express easements are subject to these equitable doctrines. See, e.g., Jewett v. Leisinger, 655 So.2d 1210, 1212 (Fla.Dist.Ct.App. 1995); Crew’s Die Casting Corp. v. Davidow, 369 Mich. 541, 120 N.W.2d 238, 240 (1963); see also Twp. of Piscataway v. Duke Energy, 488 F.3d 203, 214 (3d Cir.2007) (applying New Jersey law). At a minimum, these decisions demonstrate that there is nothing inherently unique about an expressly granted easement such that the rights created therein necessarily preclude considerations of equity. Nor would such a limitation make sense given that the Supreme Court has applied the doctrine of laches to land claims between the states where a state’s “title and rightful authority” to disputed territory is at issue. See Ohio v. Kentucky, 410 U.S. 641, 651, 93 S.Ct. 1178, 35 L.Ed.2d 560 (1973) (“ ‘The rule, long-settled and never doubted by this court, is that long acquiescence by one state in the possession of territory by another and in the exercise of sovereignty and dominion over it is conclusive of the latter’s title and rightful authority.’ ” (quoting Michigan v. Wisconsin, 270 U.S. 295, 308, 46 S.Ct. 290, 70 L.Ed. 595 (1926))); Massachusetts v. New York, 271 U.S. 65, 95, 46 S.Ct. 357, 70 L.Ed. 838 (1926) (“Long acquiescence in the possession of territory and the exercise of dominion and sovereignty over it may have a controlling effect in the determination of a disputed boundary.”).

Despite this significant evidence that the holding of Lone Star is not an accurate statement of Ohio law, our decision in Andrews relied almost exclusively on Lone Star in determining that Ohio law precludes the application of these equitable doctrines in the context of express easements. Because of this likely error, I disagree with the majority’s conclusion that we have no choice but to resolve Rutherford’s claims by reference to the rule set forth in Andrews. The approach endorsed by the majority glosses over the shortcomings of our inquiry in Andrews, and in fact compounds the error by lending credence to such an approach.

II.

According to the majority, we are “bound” to follow Andrews because no intervening Ohio state court decision “has suggested that Andrews misapplied Ohio law or reached a contrary holding.” Maj. Op. at 618-19 (citing Blaine Constr. Corp. v. Ins. Co. of N. Am., 171 F.3d 343, 350 (6th Cir.1999), and Big Lots Stores, Inc. v. Luv N’ Care, Ltd., 302 F. App’x 423, 427 (6th Cir.2008)). Underlying the majority’s position is the notion that stare decisis imposes on us an important obligation to respect and abide by the decisions handed down by prior panels of this Court. See Darrah v. City of Oak Park, 255 F.3d 301, 309-10 (6th Cir.2001). A number of circuits have reached a similar conclusion, reasoning that stare decisis “applies with equal force to cases in which state law supplies the substantive rule of decision,” Broussard v. S. Pac. Transp. Co., 665 F.2d 1387, 1389 (5th Cir.1982) (en banc), and thus holding that a prior panel decision “should be followed by other panels without regard to any alleged existing confusion in state law, absent a subsequent state court decision or statutory amendment which makes this Court’s (prior) decision clearly wrong,” Lee v. Frozen Food Express, Inc., 592 F.2d 271, 272 (5th Cir.1979) (per curiam); accord Woodling v. Garrett Corp., 813 F.2d 543, 557 (2d Cir.1987) (“A ruling of one panel of this Circuit on an issue of state law normally will not *623be reconsidered by another panel absent a subsequent decision of a state court or of this Circuit tending to cast doubt on that ruling.”).

What the majority fails to recognize, however, is that certifying a question to a state court does not implicate, much less contradict, our obligations under stare decisis. Wdien we certify a question to a state court, we are not modifying, overturning, or otherwise refusing to follow prior precedent. Rather, we are seeking guidance from a court to which we are bound to defer on issues of state law. Certifying a question regarding Ohio law to Ohio’s highest court thus does not run counter to or even implicate whatever obligations we have to follow Andrews, it merely reflects sound judicial management.

Confronted with a wrongly decided prior panel decision, our only option normally would be to follow that decision and encourage the en banc Sixth Circuit or Supreme Court to reconsider the issue. In a diversity case such as this, however, we have the additional option of certifying a question to the state’s highest court, at least where that court’s rules make the certification procedure available. If the rules of the Supreme Court of Ohio did not make the certification procedure available, then I would agree that we would be bound by Andrews. But because the certification option is available here, and in light of the interests at stake, I believe that we are compelled to exercise our discretion and provide the Supreme Court of Ohio an opportunity to address this issue.

A.

The doctrine of stare decisis et non quieta movere — meaning “to stand by things decided, and not to disturb settled points,” Black’s Law Dictionary 1414 (7th ed.1999) — reflects the principle that, for the sake of consistency and order, courts should follow prior decisions regarding the same points of law. But stare decisis “is not an ‘inexorable command.’ ” Planned Parenthood v. Casey, 505 U.S. 833, 854, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992) (quoting Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 405, 52 S.Ct. 443, 76 L.Ed. 815 (1932) (Brandeis, J., dissenting)). Rather, determining whether to follow past precedent requires an exercise of “judgment” that “is customarily informed by a series of prudential and pragmatic considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of law, and to gauge the respective costs of reaffirming and overruling a prior case.” Id. In most cases, stare decisis “is usually the wise policy, because in most matters it is more important that the applicable rule of law be settled than that it be settled right.” Burnet, 285 U.S. at 406, 52 S.Ct. 443 (Brandéis, J., dissenting). In certain circumstances, however, the balance of interests is shifted, and our responsibility to make sure that an issue is “settled right” is more important than the benefits of letting the issue remain merely “settled.” In such cases, the considerations that usually support our following prior precedent must give way to other competing interests. This is one of those circumstances.

Where, such as here, our jurisdiction is predicated on the diversity of parties, we are obliged to apply state law in resolving the substantive issues presented. See Erie R.R. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). In fulfilling that obligation, we are required to determine what in fact is the applicable state law. Our duty to properly determine state law is a serious responsibility that implicates principles at least as significant as those underlying stare decisis. In fact, unlike the “pragmatic considerations” un*624derlying stare decisis, see Casey, 505 U.S. at 854, 112 S.Ct. 2791, Erie imposes on federal courts a “duty” and “great[] responsibility” to properly determine state law, see Meredith v. Winter Haven, 320 U.S. 228, 234, 64 S.Ct. 7, 88 L.Ed. 9 (1943). In all cases arising under our diversity jurisdiction, Eñe requires that we “adjudicate the rights of the parties with the aid of such light as was afforded by the materials for decision at hand, and in accordance with the applicable principles for determining state law.” Id. at 238, 64 S.Ct. 7. This is an obligation from which we may not shrink. As the Supreme Court has emphasized, “diversity jurisdiction was not conferred for the benefit of the federal courts or to serve their convenience,” but rather was adopted by Congress “to afford to suitors an opportunity in such cases, at their option, to assert their rights in the federal rather than in the state courts.” Id. at 234, 64 S.Ct. 7.

In certain cases, our “duty” to properly determine and apply state law comes into conflict with the interests that normally would support our deferring to prior precedent. In such cases, some courts have recognized that the obligation to properly determine state law is more important than the general dictate to defer to prior federal precedent construing state law. See United States v. Maness, 23 F.3d 1006, 1008-09 (6th Cir.1994) (refusing to follow Fourth Circuit’s interpretation of North Carolina state law because “the Fourth Circuit did not follow a contrary prior state supreme court decision” even though “we would usually defer to the Fourth Circuit’s prediction of an issue of first impression regarding North Carolina state law”); Factors Etc., Inc. v. Pro Arts, Inc., 652 F.2d 278, 283 (2d Cir.1981) (emphasizing that prior federal determinations of state law should be followed “for the benefit of both the orderly development of state law and fairness to those subject to state law requirements,” but acknowledging that prior federal interpretations may be disregarded where “prior state court decisions had been inadvertently overlooked by the pertinent court of appeals”).

B.

Confronted with these competing considerations, the majority resolves the issue by suggesting that stare decisis operates in this case not as a merely prudential and pragmatic consideration, but as a binding rule. In other words, according to the majority, it is not just that we would “usually defer” to Andrews, as was the ease in Maness, but that we are required to do so.

The majority’s position does have some appeal, at least superficially, in that our case law dictates that prior reported panel decisions are binding on all subsequent panels “unless an inconsistent decision of the United States Supreme Court requires modification of the decision or this Court sitting en banc overrules the prior decision.” Salmi v. Sec’y of Health & Human Servs., 774 F.2d 685, 689 (6th Cir.1985). Even this Court’s rules provide that “[r]eported panel opinions are binding on subsequent panels. Thus, no subsequent panel overrules a published opinion of a previous panel.” 6 Cir. R. 206(c).

But this “binding” rule is not implicated by our certifying a question of Ohio state law to the Supreme Court of Ohio because that procedure does not “overrule” Andrews. While there may be sound and persuasive practical reasons not to exercise the certification option in certain cases — none of which are present here— we certainly are not precluded from doing so by Rule 206(c) or our decisions in Sal-mi Blaine Construction, or Big Lots. Indeed, none of these decisions even considers whether the unique dynamics of cases arising under our diversity jurisdiction re*625move the certification protocol from the constraints of Rule 206(c).

Significantly, the “binding” rule of Sal-mi and Rule 206(c) includes important elements that make it more constraining than the traditional doctrine of stare decisis. Like stare decisis, the requirement that a subsequent panel must follow reported decisions of prior panels implicates considerations of consistency and order. That much is obvious. What perhaps is less obvious, however, is that this rule also includes the separate notion that the authority to modify or reverse a ruling should reside only in the hands of a superi- or body, and includes as a minor premise the unextraordinary proposition that panels within the same circuit are co-equals. While I do not disagree with either this rule or its underlying premises, the majority’s contention that this rule applies unyieldingly in this case overlooks several critical factors that suggest that this binding rule operates differently in diversity cases such as this.

Most importantly, it makes sense to transform the pragmatic doctrine of stare decisis into a binding rule with respect to co-equal panels only because there are at least two higher authorities — the en banc court and the Supreme Court — from which a dissatisfied party can seek reconsideration of a wrongly decided question. As our rules suggest, it makes sense to construe reported panel decisions as “binding” on subsequent panels only because en banc reconsideration always is available to correct panel mistakes. See 6 Cir. R. 35(a) and 206(c). Whereas the rule precluding subsequent panels from overturning prior reported panel decisions promotes consistency and reliability, this availability of review functions as a safety valve that allows for the vindication of the interests of accuracy and justice. See, e.g., Adkins v. Wolever, 520 F.3d 585, 587-88 (6th Cir.2008) (following wrongly decided prior panel decision but “encouraging] the other members of our Court ... to revisit the issue en banc”), rev’d en banc, 554 F.3d 650 (6th Cir.2009). Without the potential for review by higher courts, the dictates of justice and practicality would preclude us from construing the deference we owe to prior panel decisions as strictly binding. See Burnet, 285 U.S. at 406-07, 52 S.Ct. 443 (Brandeis, J., dissenting) (recognizing that stare decisis is “usually the wise policy” because consistency and reliance interests are significant, but explaining that the Supreme Court has never hesitated to overrule prior decisions where “correction ... is practically impossible” (citations and footnotes omitted)).2

However, this obligation to defer to “higher” courts is complicated when we sit in diversity. In diversity cases, we are obliged to apply state law and, therefore, are required to defer to the decisions of state courts. See Erie, 304 U.S. at 78-80, 58 S.Ct. 817; Bovee v. Coopers & Lybrand C.P.A., 272 F.3d 356, 361 (6th Cir.2001). In fact, federal courts oftentimes confront questions of state law on which the highest court of the state has not spoken, and thus are called upon to “predict” what state law would be by looking to the decisions of the state’s intermediate and lower courts. See Prestige, 99 F.3d at 1348; see also King, 333 U.S. at 160-61, 68 S.Ct. 488 (holding that the decrees of “lower state courts” are not “controlling” but must be “attributed some weight”). Consequently, in addition *626to the “pragmatic considerations” underlying stare decisis, we are obliged in diversity cases to take into account the “mutual respect” and “comity” we owe state courts in determining and applying state law. See Burgess v. Seligman, 107 U.S. 20, 33-34, 2 S.Ct. 10, 27 L.Ed. 359 (1883).

These considerations add a dimension of complexity to our stare decisis obligations because state courts are “co-ordinate” courts, neither superior nor inferior, in our federalist system. See id. at 33, 2 S.Ct. 10 (“The Federal courts have an independent jurisdiction in the administration of State laws, co-ordinate with, and not subordinate to, that of the State courts[.]”). As the Supreme Court has explained, a federal court sitting in diversity is, “in effect, only another court of the State.” Guaranty Trust Co. v. York, 326 U.S. 99, 108, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945). This distinction is crucial. In cases involving questions of federal law, the only courts with the authority to overturn or modify a panel decision are superior federal courts. See Darrah, 255 F.3d at 309-10 (citing Salmi 774 F.2d at 689, and 6 Cir. R. 206(c)). In diversity cases, however, the decisions of “co-ordinate” state courts can force us to reconsider a prior panel’s pronouncement of state law. See Blaine Constr., 171 F.3d at 350 (recognizing that stare decisis does not apply where there is “an indication by the [state] courts that they would have decided [the issue] differently”); Big Lots, 302 Fed.Appx. at 427 (recognizing that a panel of this Court is “bound” by prior reported panel decisions “unless [state] law has measurably changed in the meantime”). This unique dynamic casts serious doubt on the majority’s suggestion that the certification procedure runs afoul of our obligation to follow Andrews.

In addition, diversity cases also include another unique feature: namely, the potential for us to seek guidance (via the certification procedure) from the court to which we must defer. With respect to questions of federal law, subsequent panels must follow prior precedent, even if wrongly decided, and are limited to encouraging en banc reconsideration. See, e.g., Adkins, 520 F.3d at 587-88. In diversity cases, however, the certification procedure provides a unique opportunity for a subsequent panel to request input from a court with the authority to reexamine and overturn otherwise-controlling prior precedent. See Bush v. Gore, 531 U.S. 98, 139, 121 S.Ct. 525, 148 L.Ed.2d 388 (2000) (Ginsburg, J., dissenting) (“Notwithstanding our authority to decide issues of state law underlying federal claims, we have used the certification devise to afford state high courts an opportunity to inform us on matters of their own State’s law because such restraint ‘helps build a cooperative judicial federalism.’ ” (quoting Lehman Bros. v. Schein, 416 U.S. 386, 391, 94 S.Ct. 1741, 40 L.Ed.2d 215 (1974))). Because the certification procedure allows us to appeal to a court to which we must defer in this context, we are not disregarding our obligations under stare decisis because we are not modifying, overturning, or otherwise refusing to follow prior precedent.3

*627Both of these unique features suggest that the traditional precepts that support transforming stare decisis from, a pragmatic doctrine into a binding rule in the context of co-equal panels of the same circuit do not apply so easily in diversity cases. See Burgess, 107 U.S. at 33 (recognizing that “[t]he existence of two co-ordinate jurisdictions in the same territory is peculiar,” and emphasizing that “mutual respect and deference” are necessary to avoid “anomalous and inconvenient” results). For example, the Supreme Court has rejected the notion that inferior courts must defer to superior courts with respect to the resolution of issues of “purely local law,” reasoning instead that the decisions of district court judges and circuit panels must be accorded greater deference in this context because they generally have greater experience interpreting a given state’s laws. See MacGregor v. State Mut. Life Assurance Co., 315 U.S. 280, 281, 62 S.Ct. 607, 86 L.Ed. 846 (1942). The Court has justified this inversion of the traditional hierarchy of deference by reasoning that, when the federal courts are called upon to apply state law, “they act ... as ‘outsiders’ lacking the common exposure to local law which comes from sitting in the jurisdiction.” Lehman Bros., 416 U.S. at 391, 94 S.Ct. 1741. This rationale is all the more compelling where, as here, the state’s highest court has yet to address an issue directly and thus the federal courts are called upon to “predict” what that court would do. See, e.g., R.R. Comm’n v. Pullman Co., 312 U.S. 496, 499, 61 S.Ct. 643, 85 L.Ed. 971 (1941) (observing, in a critical self-assessment, that even the justices of the Supreme Court are “outsiders without special competence in Texas law,” and thus “we would have little confidence in our independent judgment regarding the application of that law to the present situation”).

As a result of this inversion of deference, it is impractical to expect correction by a superior federal authority. See, e.g., MacGregor, 315 U.S. at 281, 62 S.Ct. 607. This imposes on the lower federal courts a greater responsibility to make sure that questions of state law are “settled right,” not that they are just “settled.” For that reason, even if our certifying a question to the Supreme Court of Ohio could be construed as our refusing to follow Andrews, such a deviation from our traditional stare decisis obligations is justified by our “great responsibility” to apply state law properly.

III.

Because there is significant data to suggest that the Supreme Court of Ohio would reject the rule of Lone Star, I believe that the proper course of action is to certify a question to the Supreme Court of Ohio and allow that court to weigh in on this issue directly. I recognize that we owe deference to, and generally would be bound by, Andrews because it is a prior published decision that is almost directly on point. See Darrah, 255 F.3d at 309; Blaine Constr., 171 F.3d at 350. But because we are sitting in diversity, we also have an obligation to determine state law accurately. While there may not be enough “data” to justify overturning Andrews, there is more than enough basis to justify certifying a question to the Supreme Court of Ohio to permit that court an opportunity to resolve the issue.

The United States Supreme Court has recognized that, where state law makes the certification procedure available, the decision to certify a question to á state *628court in a given case “rests in the sound discretion of the federal court.” Lehman Bros., 416 U.S. at 391, 94 S.Ct. 1741. The rules of the Supreme Court of Ohio provide for certification of questions from federal courts, see Ohio S.Ct. R. XVIII, and that court repeatedly has accepted certified questions from federal courts in the past, e.g., Genaro v. Cent. Transp., 84 Ohio St.3d 293, 703 N.E.2d 782 (1999). The question, then, is whether this is a proper case for exercising our considerable discretion in this regard. I believe it is.

Although “the mere difficulty in ascertaining local law is no excuse for remitting the parties to a state tribunal,” comity, cooperative federalism, and practical considerations regarding judicial economy are factors that legitimately may be considered in determining whether to certify a question to a state supreme court. Lehman Bros., 416 U.S. at 390-91, 94 S.Ct. 1741. Those factors support utilizing the certification procedure in this case. Although this Court concluded in Andrews that Lone Star properly reflects the state of Ohio law on this issue, for the reasons explained above, significant evidence suggests that that is not true. That reason alone is sufficient to warrant certifying this question to the Supreme Court of Ohio. See Bush, 531 U.S. at 139, 121 S.Ct. 525 (Ginsburg, J., dissenting) (encouraging certification as a means “to afford state high courts an opportunity to inform us on matters of their own State’s law”).

In rejecting this suggestion, the majority points to case law in this circuit suggesting that we should defer to Andrews because state law has not “measurably changed in the meantime.” See Maj Op. at 619 (quoting Big Lots, 302 Fed.Appx. at 427). But that is not the only context in which we may revisit prior published decisions construing state law. As courts have recognized in similar contexts, it may be appropriate to set aside prior published precedent not only where “the holding had been superseded by a later pronouncement from state legislative or judicial sources,” but also where “prior state court decisions had been inadvertently overlooked.” Factors Etc., 652 F.2d at 283 (emphasis added).

IV.

Therefore, I respectfully dissent.

Notes

. In Pedler, the Supreme Court of Ohio ultimately found that the claimant could be denied coverage based on information contained in the "plan summary” and "other material outlining his ineligibility,” holding that "[t]he failure of the appellee to specifically note the limitation on the face of the certificate is not sufficient, standing alone, to estop *621the denial of coverage where the insured had the means of acquiring knowledge of the limitation.” 490 N.E.2d at 608. In reaching that conclusion, however, the Supreme Court of Ohio subjected a written insurance contract to an equitable estoppel inquiry. The fact that the court ultimately determined that the claimant could not prevail on that claim should not distract us from the critical fact that the court recognized that the doctrine of equitable estoppel applies with equal force to written contracts.

. Justice Brandéis’ argument is limited to cases "involving the Federal Constitution” where "legislative action is practically impossible” only because legislative correction of mistakes involving matters of statutory interpretation is much more feasible than amending the Constitution, as would be required to effect a non-judicial correction of a mistake involving a matter of constitutional interpretation.

. But see Lee, 592 F.2d at 272 ("Regardless of any ambiguity the plaintiffs may find in Louisiana cases to justify such a certification, there is no ambiguity as to this Court’s view of Louisiana law because the legal issue has been squarely resolved against plaintiffs’ precise arguments by two recent Fifth Circuit decisions. Once a panel of this Court has settled on the state law to be applied in a diversity case, the precedent should be followed by other panels without regard to any alleged existing confusion in state law, absent a subsequent state court decision or statutory amendment which makes this Court’s decision clearly wrong.” (citations omitted)). In Lee, however, the Fifth Circuit did not con*627front the same tension we face here because that panel determined that "the analysis in our prior decisions appears to us to be correct.” Id.