In this diversity case, the Allstate Insurance Company (“Allstate”), as subrogee of Sam Lakhia, appeals from the district court’s dismissal of its claim for property damage founded on the doctrine of strict liability for a defective product. There is a disagreement among the intermediate courts of appeals in Illinois with respect to the applicable statute of limitations. The United States District Court for the Northern District of Illinois decided that it should follow the rule established by the state intermediate court for the state appellate district in which the lawsuit would have been brought. Accordingly, it determined that the present lawsuit had been filed beyond the time permitted by the statute of limitations and dismissed the action. Because we believe that the district court erred in its belief that it was required to follow the rule of the state intermediate court of appeals for the state appellate district in which the suit would have been brought, we reverse its decision on that issue. Finally, because of the importance of the question to the jurisprudence of Illinois, we certify to the Supreme Court of Illinois the issue of the applicable statute of limitations.
I
BACKGROUND
Prior to December 18,1994, Sam Lakhia or a member of his family purchased a torchiere halogen lamp in a Menards store in Hillsdale, Illinois. The lamp was placed in the family home in Bellwood, Illinois and situated along the south wall near the basement stairs for the purposes of providing ambient light. On December 18, there was a fire in the Lakhia home that resulted in a claim by Mr. Lakhia for $144,799.05 for property damage and related living expenses incurred as a result of the fire. Allstate, Lakhia’s insurer and subrogee, paid the claim and then brought this action against Menards on March 10, 1999. The jurisdiction of the district court was based on the diverse citizenship of the *632parties.1 Menards filed a motion to dismiss on the ground that the action was barred by the applicable statute of limitations.
In a hearing before the district court, the parties disagreed with respect to the applicable statute of limitations period for a products liability action. In Allstate’s view, the applicable statute of limitations was five years. It relied upon the decision of the Second District in American Family Insurance Co. v. Village Pontiac-GMC, Inc., 182 Ill.App.3d 385, 131 Ill.Dec. 484, 538 N.E.2d 859 (1989). Menards, on the other hand, submitted that the applicable statute of limitations was two years. It relied on the decisions of the First District in McLeish v. Sony Corp. of America, 152 Ill.App.3d 628, 105 Ill.Dec. 648, 504 N.E.2d 933 (1987) and Calumet Country Club v. Roberts Environmental Control Corp., 136 Ill.App.3d 610, 91 Ill.Dec. 267, 483 N.E.2d 613 (1985).2
In discharging their responsibility to ascertain the content of state law in diversity cases, the sitting judges of the Northern District of Illinois have developed two conflicting approaches when confronted with no controlling decision of the Supreme Court of Illinois and with conflicting decisions of the Illinois Appellate Court.3 Some judges have followed the standard *633practice for a district court sitting in diversity and have attempted to predict how the Supreme Court of Illinois would resolve the question.4 Others have deemed themselves bound by the prevailing rule of the state appellate district in which the suit would have been brought in state court,5 reasoning that this approach would prevent forum shopping. The district court in this case had followed the latter approach in earlier cases and decided that, in the absence of explicit guidance from this court, it would not alter its course. Accordingly, it followed the decisions of the First District and held that the applicable statute of limitations was two years and dismissed the action.6
II
DISCUSSION
A.
We first address whether the district court was correct in its decision to consider itself bound by the view of the state intermediate appellate court with jurisdiction over the place where the action would be brought in the state court. Both parties before us, and all of the judges of the United States District Court for the Northern District of Illinois who have addressed the matter, acknowledge that the ultimate responsibility of the district courts is to apply the law of the state in which the court sits with respect to substantive matters. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). It is further beyond dispute that, as a general rule, statutes of limitation are considered to be substantive matters for purposes of the Erie doctrine. See Guaranty Trust v. York, 326 U.S. 99, 110, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945).
As we have noted already, the district court, relying on its past practice as well as the practice of some other judges sitting in this particular district, took the view that, in the absence of a definite ruling by the Supreme Court of Illinois and faced with a conflict among the state’s intermediate appellate courts, it was obliged to follow the rule established by the Illinois Appellate Court with jurisdiction over the *634place where the suit would have been filed, if it had been filed in state court. The view of the district court finds support in several published cases of the Northern District of Illinois. See, e.g., Abbott Labs. v. Granite State Ins. Co., 573 F.Supp. 193 (N.D.Ill.1983); Comm’l Discount Corp. v. King, 552 F.Supp. 841 (N.D.Ill.1982). As Menards points out in its brief, these cases consider the main focus of Erie to be a concern that “ ‘for the same transaction the accident of suit by a nonresident litigant in a federal court instead of a State court a block away should not lead to a substantially different result.’ ” Abbott Labs., 573 F.Supp. at 197 (quoting Guaranty Trust, 326 U.S. at 109, 65 S.Ct. 1464). The same theme, noted the district court in Abbott Laboratories, was expressed by the Supreme Court in Klaxon v. Stentor Manufacturing Co., 313 U.S. 487, 496, 61 S.Ct. 1020, 85 L.Ed. 1477 (1941), when the Supreme Court wrote that, unless state choice of law rules are treated as anything other than substantive, “the accident of diversity of citizenship would constantly disturb equal administration of justice in coordinate state and federal courts sitting side by side.” Klaxon, 313 U.S. at 496, 61 S.Ct. 1020. The rule articulated in Klaxon dealt with a state’s choice of law rules governing when it would apply its own substantive law and when it would apply the substantive law of another state. The district court in Abbott Laboratories maintained, however, that Erie’s concern with the equal administration of justice requires that the same approach be taken with respect to a state’s “internal choice of law rule that requires the state trial courts in Illinois to follow the rule of the appellate court for the district in which it is located.” 7 Abbott Labs., 573 F.Supp. at 195. Thus, when “‘state law’ itself contains a definitive rule as to the way to ascertain ‘state law’ in case of intrastate appellate court disputes, we must follow that first ‘state law’ in order to learn the second ‘state law.’ ” Comm’l Disc. Corp., 552 F.Supp. at 851. In short, said the district court, the “essential theory” of Erie is that “ ‘a federal court must decide substantive questions in diversity cases in the same way that a state trial judge in the same location would.’ ” Id. at 847 (quoting Nat’l Can Corp. v. Whittaker Corp., 505 F.Supp. 147, 148 n. 2 (N.D.Ill.1981)).
In assessing this perspective, we believe that the most appropriate course is to start with Erie itself. There can be no doubt that, in setting forth the rule that a district court sitting in diversity must apply the law of the state in which it sits, the Supreme Court was concerned with forum shopping. See Erie, 304 U.S. at 74-75, 58 S.Ct. 817. It is necessary, however, to go behind this concern and to perceive the reasons for the Supreme Court’s concern. Erie is not simply the product of a search for a salutary rule of judicial administration, but a holding that the Supreme Court explicitly said was constitutionally corn-*635pelled: “If only a question of statutory construction were involved, we should not be prepared to abandon a doctrine so widely applied throughout nearly a century. But, the unconstitutionality of the course pursued has now been made clear, and compels us to do so.” Erie, 304 U.S. at 77-78, 58 S.Ct. 817. Noting the federal government’s lack of constitutional authority to create substantive rules of law, the Supreme Court viewed its holding as restoring to the state governments the prerogative of fashioning principles of law in matters beyond the limited constitutional competence delineated for the federal government in the Constitution:
Except in matters governed by the Federal Constitution or by acts of Congress, the law to be applied in- any case is the law of the State. And, whether the law of the State shall be declared by its Legislature in a statute or by the highest court in a decision is not a matter of federal concern. There is no general federal common law. Congress has no power to declare substantive rules of common law applicable in a State whether they be local in their nature or “general,” be they commercial law or part of the law of torts. And no clause in the Constitution purports to confer such a power upon the federal courts.
Id. at 78, 58 S.Ct. 817; see also Bernhardt v. Polygraphic Co. of Am., 350 U.S. 198, 202, 76 S.Ct. 273, 100 L.Ed. 199 (majority), 208 (Frankfurter, J., concurring) (1956). To maintain the constitutional line between limited federal lawmaking power and state residual lawmaking power, Erie requires that the federal court follow an analytical path that ensures that the law applied in federal proceedings is the law that “ultimately would be applied were the case to be litigated in the state courts.” Roberts v. Western-Southern Life Ins. Co., 568 F.Supp. 536, 539-40 (N.D.Ill.1983) (emphasis in original).8
Application of the approach employed by the district court in this case would not achieve this result because it would deny the litigant any opportunity to receive a judicial assessment about what the supreme court of the state might determine the law to be through a resolution of conflicting precedent in the lower state appellate courts. The litigant in an Illinois state court is no doubt bound by the approach articulated in People v. Thorpe, 52 Ill.App.3d 576, 10 Ill.Dec. 351, 367 N.E.2d 960, 963 (1977), and Garcia v. Hynes & Howes Real Estate, Inc., 29 Ill.App.3d 479, 331 N.E.2d 634, 636 (1975), and therefore can expect to be bound in the trial court by the substantive rule of law articulated by the state appellate court for the district in which the trial court is located. Upon rendition of judgment by the trial court, however, that litigant has the right to appeal to the intermediate appellate court and to convince that court that the prevailing rule ought to yield to that embraced by an appellate court in another district. Absent satisfactory relief in that court, yet a further attempt may be made to convince the state’s highest tribunal to resolve the conflict among the districts. For the litigant in federal court, however, there is no such direct path to the Supreme Court of Illinois. If the mandate of Erie is to be satisfied and the law ultimately employed is to be the law of the state, the federal court, exercising its authority to hear diversity cases,9 must make a predictive judgment as to how the supreme court of the state would decide the matter if it were presented presently to that tribunal.
*636In many instances, therefore, application of the rule employed in Abbott Laboratories would result in the precedent of the state intermediate appellate court being given more effect in federal court than it would receive in the state judicial system. Under the rationale of that case, both the district court and this court would be bound by the precedent of the state intermediate appellate court. If the case were brought in the state system, however, that court would be free to reevaluate its own precedent and determine that, despite the usual constraints of stare decisis and precedent, that decision ought not stand. See Geri J. Yonover, Ascertaining State Law: The Continuing Erie Dilemma, 38 DePaul L.Rev. 1, 33 (1988).
Although the Supreme Court of the United States has not had to confront the precise issue before us today, its decisions make it clear that the Court views the Erie doctrine as imposing on federal courts the responsibility to determine the content of state law at the state, not the local, level. Eñe itself appears to have contemplated that the duty of the federal court, sitting in diversity, is to determine the content of state law as the highest court of the state would interpret it. See Erie, 304 U.S. at 78, 80, 58 S.Ct. 817; see also Wichita Royalty Co. v. City Nat'l Bank, 306 U.S. 103, 59 S.Ct. 420, 83 L.Ed. 515 (1938). In Bernhardt, the Court also focused on the clarity of the entire state’s decisional law in ascertaining the content of state law. See Bernhardt, 350 U.S. at 205, 76 S.Ct. 273. Most recently, and perhaps most pointedly, in holding that federal courts of appeals must review de novo the district court’s determination of the content of state law, the Supreme Court noted that deferential review invites “divergent development of state law among the federal trial courts even within a single state.” Salve Regina College v. Russell, 499 U.S. 225, 234, 111 S.Ct. 1217, 113 L.Ed.2d 190 (1991).
Finally, we do not think that the Supreme Court’s. decision in Klaxon is of any help to Menards. A fair reading of Klaxon makes clear that its focus is on ensuring that, in determining whether to apply the substantive law of the state in which it sits or the substantive law of another state, a district court should follow the choice of law rules of the forum state to ensure that the replacement of one state’s law by another is based on a state policy decision and not a federal one. As the Court stated:
Any other ruling would do violence to the principle of uniformity within a state, upon which the Tompkins decision is based. Whatever lack of uniformity this may produce between federal courts in different states is attributable to our federal system, which leaves to a state, within the limits permitted by the Constitution, the right to pursue local policies diverging from those of its neighbors. It is not for the federal courts to thwart such local policies by enforcing an independent “general law” of conflict of laws. Subject only to review by this Court on any federal question that may arise, Delaware is free to determine whether a given matter is to be governed by the law of the forum or some other law.
Klaxon, 313 U.S. at 496-97, 61 S.Ct. 1020.10
As one scholar has written:
All that was decided in Klaxon was that the federal court in Delaware should *637follow the conflict of laws rules prevailing in the state in which it sits. The conflict of laws rules contemplated by Klaxon are those rules, whether they be traditional First Restatement of Conflict of Laws method or one of the more modern approaches, which would direct the forum jurisdiction to apply the law of another jurisdiction. The intrastate precedential effect of intermediate state court decisions on state courts as exemplified by Thorpe/Garcia is an apple, or better yet an asparagus, compared to Klaxon’s orange!
Yonover, supra at 35.
Although we believe that the task of the federal court sitting in diversity is to ascertain the substantive content of state law as it either has been determined by the highest court of the state or as it would be by that court if the present case were before it now, we pause to emphasize that this determination in no way implies any erosion of our precedent that, in the absence of prevailing authority from the state’s highest court, federal courts ought to give great weight to the holdings of the state’s intermediate appellate courts and ought to deviate from those holdings only when there are persuasive indications that the highest court of the state would decide the case differently from the decision of the intermediate appellate court. See State Farm Mut. Auto. Ins. Co. v. Pate, 275 F.3d 666, 669 (7th Cir.2001); Lexington Ins. Co. v. Rugg & Knopp, Inc., 165 F.3d 1087, 1090 (7th Cir.1999); Allen v. Transamerica Ins. Co., 128 F.3d 462, 466 (7th Cir.1997). See generally E. Chemerinsky, Federal Jurisdiction § 5.3 at 323-26 (3d ed.1999) (discussing Supreme Court authorities); Yonover, supra at 5 n. 21. As the Supreme Court has held, “[wjhere an intermediate appellate state court rests its considered judgment upon the rule of which it announces, that is a datum for ascertaining state law which is not to be disregarded by a federal court unless it is 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, 237, 61 S.Ct. 179, 85 L.Ed. 139 (1940).11
In sum, we adhere today to the general rule, articulated and applied throughout the United States, that, in determining the content of state law, the federal courts must assume the perspective of the highest court in that state and attempt to ascertain the governing substantive law on the point in question. See Allen, 128 F.3d at 466-67. See generally 19 Charles Alan Wright, Arthur R. Miller & Edward H. Cooper, Federal Practice and Procedure: Jurisdiction 2d § 4507 at 157-61.
B.
As we have noted earlier, federal courts have, under Article III, the authority and, indeed, the duty to ascertain the content of state substantive law when called upon to do so in the exercise of the diversity jurisdiction of the federal courts. Exercising this responsibility is no easy task. As one jurist-author has noted, “[fjinding the applicable state law ... is a search that often proves elusive.”12 Moreover, although of*638ten lost in the debate about the continued importance of diversity jurisdiction in our constitutional scheme — a debate that usually focuses on the number of cases involved — there is a significant impact on the health of our federalism as federal courts interpret, with increasing frequency, issues of state law.13 As the same author pointed out, there is some evidence that federal courts are not the best prognosticators of the future course of state doctrinal issues.14 Unless and until corrected, these missteps by federal courts can have a grave impact on the principled and orderly growth of state law principles. These decisions “inevitably skew the decisions of persons and businesses that rely on them and inequitably affect the losing federal litigant who cannot appeal the decision to the state supreme court; they may even mislead lower state courts that may be inclined to accept federal predictions as applicable precedent.” 15 It may be that, in the process of ascertaining the content of state law, federal judges make, on occasion, significant contributions to the growth of state jurisprudence.16 Yet, the potential for significant intrusion, sometimes with disastrous results, counsels that the task be undertaken with great care, thoroughness and a full realization of the impact that the process, even when executed adroitly, has on “Our Federalism.” See Younger v. Harris, 401 U.S. 37, 44, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971).
Certification of a controlling issue of state law to the highest court of the state is one method of reducing the possibility of error in making an “Erie guess.” Many states have now enacted, through their state constitutions, statutes, or rules of court, specific provisions that permit a federal appellate court, and sometimes a federal district court, to seek such guidance.17 In this circuit we are the fortunate recipients of this option by all three states within the circuit. See Ind. R.App. P. 64; Ill.S.Ct. Rule 20; Wise. Stat. ch. 821.01. However, as we pointed out recently in Pate, although certification is a “useful tool of cooperative federalism,” Pate, 275 F.3d at 671, because it permits the federal court to obtain a definitive ruling on the content of state law, see City of Houston v. Hill, 482 U.S. 451, 470, 107 S.Ct. 2502, 96 L.Ed.2d 398 (1987), the use of the procedure is “not without its costs to the litigants and to the state court which already must contend with a crowded docket of its own.” Pate, 275 F.3d at 671. Consequently, we wrote, “a respect for the burdens of our colleagues on the state bench and concern for the litigants before us counsel that we approach the decision to certify with circumspection.” Id. We then set out a series of factors that federal courts ought to consider before deciding to certify an uncertain question of state law. See id. at 672-73. First, we said that the federal court ought to weigh the degree of uncertainty that exists on the state law issue. See id. We noted that, in the past, our cases have directed that a court also consider whether the issue presents a matter of public concern, whether the issue is likely to recur, whether the issue is outcome determinative, and whether the state supreme court has had an opportunity to *639address the issue in other cases. See id. at 672.18 We must also consider whether the supreme court of the state would consider the issue an important one in the growth of the state’s jurisprudence19 and whether resolution of the issue definitively will benefit future litigants20 or will almost exclusively affect the citizens of the state.21 See Pate, 275 F.3d at 672. Notably, we also said that, in determining whether an intrusion on the time of our colleagues on the state court is justifiable, we shall be more inclined to certify the question when the intermediate courts of the state are in disagreement on the issue or the issue is one of first impression for the court of last resort. See id. On the other hand, we have said that fact specific issues, as well as issues upon which there is no serious doubt or that are not dispositive of the case, are not candidates for certification. See id.
After giving the matter considerable study, we believe that the present case justifies certification to the Supreme Court of Illinois. Statutes of limitations reflect significant policy choices by the state and have grave consequences for the administration of justice within the state. Certainly, our colleagues on the Supreme Court of Illinois are far more familiar with the policy choices that have been made, and have far more direct responsibility for the administration of justice within the state than do the members of this court. Moreover, as we have noted earlier, the intermediate appellate courts in the state are in disagreement about the applicable statute of limitations. The intrusion of unguided federal precedent into the situation will only serve to destabilize further the state of Illinois jurisprudence on this issue and make it more difficult for members of the Illinois bar to counsel their clients accurately. The issue is, moreover, a recurring one and is likely to arise with significant frequency both in state and federal forums. To this date, moreover, it appears that the Supreme Court of Illinois has not had an opportunity to address the question squarely. Before deciding to certify this issue, we asked counsel to confirm that none of the intermediate court decisions on this issue have been presented to the Supreme Court of Illinois.
Conclusion
Accordingly, pursuant to Circuit Rule 52 and Rule 20 of the Supreme Court of Illinois, having determined that there is no controlling precedent from the Supreme Court of Illinois on an issue that may be determinative of this litigation, we respectfully certify the following question:
What is the applicable statute of limitations in Illinois for an action for damage to property based on the doctrine of strict liability in tort when that action is brought within the applicable statute of repose?
Question CeRtified.