Judge, dissenting.

The effect of today’s decision is to forever disable the district courts of this circuit from exercising their discretion to entertain declaratory judgment actions brought by insurers in cases already pending in state courts. That result is particularly unfortunate because it could have been avoided had the majority not been so anxious to take the initial determination of jurisdiction out of the hands of the district court where, in my view, it belongs. I would have remanded the case to the district court with instructions to reconsider the matter in light of Chamberlain v. Allstate Co., 931 F.2d 1361 (9th Cir.1991). But the majority prefers to conduct that review itself. Because I disagree with that action, as well as with the analysis that is its product, I dissent.

This court reviews de novo a district court’s decision to exercise its jurisdiction under the Declaratory Judgments Act, 28 U.S.C. § 2201 (1988) (“the Act”). Chamberlain, 931 F.2d at 1366 . It follows, the majority concludes, that we have the option to review the district court’s jurisdictional decision now or “remand and delay doing so until after another appeal is taken.” Majority at 1370. (citing Allstate Ins. Co. v. Mercier, 913 F.2d 273, 277 (6th Cir.1990)). It chooses the latter course, assuming that if we remand this case, whatever action the district court takes will be appealed and we will thus “still ultimately be obligated to *1375 exercise our own discretion in considering the propriety of the district court’s grant or denial of declaratory relief.” Majority at 1370. This is a great victory for efficiency, but a defeat for the integrity of our federal court system.

We do not know why the district court chose to exercise its jurisdiction. If it considered the factors that Chamberlain derived from Brillhart v. Excess Ins. Co., 316 U.S. 491 , 62 S.Ct. 1173 , 86 L.Ed. 1620 (1942) for determining when district courts may properly exercise jurisdiction over suits brought under the Act, it developed no record of that consideration. This court, therefore, has no record upon which to rely in conducting its own evaluation of those factors. But that is of no concern to the majority. How ironic it is that at the same time it draws on the lessons of Brill-hart to admonish the district court for interfering in the domain of the state courts, the majority ignores that decision’s counsel on the proper function of appellate courts.

In Brillhart , the district court refused to take jurisdiction over a declaratory judgment action brought by a reinsurer who had been made a party to a state garnishment proceeding. The court of appeals overruled the district court, holding that dismissal of the suit was an abuse of its discretion. Id. at 492-94 , 62 S.Ct. at 1174-75 . The Supreme Court reversed and remanded to the district court with instructions to reconsider its decision in light of the principles the court articulated in its opinion. Id. at 498 , 62 S.Ct. at 1177 .

The Court remanded, rather than deciding itself whether jurisdiction was proper, because the record suggested that the district court had not considered the factors the Court concluded were essential to that determination. Specifically, the district court had not considered “whether, under applicable local law, the claims sought to be adjudicated by the respondent in this suit for declaratory judgment had either been foreclosed by Missouri law or could adequately be tested in the garnishment proceeding pending in the Missouri state court.” Id. at 495-96 , 62 S.Ct. at 1176 . The Court made clear why the district court, not the Supreme Court, should make that determination in the first instance.

Id. at 496-97 , 62 S.Ct. at 1176 .

The circumstances of this case are no different from those of Brillhart and Justice Frankfurter’s reasoning counsels as strongly in favor of remand here as it did there. The district court in this case apparently never considered the Brillhart factors and thus never developed a record evaluating them. Such a record is invaluable to our own de novo review of the court’s exercise of jurisdiction. Indeed, the speculative nature of both the majority’s and my own application of the Brillhart principles to this case demonstrates the hazards that befall appellate courts when they deceive themselves that they are better equipped to do the job of trial courts than are trial courts themselves. I would remand to the district court with instructions to reconsider the matter in light of Chamberlain .

Though I would not review the district court’s jurisdictional determination without a proper foundation, the majority has chosen not to delay, and I therefore feel constrained to register my disagreements with its analysis. First, it is true, as the majority states, that when a federal declaratory judgment action is brought while a state court action is pending, there is a presumption that the entire matter should be heard in state court. Majority at 1370 (citing Chamberlain, 931 F.2d at 1366-67 ). Chamberlain goes on to make clear, however, that

Chamberlain, 931 F.2d at 1367 (quoting Brillhart, 316 U.S. at 495 , 62 S.Ct. at 1176 ). At the time appellant filed his motion for a stay in this case, the federal proceeding had progressed further than the parallel state proceeding and either court could have afforded complete relief. If these circumstances have not changed, on remand the district court might well conclude, despite the presumption against jurisdiction when a parallel state proceeding has been commenced, that concerns of judicial administration and fairness to the litigants weigh in favor of retaining jurisdiction.

In addition to these general concerns, Chamberlain derives from Brillhart several specific factors the district court should consider in deciding whether to exercise its jurisdiction. I object to certain aspects of the majority’s analysis of those factors. The first Brillhart factor identified in Chamberlain is the need “to avoid having federal courts needlessly determine issues of state law.” Chamberlain, 931 F.2d at 1367 . The majority is of the opinion that this factor counsels against jurisdiction because the same issue is at stake in the state proceeding, and California law provides the rule of decision. What the majority fails to note is that the first Brill-hart factor is implicated only when complex state law issues are presented. See id. (citing Brillhart, 316 U.S. at 497 , 62 S.Ct. at 1176 ).

There is no complex issue of state law presented in this case. Rather, the suit boils down to a factual dispute over whether, at the time the claim was filed, appellant had an ownership interest in the policy. The mere incantation of California’s “complex scheme of insurance regulation” cannot change that fact, nor can it make the policies of Burford abstention relevant to the majority’s analysis. See Pinhas v. Summit Health, Ltd., 880 F.2d 1108, 1115 (9th Cir.1989) (“Burford abstention is appropriate when a federal court is presented with difficult questions of state law bearing on policy problems of substantial public import whose importance transcends the result in the case then at bar.”) (citation and internal quotation omitted). Surely the majority does not mean to suggest that every breach of contract claim against an insurer requires a court to plumb the depths of some byzantine administrative apparatus. Because the district court was not called upon to decide unsettled questions of state law, I suggest that the first Brillhart factor has not been implicated.

Under the second Brillhart factor, the district court should consider “that parties could attempt to avoid state court proceedings by filing declaratory relief actions in federal court.” Chamberlain, 931 F.2d at 1367 . The majority relies on Transamerica Occidental Life Ins. Co. v. Digregorio, 811 F.2d 1249 , 1254 n. 4 (9th Cir. 1987), to characterize this factor “as relating to ‘the “defensive” or “reactive” nature of a federal declaratory judgment suit.’ ” Majority at 1371. It then suggests “that if a declaratory judgment suit is defensive or reactive, that would justify a district court’s decision not to exercise jurisdiction.” Id. While this position is not necessarily inconsistent with Digregorio , I am *1377 troubled by the majority’s attempt to translate the Digregorio analysis into a per se rule: “Whether the federal declaratory judgment action regarding insurance coverage is filed first or second, it is reactive, and permitting it to go forward when there is a pending state court case presenting the identical issue would encourage forum shopping in violation of the second Brill-hart principle.” Id. at 1363 .

I do not believe Digregorio supports such a rule. The case never actually discusses “reactive” litigation. Indeed, its only reference to “reactive” litigation appears in a parenthetical reference to Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U.S. 1 , 103 S.Ct. 927 , 74 L.Ed.2d 765 (1983) — a case not involving the Declaratory Judgments Act — in a footnote that stands for nothing more than “the limited proposition that Brillhart survives Colorado River and authorizes jurisdictional discretion in Declaratory Judgment [sic] Act suits.” 811 F.2d at 1254 n. 4.

All that Digregorio actually says about “forum shopping” is that “a district judge has discretion to decline jurisdiction in favor of pending state court litigation when a party seeks to use the Declaratory Judgment [sic] Act to deprive a plaintiff of his choice of forum or to encourage a race to judgment.” 811 F.2d at 1253 . This is consistent with Chamberlain’s statement of the second Brillhart factor as well as its suggestion that forum shopping should be avoided “by requiring district courts to inquire into the availability of state court proceedings to resolve all issues without federal intervention.” 931 F.2d at 1367 .

Appellee’s declaratory action was arguably “defensive.” That fact, however, is only one of several factors that a court must balance in determining whether to entertain an action for declaratory relief. It is not, as the majority suggests, disposi-tive.

Finally, the majority suggests that allowing the federal action to go forward would frustrate the third Brillhart factor: avoiding “[gratuitous interference with the orderly and comprehensive disposition of a state court litigation. 316 U.S. at 495 , 62 S.Ct. at 1176 . Because no substantial progress had been made in the state court action, I do not believe this factor is implicated. Furthermore, I question whether the presence of “Doe Defendants” in the state action has any bearing on the third factor, the policy disfavoring piecemeal litigation, or the diversity removal provision of 28 U.S.C. § 1441 (b), (c) (1988). I am at a complete loss as to how the presence of fictitious defendants in appellant’s state court action prevents a federal court from granting complete relief to the actual parties.

In my opinion, these flaws in the majority’s analysis substantially undermine its conclusion that it would be improper for the district court to exercise its discretion to take jurisdiction over appellee’s declaratory judgment action. More importantly though, these flaws demonstrate exactly why it is imperative to allow district courts to exercise their discretion in the first instance. Had we allowed the district court to consider the jurisdictional question in light of Chamberlain , it would have established a record illuminating the applicability of the Brillhart principles to the facts of this case and enabled us to engage in an intelligent de novo review of its determination. Without a record, we are left to fumble in the dark for the correct decision.

Perhaps my greatest concern, however, is the practical consequence of the majority’s analysis. The majority pays lip service to the discretion afforded district courts under Chamberlain , holding only that in this case “issuance of the declaratory judgment with respect to the insurance coverage dispute is contrary to the exercise of sound judicial discretion.” Majority at 1374. (emphasis added). But the reality is that the majority has established a per se rule against federal district courts taking jurisdiction over declaratory judgment actions brought by insurers. By invoking California’s “complex scheme of insurance regulation” without regard to the facts of this case, by ignoring the relative postures of the state and federal proceedings in the case, and by making the “defensiveness” of *1378 the declaratory judgment action dispositive, the majority leaves the district courts no room to exercise their discretion. Under its analysis, two of the three Brillhart factors will always counsel against jurisdiction, and the third will be dispositive in any case that can be deemed “defensive.” There is no escaping the conclusion that in the future, district courts will feel bound by this decision and choose not to take jurisdiction over declaratory judgment actions by insurers.

I respectfully dissent.