LEVAL, District Judge.

This is an action for declaratory judgment brought by the Travelers Indemnity Company of Illinois (Travelers), the issuer of comprehensive general liability insurance (CGL) covering Metropolitan Life Insurance Company (Met), to determine the obligations of Travelers to Met under the policies. Named as defendants are not only Met, the insured, but also National Union Fire Insurance Company of Pittsburgh, PA (National), the issuer of an umbrella and excess policy covering Met. Defendant Met moves to dismiss the action under Rule 12 for lack of diversity among the parties or alternatively, on grounds of forum non conveniens. National joins Travelers in opposing the motion.

Plaintiff Travelers is an Illinois corporation with its principal place of business in Connecticut. Defendant Met is a New York corporation with its principal place of business in New York. Defendant National is a Pennsylvania corporation with its principal place of business in New York. Thus, as the complaint is pleaded, there is complete diversity as between plaintiff and defendants. Met, however, contends that the parties have been misaligned. It contends National was improperly joined as a defendant, that National's interests are aligned with Travelers', and that National should therefore be realigned as a plaintiff. If this were done, the New York citizenship of plaintiff National and defendant Met would preclude diversity jurisdiction. Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267, 2 L.Ed. 435 (1806).

In 1989 and 1990, Travelers provided Met with CGL coverage up to a limit of $10,000,000 per occurrence with a yearly cap of $25,000,000. Travelers contends its policies expressly exclude Errors and Omissions (E O) coverage, which is a form of professional malpractice insurance, as well as coverage for losses arising out of the conduct of Met's insurance and related financial services businesses.

For the year 1989 National provided Met with "excess" umbrella coverage with a limit of $10,000,000. National contends its policy also excludes E O coverage. Insofar as pertains to this action, the scope of coverage and exclusion under the National policy are very similar to Travelers' although not written in identical words.

During 1990 and 1991, liabilities were asserted against Met by third parties in three Texas state court actions: Herzing, et al. v. Metropolitan Life Insurance Company, et al., No. C-1681-90-A, District Court of Hidalgo County, 92nd Judicial District; Bond, et al. v. Metropolitan Life Insurance Company, et al., No. C-2584-91-A, District Court of Hidalgo County, 92nd Judicial District; and Heward, et al. v. Metropolitan Life Insurance Company, et al., No. C-91-020325, District Court of Harris County, 190th Judicial District. The plaintiffs in these actions assert that in 1989 and 1990 a sales representative of Met, Donald Bryant, Jr., fraudulently promised that Met would arrange a series of "self-liquidating, non-recourse loans" for over $27,000,000 in return for the payment of commitment fees totaling approximately $82,000, that the commitment fees were paid but that the loans were never provided. The Texas suits allege claims in tort and contract, as well as violations of Texas statutes, and assert liability on the part of Met in the hundreds of millions of dollars.

Met demanded that Travelers defend the Texas claims. Travelers first denied any coverage or obligation. Travelers then agreed to defend the Herzing action and to pay Met's defense costs in the other two suits, in each case under a reservation of rights.

After providing the defense of the Herzing action for a substantial period of pre-trial proceedings, Travelers filed this declaratory judgment action seeking adjudication of its disclaimer of obligations.

One week later, Met filed an action in the Texas state court seeking declaratory judgment against Travelers, National, and International Insurance Company ("International"), the insurer that provided Met with umbrella coverage for 1990.

In this action, National has counter-claimed against Travelers, cross-claimed against Met and brought a third-party action against International. National makes numerous contentions, principally (i) that the Texas occurrences are not of the type that its policy covers; (ii) that if they are, they relate to 1990 when International rather than National provided coverage, and (iii) that, in any event, Travelers has primary coverage which must be exhausted before any obligation of National arises.

The leading authority on when to realign parties to determine the existence of diversity remains Justice Frankfurter's 1941 decision in City of Indianapolis v. Chase National Bank, 314 U.S. 63, 62 S.Ct. 15, 86 L.Ed. 47 (1941). The teaching of that opinion is, however, far from clear. Despite the passage of more than half a century, courts remain divided on how to apply the test.

Frankfurter, writing for the majority, stated that for diversity jurisdiction to be sustained, there must be an "actual, substantial controversy" between citizens of different states. Id. at 69, 62 S.Ct. at 16. Whether the necessary "collision of interests" is present, he continued, must be "ascertained from the principal purpose of the suit and the primary and controlling matter in dispute." Id. "It is [the federal courts'] duty," he added, ". . . to look beyond the pleadings and arrange the parties according to their sides in the dispute." Id. (citations omitted).

The first difficulty courts have had in attempting to apply the Indianapolis test is in deciding whether the existence of a colliding controversy between parties pleaded as adverse must be determined on the basis of the "principal purpose of the suit" ("primary and controlling matter in dispute") or whether it is sufficient to find "actual substantial controversy" between the parties regardless whether the controversy would be identified as the "principal purpose of the suit."

Justice Frankfurter unquestionably stated that the finding of collision must relate to the "primary and controlling matter in dispute." However, in that particular case, unlike some others, there was an identifiable primary and controlling matter in dispute as to which opposing parties had the same interests. Justice Frankfurter noted that the interests of two parties (although they were pleaded on adverse sides) turned identically on the validity or invalidity of a lease. Accordingly, the Court ruled that realignment was required, and as a result diversity was destroyed, depriving the federal courts of jurisdiction.

Although Justice Frankfurter seems to assert that the adversity must relate to the primary issue in dispute, he, nonetheless, in footnote 3, explored whether there were other matters in controversy between the two questioned parties that would justify regarding them as adversaries. He found that there were not. Opponents of the "primary purpose" test argue that if the primary purpose were indeed controlling, then there would have been no reason for Justice Frankfurter to explore the existence of conflict as to subordinate questions.

It is unclear how the Indianapolis rule should be applied in complex suits where there is real adversity between two adverse parties although not pertaining to the primary matter in dispute, and in suits where it is not possible to identify with clarity what is the principal purpose of the suit.

Several circuits (the 3rd, 5th, 6th and 9th) have declared in favor of the "primary purpose" test, reading the Indianapolis opinion literally. Employers Insurance of Wausau v. Crown Cork Seal Co., 942 F.2d 862 (3d Cir. 1991); United States Fidelity Guaranty Co. v. Thomas Solvent Co., 955 F.2d 1085 (6th Cir. 1992); see also Continental Airlines, Inc. v. Goodyear Tire Rubber Co., 819 F.2d 1519, 1523 n. 2 (9th Cir. 1987); Lowe v. Ingalls Shipbuilding, 723 F.2d 1173, 1178 (5th Cir. 1984); Cf. U.S.I. Properties Corp. v. M.D. Const. Co., Inc., 860 F.2d 1, 4 (1st Cir. 1988). cert. denied, 490 U.S. 1065, 109 S.Ct. 2064, 104 L.Ed.2d 629 (1989) ("our task is to determine the primary and controlling matter in dispute, and then determine whether any actual collision in interests remain").

The Seventh, Eighth and Tenth Circuits, in contrast, have interpreted City of Indianapolis to mean that realignment of parties by courts is improper if there is any "actual and substantial conflict" between the parties. American Motorists Insurance Co. v. Trane Co., 657 F.2d 146 (7th Cir. 1981); Farmers Alliance Mutual Insurance Co. v. Jones, 570 F.2d 1384 (10th Cir.), cert. denied, 439 U.S. 826, 99 S.Ct. 97, 58 L.Ed.2d 119 (1978); Universal Underwriters Insurance Co. v. Wagner, 367 F.2d 866 (8th Cir. 1966); Cf. Smith v. Sperling, 354 U.S. 91, 97-98, 77 S.Ct. 1112, 1115-16, 1 L.Ed.2d 1205 (1957) ("There is jurisdiction if there is real collision between [the parties]").

The Second Circuit has not ruled directly on which test to follow. In Lewis v. Odell, 503 F.2d 445 (2d Cir. 1974), the court ruled noncommittally that "federal courts are required to realign parties according to their real interests so as to produce an actual collision of interests." Id. at 447 (citing City of Indianapolis, 314 U.S. at 69, 62 S.Ct. at 16). District courts in the Second Circuit have endorsed a "collision of interest" test, rather than a "primary purpose" test. Syms, Inc. v. IBI Security Serv. Inc., 586 F. Supp. 53, 56 (S.D.N Y 1984); Am. Mutual Liability Ins. Co. v. Flintkote Co., 565 F. Supp. 843 (S.D.N.Y. 1983); Irving Trust Co. v. Century Export and Import, S.A., 464 F. Supp. 1232, 1241 (S.D.N.Y. 1979); Sands v. Geller, 321 F. Supp. 558, 562 (S.D.N.Y. 1971); See also Maryland Casualty Co. v. W.R. Grace Co., 88 Civ. 2613 (SWK) (Bernikow, M.J.) (Report and Recommendation, May 12, 1992).

A second, and at times equally troublesome problem, is how to appraise the existence or non-existence of "actual collision of interests." Should it be determined taking at face value the assertions of the pleadings? Should the court accept as "actual [and] substantial" any conflict asserted by counsel to exist, regardless whether stated in the pleadings, even though it may be hypothetical and conjured up only to control the issue of jurisdiction? Should the court inquire behind the pleadings and the assertions of counsel in an effort to test whether the asserted conflict is real or imaginary? In cases having some similarity to this one, the Seventh Circuit in American Motorists v. Trane, supra, and a judge of this District Court in Am. Mutual Liability Ins. Co. v. Flintkote, supra, have adopted the approach of accepting counsel's assertion of the existence of conflict at face value, without inquiry as to whether such conflict is real or hypothetical.

Examining the relationship between Travelers and National in the context of the litigation, one finds little indication of colliding interests. Met contends that as issuers of virtually identical insurance coverage (although at different levels), Travelers and National have a common interest in obtaining a ruling that their policies do not cover the incidents involved in the Texas litigation. Travelers' complaint against Met and National seeks only a declaration of Travelers non-liability to Met and makes no suggestion of adversity between Travelers and National; indeed, its only functional reference to National is the statement that National issued umbrella coverage to Met.

At oral argument of the motion, the court asked Travelers' counsel to describe the adversity between Travelers and National, as well as the circumstances that would bring the adversity to fruition. Travelers answered that because National's policy was of the umbrella type, National was exposed for both excess and primary coverage and would therefore have an interest in urging a court to find Travelers liable if National were found liable so that National would benefit from Travelers' primary layer of protection.

Accepting this theoretical adversity, counsel were asked whether there were identifiable differences in the language of coverage and exclusion that could arguably sustain different results as between National's coverage and Travelers'. Counsel could name none.

In the end, counsel advanced the following hypothetical adversity: In this New York court, Travelers might be held not liable; in a Texas court (through a ruling incompatible with the New York ruling), National might be held liable to Met; if National was not bound by Travelers' earlier established ruling of non-liability then National might seek a ruling imposing primary liability on Travelers. Thus the hypothetical adversity arises only from the danger that different courts might reach incompatible results on the same (or substantially the same) question, after National and Travelers initially each argued the same side of the question. This is an adversity that can arise only if Travelers and National are not present in the same litigation and two courts reach opposite results on what is functionally the same question. I conclude that it is an adversity that is more hypothetical and conjectural than real. For until this unlikely scenario is reached, both insurers will argue the same side of the question — both taking the position that Travelers' (and likewise National's) policy does not cover the Texas incidents.

Applying the various versions of the Indianapolis test to the particular degree of adversity that is present here, I find it does not matter whether the court adopts as a standard the "primary issue in dispute" or the existence of any "actual [and] substantial" collision of interests. In either case, Travelers' contention of adversity flunks the test. As to the primary question in the litigation, the coverage of the Texas cases by Travelers' policies, Travelers and National are aligned rather than adverse. Nor is there any conflict between them on any real and substantial issue in dispute. It is only if one goes as far as Trane and Flintkote to accept as sufficient a hypothetical, conjectural conflict asserted by counsel primarily for support on the jurisdictional argument that the relative positions of Travelers and National would pass the adversity test.

I cannot accept this approach of Trane and Flintkote, for two reasons. First, it seems incompatible with the Supreme Court's holding in Indianapolis. Notwithstanding the ambiguity of Indianapolis as to whether the justifying conflict must be found in the "primary" issue or may be found in subordinate issues, there is no ambiguity in the Supreme Court's insistence that the conflict be an "actual" and "substantial collision of interests." This requirement seems incompatible with an adversity that is unlikely to eventuate but is conjured up by counsel in order to achieve a jurisdictional objective.

Secondly, this approach lends itself to manipulation that can defeat the legitimate purposes of the diversity jurisdiction — to protect foreign litigants from regional bias. In almost any dispute involving several participants, no matter how identically aligned two litigants are in practical reality, an imaginative lawyer can describe circumstances in which there might be adversity between them, and use this artifice either to achieve federal jurisdiction or to defeat it in precisely the circumstances for which the diversity jurisdiction was intended. See Burgess v. Seligman, 107 U.S. 20, 34, 2 S.Ct. 10, 21, 27 L.Ed. 359 (1883); Smith v. Metropolitan Property and Liability Ins. Co., 629 F.2d 757 (2d Cir. 1980).

Assume for example the following hypothetical case: Driver, an employee of Company, both being citizens of Farm State, injures Pedestrian, a citizen of Industrial State. We would assume that Pedestrian is aligned on one side of the controversy while Driver and his Company are on the other. But suppose that Driver sues Pedestrian in the local courts of his own Farm State for a declaration of no negligence and names his employer Company as a defendant. Pedestrian removes to federal court insisting that Company and Driver should be realigned to the same side of the controversy. Driver answers that they are adverse and explains (in terms similar to those advanced by Travelers) as follows: Driver might obtain a judgment of non-liability to Pedestrian in Farm State; Pedestrian might sue Company at his home in Industrial State and win a judgment establishing Company's liability, based on Driver's negligence; Company might then sue Driver for indemnity, asserting that Driver's negligence caused Company's liability. Driver would insist that he must join Company in the initial suit, as a defendant, to protect himself by binding Company to whatever judgment the Farm State court renders on Driver's negligence.

That chain of events could indeed happen, and would result ultimately in adversity between Company and its Driver. But to allow such unlikely and conjectural events to determine adversity and thus destroy Pedestrian's access to the federal courts would substantially undermine the protections of diversity jurisdiction and would be incompatible with the Supreme Court's insistence in Indianapolis that the parties be aligned in such fashion that produces real, substantial conflict between plaintiffs and defendants. I therefore decline to accept the Trane/Flintkote extension of the controversy test.

The controversy depicted by Travelers, based on inconsistent rulings in Texas and New York, in its effort to gain federal court jurisdiction is barely less conjectural than that advanced by Driver in the hypothetical to defeat Pedestrian's access to federal court. Neither, in my view, satisfies the Indianapolis test.

I conclude that Travelers and National must be realigned to the same side of the controversy. Travelers seeks a ruling that its policies, by virtue of the E O exclusion, do not cover the risks asserted against Met in the Texas litigation. National, which extends similar coverage to Met through its umbrella policy, has a similar interest to Travelers' to obtain a ruling that Travelers' coverage, and National's similar coverage, do not extend to the Texas claims. Once National is realigned among the plaintiffs, its New York citizenship precludes a finding of diversity as defendant Met is also a citizen of New York.

Accordingly, Met's motion is granted. The action must be dismissed under Rule 12(b)(1) of the Federal Rules of Civil Procedure for lack of subject matter jurisdiction.

SO ORDERED.