This appeal presents a series of jurisdictional questions. The underlying dispute is a state law claim involving a mortgage foreclosure action. It gained entry to the federal system via the Financial Institutions Reform, Recovery, and Enforcement Act of 1989 (FIRREA), Pub.L. No. 101-73, 103 Stat. 183 (1989), which extends federal jurisdiction to "any civil action, suit, or proceeding to which the [Resolution Trust Corporation (RTC) ] is a party." 12 U.S.C. § 1441a(l)(1). Invoking this provision, the RTC filed suit in federal court, seeking foreclosure and related relief on various mortgages. The RTC subsequently sold its interest in the underlying loans and was dismissed from the case. The district court entered summary judgment against the debtor, who appeals on the ground that federal jurisdiction failed when the RTC was dismissed.

We will consider in turn the three jurisdictional issues raised by the parties. We begin by rejecting a mootness challenge to our appellate jurisdiction. We next determine that we do not have continuing jurisdiction under § 1441a(l)(1). We then reject the invocation of the "black letter rule" that jurisdiction is only determined at the time of the filing of the complaint. We conclude, however, by looking to the supplemental jurisdiction statute, 28 U.S.C. § 1367. Because we find that this particular case should fall within the supplemental jurisdiction of the district court, we will affirm the grant of summary judgment, even though we found that jurisdiction under § 1441a(l)(1) had terminated.

On December 23, 1987, Preferred Entity Advancements, Inc., ("Preferred Entity") and DAML Realty Corp. ("DAML") executed and delivered to BRT Realty Trust ("BRT") five promissory notes in the original amounts of $67,965,270; $9,447,920; $2,880,151; $706,659; and $4,000,000. These notes were secured by mortgages on tracts of land in New Jersey and New York. On the same date, BRT assigned a 95% interest in the loan to FarWest Savings and Loan Association ("FarWest").

The Notes and Mortgages contained clauses providing for acceleration and foreclosure in the event of default. On or about March 1, 1991, Preferred Entity and DAML defaulted on their obligations. Meanwhile, on February 15, 1991, the RTC had been appointed receiver for FarWest. On June 18, 1991, BRT and the RTC filed an action in the Superior Court of New Jersey, Chancery Division, Hudson County, seeking foreclosure and related relief on the mortgages secured by New Jersey property.

On September 23, 1992, the RTC acquired BRT's remaining 5% interest in the loan. The RTC then dismissed the New Jersey action and, on July 30, 1993, filed this suit in the United States District Court for the District of New Jersey. Jurisdiction rested on 12 U.S.C. § 1441a(l),1 which grants the federal courts jurisdiction over any proceeding to which the RTC is a party.

No discovery or further action took place. On October 5, 1994, New Rock Asset Partners, L.P., ("New Rock") acquired all of the RTC's right, title, and interest in the loans and mortgages. On December 9, 1994, New Rock filed an Amended Complaint stating, inter alia:

1. The Jurisdiction of this Court is invoked pursuant to the Financial Reform, Recovery, and Enforcement Act of 1989, 12 U.S.C. § 1441a(l).

....

3. Plaintiff, New Rock, is the sole owner and holder of all right, title and interest in the Indebtedness (as defined herein) and the right to repayment thereof, together with all of the collateral security granted for repayment of the Indebtedness, pursuant to the Mortgage Assignments (as defined herein).

....

31. New Rock has succeeded to all of RTC's right, title and interest in the Rent Order [obtained against defendants in a previous state court action].

App. at 28-29, 35. On December 14, 1994, New Rock obtained an order substituting itself as plaintiff and dismissing the RTC from the case.

Two day later, New Rock moved for partial summary judgment and final judgment of foreclosure. Preferred Entity responded by contesting subject matter jurisdiction and the certifications on which New Rock based its summary judgment motion. New Rock then supplemented its motion with additional certifications. The district court denied Preferred Entity's jurisdictional challenge and granted New Rock's summary judgment motion. Preferred Entity appealed.

Since the filing of the appeal, New Rock has executed on its judgment and purchased the New Jersey property at a sheriff's sale conducted on August 10, 1995, by the Sheriff's Office for Hudson County. New Rock is currently pursuing various actions in New York state courts to foreclose on the New York properties.II.

The propriety of federal court jurisdiction forms the nub of this case. The district court asserted jurisdiction based solely on 12 U.S.C. § 1441a(l). We exercise appellate jurisdiction over the district court's judgment and final order pursuant to 28 U.S.C. § 1291. Our review of the district court's jurisdictional determinations is plenary. Wujick v. Dale & Dale, Inc., 43 F.3d 790, 792 (3d Cir.1994).

We will first address New Rock's argument that this appeal has been "mooted in part" because the subject property has been purchased at foreclosure sale and the validity of the foreclosure sale can no longer be disputed. New Rock combines this argument with a citation to our decision in National Iranian Oil Co. v. Mapco Int'l, Inc., 983 F.2d 485 (3d Cir.1992), contending that this appeal is not moot as to the judgement's collateral estoppel and res judicata effects. This "moot in part, yet not moot in part" argument represents a fundamental misunderstanding of mootness doctrine and National Iranian Oil. We will attempt to clarify the matter.

In arguing mootness because of the foreclosure sale, New Rock relies on a Seventh Circuit case, Federal Deposit Ins. Corp. v. Meyer, 781 F.2d 1260 (7th Cir.1986), which states the rule in the Seventh Circuit: "In the absence of a stay of the enforcement of a judgment, if a district court judgment authorizes the sale of property and the property is sold to a good faith purchaser during the pendency of the appeal, the sale of property moots the appeal...." Id. at 1263. In the case before us, foreclosure and sale have taken place. New Rock contends that this aspect of the appeal is therefore moot. We do not agree.

We in the Third Circuit have never addressed the issue of whether foreclosure and sale, purely and simply, would render an appeal moot. It is possible that we might come to that conclusion in an appropriate case after examining the full effects on the dispute of such a foreclosure and sale. But, before so concluding, our precedents require that we first determine if there is still the possibility of granting any effective relief. See National Iranian Oil, 983 F.2d at 489 ("A case is not moot if there is a real and substantial controversy admitting of specific relief through a decree of conclusive character") (citations omitted); In re Joshua Slocum, Ltd., 922 F.2d 1081, 1085-86 (3d Cir.1990) (declining to moot appeal in landlord-tenant dispute where landlord failed to obtain stay; court could still grant effective relief); Main Line Fed. Sav. & Loan Ass'n v. Tri-Kell, Inc., 721 F.2d 904, 907 (3d Cir.1983) ("the determination that a case is moot requires that there be nothing gained by reaching a decision"); see also New Jersey Turnpike Auth. v. Jersey Cent. Power & Light, 772 F.2d 25, 30 (3d Cir.1985) (discussing mootness in terms of inability to grant effective relief). If effective relief can be granted, then this appeal is not moot.

We have commented in dictum on the possibility of effective relief remaining when a party has challenged a court's jurisdiction over a judgment used to foreclose property. In Raymark Indus., Inc. v. Lai, 973 F.2d 1125 (3d Cir.1992), we rejected the plaintiff's contention that because money that the defendant had posted in lieu of a supersedeas bond had been disbursed, meaningful relief was precluded and the defendant's appeal was moot. We noted that if the state court order was improper and void ab initio, then the defendant could seek to "undo the harm it suffered." Id. at 1128. We analogized the defendant's position

to that of an appellant who has not obtained a stay of execution on the underlying judgment pending appeal when the appellee executes on its judgment while the appeal is pending. The execution does not render the appeal moot since a reversal would allow the appellant to seek either a money judgment or return of the funds or property seized in the execution.

Id. at 1129. This situation is similar to the case before us.

Applying the effective relief test, we have little difficulty finding this appeal justiciable. If the district court lacked subject matter jurisdiction and its order were void ab initio, then as indicated in Raymark, Preferred Entity could seek a variety of relief both by attempting to recover damages for the seizure of the New Jersey property and by resisting the foreclosure action in New York. Given this potential for an effective remedy, the current appeal is not moot.

Because we find that the foreclosure sale has not mooted the appeal, we do not need to address New Rock's "partial mootness" theory. New Rock would have us moot the propriety of the district court's granting of summary judgment while still giving collateral estoppel effect to that judgment. We do not wish impliedly to give our stamp of approval to such a concept. We will therefore briefly point out the invalidity of the theory.

In making its argument on "partial mootness," New Rock points to language in National Iranian Oil which supports its position that a judgment having possible collateral legal consequences, including collateral estoppel effect in similar actions, is not moot. 983 F.2d at 490. The problem with New Rock's argument is that the example of collateral estoppel is only one of several examples that we gave in National Iranian Oil of effects, such as a viable claim for damages or a likelihood that the parties will relitigate the same issue, that will support a finding that a matter is not moot. If any one of these factors is established, the entire judgment is saved from mootness. It is not just a portion of the judgment which survives. If, however, the appeal is moot, it is entirely moot and it will have no res judicata effect. Id. at 489 (citing United States v. Munsingwear, Inc., 340 U.S. 36, 39-41, 71 S. Ct. 104, 106-07, 95 L. Ed. 36 (1950); Clarendon Ltd. v. Nu-West Indus., Inc., 936 F.2d 127, 130 (3d Cir.1991)).

The need that a judgment have a res judicata effect may be enough to support a determination that a judgment is not moot. We recognized in National Iranian Oil that such a need for preclusive effect may be a sufficient reason to reject mootness. 983 F.2d at 490. However, if we come to such a conclusion, i.e., that we will reject mootness so that a judgment has res judicata effect, then a fortiori the entire matter is not moot. New Rock cannot have a res judicata effect to assist it in the New York foreclosure actions and at the same time have a declaration that the granting of summary judgment by the district court, or the propriety of that judgment, is moot. A case is either moot or not. Because the potential for effective relief remains, this case is not moot.

We now turn to the central jurisdictional issues raised in this appeal.

Preferred Entity challenges the district court's granting of summary judgment based on the absence of federal jurisdiction. New Rock's sole asserted basis for jurisdiction is 12 U.S.C. § 1441a(l), which grants the federal courts original jurisdiction over any case to which the RTC is a party. Preferred Entity argues that because New Rock intervened in the suit and filed an amended complaint naming itself as sole plaintiff, the RTC is no longer a party to the action and § 1441a(l) cannot support jurisdiction. Although this question presents a difficult problem of statutory interpretation, we conclude that Preferred Entity is correct in its reading of § 1441a(l).

Preferred Entity's contention forces us to explore a gray area of FIRREA jurisdiction. It seems clear that jurisdiction under § 1441a(l) would not attach had the RTC sold the loans to New Rock before litigation or had New Rock filed its own action. In such a scenario, the RTC would never have been a party to the case, and the RTC cannot pass its jurisdictional rights by contract. It seems equally clear that federal jurisdiction would attach if the RTC had remained a party to the case, regardless of its capacity. Spring Garden Assoc., L.P. v. Resolution Trust Corp., 26 F.3d 412, 415-17 (3d Cir.1994) (finding jurisdiction where RTC substituted itself for defendant and removed entire action to federal court); Resolution Trust Corp. v. Nernberg, 3 F.3d 62, 66-67 (3d Cir.1993) (finding jurisdiction where RTC substituted itself for plaintiff and removed after state court judgment had been appealed). The issue is therefore whether § 1441a(l) provides for continuing jurisdiction where the RTC was once a party but has since been dropped from the case.

This question raises a matter of first impression for this circuit. Indeed, few courts have ever considered it. Of the federal appellate tribunals, only the United States Court of Appeals for the Fifth Circuit has addressed the issue. Federal Sav. & Loan Ins. Corp. v. Griffin, 935 F.2d 691 (5th Cir.1991), cert. denied, 502 U.S. 1092, 112 S. Ct. 1163, 117 L. Ed. 2d 410 (1992). Several district courts have reached it, including the Western District of Pennsylvania. Skaro v. Eastern Sav. Bank, 866 F. Supp. 229 (W.D.Pa.1994). Skaro relied almost exclusively on Griffin, as did the district court here. By contrast, in Mill Investments, Inc. v. Brooks Woolen Co., Inc., 797 F. Supp. 49 (D.Me.1992), the United States District Court for the District of Maine discussed Griffin thoroughly and reached contrary conclusions. After conducting our own independent analysis of the matter, we find that we disagree with Griffin and agree with Mill Investments.2

The scope of § 1441a(l) presents a question of statutory interpretation. This process begins with the plain language of the statute. Santa Fe Medical Services, Inc. v. Segal (In re Segal), 57 F.3d 342, 345 (3d Cir.1995); Spring Garden, 26 F.3d at 415; see United States Trustee v. Price Waterhouse, 19 F.3d 138, 141 (3d Cir.1994). "Where ... the statute's language is plain, 'the sole function of the court is to enforce it according to its terms.' " United States v. Ron Pair Enters., Inc., 489 U.S. 235, 241, 109 S. Ct. 1026, 1030, 103 L. Ed. 2d 290 (1989) (quoting Caminetti v. United States, 242 U.S. 470, 485, 37 S. Ct. 192, 194, 61 L. Ed. 442 (1917)). Plain meaning is conclusive, "except in the 'rare cases [in which] the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters.' " Id. at 242, 109 S.Ct. at 1031 (alteration in original) (quoting Griffin v. Oceanic Contractors, Inc., 458 U.S. 564, 571, 102 S. Ct. 3245, 3250, 73 L. Ed. 2d 973 (1982)).

In determining the plain meaning of FIRREA, we have consistently looked to its legislative history. Smith v. Fidelity Consumer Discount Co., 898 F.2d 907, 911 (3d Cir.1990); see Hudson United Bank v. Chase Manhattan Bank of Conn., N.A., 43 F.3d 843, 849 n. 14 (3d Cir.1994) ("As this is a matter of statutory construction, consideration of the legislative history would be appropriate."). We have also examined "the atmosphere in which [the statute] was enacted." Carteret Savings Bank, F.A. v. Office of Thrift Supervision, 963 F.2d 567, 574 (3d Cir.1992). Only if the plain language of the statute remains ambiguous after these steps will we "resort to other rules of statutory construction...." Resolution Trust Corp. v. Nernberg, 3 F.3d 62, 64 (3d Cir.1993) (finding ambiguity in removal provision of § 1441a(l)(3)(A) and (C)).3

These principles establish the requisite steps in our inquiry. We note at the outset that we reach our conclusion based on plain meaning. We find neither a "literal application of the statute [that] would produce a result demonstrably at odds with the intentions of its drafters," Ron Pair Enter., 489 U.S. at 242, 109 S. Ct. at 1031, nor an ambiguity that forces us to invoke general canons of statutory construction, Nernberg, 3 F.3d at 64.

We begin with the provision itself. Both the initial and amended complaints ground jurisdiction with a general cite to § 1441a(l). This section, entitled "Power to remove; jurisdiction," contains three parts. Section 1441a(l)(1) sets out a general jurisdictional grant creating original jurisdiction in the federal courts. Section 1441a(l)(2) provides for substitution of the RTC as a party for the RTC's predecessors in thrift supervision, the now-defunct Federal Saving and Loan Insurance Corporation ("FSLIC"), the equally defunct Federal Home Loan Bank Board ("FHLBB"), and the reconstituted and redirected Federal Deposit Insurance Corporation ("FDIC"). Section 1441a(l)(3) sets forth specific procedures for removing actions from state court where the RTC is a party. The vast majority of case law addresses this final provision. See, e.g., Wujick v. Dale & Dale, Inc., 43 F.3d 790, 792-94 (3d Cir.1994) (addressing jurisdictional sufficiency of removal by RTC; dismissing claim against RTC due to plaintiff's failure to exhaust remedies).

Despite New Rock's general citation to § 1441a(l), neither part (2) nor part (3) is relevant. The RTC began the case as a party, obviating the need for § 1441a(l)(2). The case was originally filed in federal court, eliminating the need for § 1441a(l)(3). This case turns on § 1441a(l)(1).

(1) In general

Notwithstanding any other provisions of law, any civil action, suit, or proceeding to which the [RTC] is a party shall be deemed to arise under the laws of the United States, and the United States district courts shall have original jurisdiction over such action, suit, or proceeding.

12 U.S.C. § 1441a(l)(1).

Preferred Entity contends that the plain language of § 1441a(l)(1) provides for jurisdiction only over cases where the RTC is a party but not where it was a party. In the current action, the RTC transferred its loans to New Rock, the court dismissed the RTC from the case, and New Rock filed an amended complaint naming itself as the sole plaintiff. Preferred Entity contends that the RTC consequently is no longer a party and jurisdiction no longer exists.

New Rock argues instead that the case is controlled by the principle that jurisdiction is determined at the time the action is commenced, both as a matter of statutory interpretation and, as discussed in Part III.B.2, a matter of black letter law. Under § 1441a(l)(1), this result is obtained by reading the statute to create federal jurisdiction which, once established at the outset by the presence of the RTC as a party, continues throughout the litigation, whether or not the RTC remains as a party.

We can divine no conclusive method of deciding between the two alternative readings of § 1441a(l)(1) from the text of the statute alone.4 Cf. Hudson United Bank, 43 F.3d at 849 ("It is true that FIRREA is awkwardly written and difficult to interpret."). Nevertheless, we find Preferred Entity's reading more persuasive. It encompasses the presence of the RTC as a party. Moreover, it reads the statute narrowly rather than expansively. See discussion, infra. Were we forced to rely solely on the words of the statute, we would agree with Preferred Entity. But other sources remain.

Following Smith v. Fidelity Consumer Discount Co. and Hudson United Bank, we next turn to legislative history to clarify plain meaning. Unfortunately, the legislative history of § 1441a(l)(1) provides minimal assistance. The current provision became law as 1989 Pub.L. No. 101-73 § 53(l )(1), 103 Stat. 389. The House Report's lone comment reads:

Subsection (o) [sic.] provides that suits by or against the RTC shall arise under the laws of the United States and can be removed to the District Court of the District of Columbia or if the suit arises out of actions by the RTC with respect to an institution for which a conservator or receiver has been appointed in the district court in which the institution's principal place of business is located.

H.R.Rep. No. 101-54(I), 101st Congress, 1st Sess. 362 (1989), reprinted in 1989 U.S.C.C.A.N. 86, 158. This passage does little more than reiterate the language of the bill and demonstrate that the provision was concerned first and foremost with removal. The substitution of the phrase "by or against the RTC" provides some support for the narrow reading proposed by Preferred Entity, but the same arguments for competing interpretations apply. Like the plain language of the statute, the legislative history favors Preferred Entity, but only slightly.

Stronger support for a narrow reading flows from the next source to which we have looked in interpreting FIRREA, "the atmosphere in which [the statute] was enacted." Carteret Savings, 963 F.2d at 574. Carteret describes that atmosphere.

In 1989, the thrift industry was in crisis. As the House Report noted, "[t]he nation's thrift industry and its deposit insurance fund, the [FSLIC] are currently in precarious financial condition and consumer confidence in the savings and loan industry is waning." H.R.Rep. No. 54, 101st Cong., 1st Sess., pt. 1, at 302 (1989), reprinted in 1989 U.S.C.C.A.N. 86, 98. The 2,949 FSLIC-insured savings institutions holding deposits of $971 billion and assets of $1.35 trillion lost $12.1 billion in 1988. Id. at 303, reprinted in 1989 U.S.C.C.A.N. at 99. The FSLIC was in a combined deficit position of at least $56 billion by the end of 1988. Id. at 304, reprinted in 1989 U.S.C.C.A.N. at 100. Rapidly declining consumer confidence led to record deposit withdrawals by consumers. Id. at 305, reprinted in 1989 U.S.C.C.A.N. at 101. Congress believed that the Bank Board, inter alia, had repeatedly understated the magnitude of the problem. Id.

Id. at 574-75.

In this atmosphere of crisis, Congress passed FIRREA to serve several important purposes. As framed by the statute, these purposes included: "(7) To establish a new corporation, to be known as the Resolution Trust Corporation, to contain, manage, and resolve failed savings associations." 1989 Pub.L. No. 101-73 § 101, 103 Stat. 187. The House Report stated this goal in more general terms: "The primary purposes of [FIRREA] are to ... establish organizations and procedures to obtain and administer the necessary funding to resolve failed thrift cases and to dispose of the assets of these institutions ... and, enhance the regulatory enforcement powers of the depositor institution regulatory agencies...." H.R.Rep. No. 101-54(I), 101st Congress, 1st Sess. 308 (1989), reprinted in 1989 U.S.C.C.A.N. 86, 104. To serve these purposes, FIRREA established the RTC "to manage and dispose of assets acquired from failed thrifts." Id.

Although not expressly discussed in the legislative history, § 1441a(l)(1)' § grant of federal jurisdiction relates to these purposes in obvious ways. It gives the RTC the benefit of a federal forum and a uniform body of federal law for its receivership activities. The federal forum is also a boon to the RTC both in pursuing claims and defending actions against the thrifts over which it had assumed control. The broader scope of a federal remedy similarly boosts the RTC's enforcement authority. Federal jurisdiction thus helps the RTC "manage and dispose of assets acquired from failed thrifts." Id.

The role of federal jurisdiction in assisting the RTC in its management role and in disposing of thrift assets also indicates that once this has been accomplished, the reasons for federal jurisdiction end. Once the RTC has successfully managed a thrift and either restored it to solvency or transferred its assets to willing buyers, the agency's role--and hence the logic of jurisdiction--no longer exists. This reading indicates why the terms of the statute limit federal jurisdiction to cases in which the RTC "is a party." The RTC will presumably only be a party where it is engaged in active management and disposal of thrifts and thrift assets. The RTC will no longer be a party--and jurisdiction will no longer apply--once the RTC has managed a thrift and its assets have been disposed.

The atmosphere surrounding FIRREA and the purposes of the statute thus provide additional support for Preferred Entity's reading of § 1441a(l)(1). Once New Rock became "the sole owner and holder of all right, title and interest in the Indebtedness," and once it "succeeded to all of RTC's right, title and interest in the Rent Order [obtained against defendants in a previous state court action]," the RTC's interest in the loans had been managed and disposed. The RTC no longer had any role in the action, and the agency was dropped from the case. Similarly, there was no longer any reason for federal jurisdiction, and § 1441a(l)(1)'s power lapsed. Given the purposes of FIRREA, Preferred Entity's jurisdictional stance is correct.

Based on the language of § 1441a(l)(1), its legislative history, and the background and purpose of FIRREA, we conclude that the plain meaning of the statute precludes continuing jurisdiction over an action where the RTC is no longer a party. Section 1441a(l)(1) will not support jurisdiction in this case.

In Federal Sav. & Loan Ins. Corp. v. Griffin, 935 F.2d 691 (5th Cir.1991), the Fifth Circuit reached the opposite conclusion.5 Griffin involved a similar scenario in which the FSLIC, acting as receiver for a failed thrift, removed a contract action to federal court. With the passage of FIRREA, the FDIC replaced the FSLIC. Jurisdiction was based on 12 U.S.C. § 1819, the FDIC's jurisdictional provision corresponding to the RTC's § 1441a(l). See Spring Garden, 26 F.3d at 416 n. 7 (noting parallels between statutes); Nernberg, 3 F.3d at 66 n. 2 (same). The FDIC assigned the contract action to a thrift that had acquired the assets and liabilities of the failed institution, after which it "apparently no longer pursu[ed] any claims." 935 F.2d at 694. This left the thrift and the private defendant as the only parties. Griffin nevertheless asserted jurisdiction.

Griffin based its conclusion primarily on policy concerns.6 Without citation to caselaw or legislative history, the court concluded that

policy reasons for insuring federal jurisdiction over cases involving the actions of failed thrifts continue when the FDIC is voluntarily dismissed as a party and the owner of the failed thrift's assets remains. A transferee from FSLIC or FDIC, as successor of their interests, is still entitled to the protection of federal courts applying D'Oench, Duhme, even when FSLIC or FDIC is voluntarily dismissed.

935 F.2d at 696.

In response to Griffin, the district court in Mill Investments, Inc. v. Brooks Woolen Co., Inc., 797 F. Supp. 49 (D.Me.1992), explored the policies behind FIRREA's jurisdictional grant and reached the same conclusions about its plain meaning that we have reached here:

FIRREA was enacted to deal with a banking crisis and "to smooth the modalities by which rehabilitation might be accomplished." Serge Marquis v. Federal Deposit Insurance Corp., 965 F.2d at 1154. It is clear to the Court that this policy is not advanced in any significant way by retaining federal jurisdiction once the failed bank's assets have been assigned to a private company. The expanded federal jurisdiction and other procedural protections of FIRREA may "tremendously increase the FDIC's ability to carry out its regulatory and enforcement responsibilities under FIRREA." Matter of Meyerland Co., 960 F.2d 512, 519 (5th Cir.1992). While the procedural protections also allow, "the FDIC to effectively and aggressively protect a failed bank's interests and assets," id. at 519-20, it can no longer do so when it is no longer a party to the case and when those assets have successfully been assigned to another. In essence, one of the goals of the statute has been achieved on a micro level once the assets have been assigned.

797 F.Supp. at 53-54. We agree. Contrary to Griffin 's naked assertion, the policy reasons for federal jurisdiction end when the FDIC or RTC leaves the case.

Mill Investments also dealt with Griffin 's claim that the need for federal enforcement of the D'Oench, Duhme doctrine supported jurisdiction. "D'Oench, Duhme is a federal estoppel doctrine which prohibits borrowers or guarantors from using secret or unrecorded side agreements to defend against efforts by FDIC or its assignees to collect on promissory notes it has acquired from a failed bank." Id. at 54 (citation omitted); see generally Adams v. Madison Realty & Dev., Inc., 937 F.2d 845, 852-54 (3d Cir.1991) (discussing D'Oench Duhme ). The Mill Investments court saw no reason why state courts could not enforce these defenses. 797 F. Supp. at 54. We also have faith in the competence of state tribunals. Griffin 's D'Oench, Duhme policy rationale is not convincing.

As a result, we find Griffin 's position on § 1441a(l)(1) unpersuasive. By contrast, our reading of the statute is consistent with the purpose of FIRREA as expressed in the statute and its legislative history. We also note that our reading is consistent with general policies underlying federal jurisdiction. These principles include the limited nature of federal jurisdiction and the goal of not interfering in the business of the states.

The limited nature of federal jurisdiction needs little discussion. This principle marks a fundamental precept of the American court system. Owen Equip. & Erection Co. v. Kroger, 437 U.S. 365, 374, 98 S. Ct. 2396, 2402-03, 57 L. Ed. 2d 274 (1978). Interpreting § 1441a(l)(1) narrowly comports with this general rule. See Lyon v. Whisman, 45 F.3d 758, 763-64 (3d Cir.1995) (interpreting narrowly jurisdiction under Fair Labor Standards Act); see also Shamrock Oil & Gas Corp. v. Sheets, 313 U.S. 100, 108-09, 61 S. Ct. 868, 872, 85 L. Ed. 1214 (1941) (interpreting jurisdictional statute narrowly).

A narrow reading of § 1441a(l)(1) also comports with the need to avoid interfering in state court matters. In BFP v. Resolution Trust Corp., 511 U.S. 531, 114 S. Ct. 1757, 128 L. Ed. 2d 556 (1994), the Supreme Court recognized this interest in considering the impact of a federal bankruptcy provision on state foreclosure law:

Federal statutes impinging on important state interests cannot be construed without regard to the implications of our dual system of government. When the Federal Government takes over local radiations in the vast network of our national economic enterprise and thereby radically readjusts the balance of state and national authority, those charged with the duty of legislating must be reasonably explicit. It is beyond question that an essential state interest is at issue [in property foreclosures].... To displace traditional State regulation in such a manner, the federal statutory purpose must be clear and manifest. Otherwise, the [statute] will be construed to adopt, rather than to displace, pre-existing state law.

Id. at ---- - ----, 114 S. Ct. at 1764-65 (citations and alterations omitted).

We have previously expressed similar concerns about § 1441a(l).

We note with some uneasiness that ... the Resolution Trust removal statute does not exclude [purely state law] cases. The language of the statute thus allows Resolution Trust to remove routine collection and foreclosure cases to the already overburdened federal courts.... It is a serious question whether such litigation is properly the[ir] role.

Nernberg, 3 F.3d at 68 n. 3. In this passage, we were commenting on post-judgment removal from state court where the RTC had been substituted as a party. In such a scenario, the RTC becomes an active participant in the case, injecting a federal element and creating a basis for removal. We nonetheless questioned the role of the federal courts in resolving such a dispute.

The concerns expressed in Nernberg are equally appropriate and even accentuated in the current context, where the RTC was once a party to the case but has now been dismissed, leaving a purely state law matter. Extending jurisdiction to federalize this class of foreclosure actions absent a "clear and manifest" legislative intent would conflict with the Supreme Court's ruling in BFP. 511 U.S. at ----, 114 S.Ct at 1765. Our examination of the plain meaning of § 1441a(l)(1) shows that no such "clear and manifest" intent exists. Our interpretation of the statute thus coheres with the federal goal of avoiding interference in state concerns.

For these reasons, we conclude that once the RTC left the case, § 1441a(l)(1) could no longer support federal jurisdiction. We will reverse the district court's exercise of jurisdiction pursuant to this provision.

Having determined that the language of § 1441a(l)(1) will not support continuing jurisdiction, we must next address New Rock's argument that the "black letter rule" that jurisdiction is determined at the time of filing preserves jurisdiction after the RTC's dismissal. We disagree.

The principle that jurisdiction is determined at the outset of the action is simply insufficient to support the continuing applicability of § 1441a(l)(1) to this case. One basic difficulty with this argument is that the letter and spirit of the rule apply most clearly to diversity cases. The Supreme Court set out the rule in the diversity context. St. Paul Mercury Indem. Co. v. Red Cab Co., 303 U.S. 283, 286, 290-92, 58 S. Ct. 586, 590, 591-92, 82 L. Ed. 845 (1938). In addition, the Court crafted the rule for the removal of actions from state court, which involves a more lenient standard not relevant here. Id. Most importantly, the policies behind removal and the risks of manipulative behavior played a significant role in the Court's decision. St. Paul focused primarily on the monetary threshold for federal jurisdiction, observing that the time of filing rule prevented plaintiffs from subsequently amending their complaint to plead a lesser amount and avoid removal. Id. at 294, 58 S.Ct. at 592-93. Similar concerns applied to changes of parties that would potentially destroy diversity of citizenship. Id. at 294-95, 58 S.Ct. at 592-93. From the outset, the underlying concern of the time of filing rule was the risk that parties would deploy procedural tactics to manipulate federal jurisdiction.

The rule that jurisdiction is assessed at the time of the filing of the complaint has been applied only rarely to federal question cases. Moreover, in these rare cases, the rule has often been applied axiomatically, without extensive discussion or analysis. See Rosa v. Resolution Trust Corp., 938 F.2d 383, 392 n. 12 (3d Cir.), cert. denied, 502 U.S. 981, 112 S. Ct. 582, 116 L. Ed. 2d 608 (1991); see also F. Alderete General Contractors, Inc. v. United States, 715 F.2d 1476, 1480 (Fed.Cir.1983) (observing in government contracts action that "the decision below is at variance with the long-standing rule in the Federal courts that jurisdiction is determined at the time the suit is filed and, after vesting, cannot be ousted by subsequent events, including action by the parties"). Even in the federal question context, however, the focus of the time of filing rule has been on preventing manipulation of jurisdiction when a claim is removed. As we observed in Westmoreland Hospital Ass'n v. Blue Cross of Western Pa., "a subsequent amendment to the complaint after removal designed to eliminate the federal claim will not defeat federal jurisdiction." 605 F.2d 119, 123 (3d Cir.1979) (emphasis added), cert. denied, 444 U.S. 1077, 100 S. Ct. 1025, 62 L. Ed. 2d 759 (1980). Along with the obvious goal of judicial efficiency, we perceive the risk of strategic behavior as the primary rationale behind the time of filing rule.

Manipulation of jurisdiction is simply not at issue in this case. There is no suggestion of manipulation, nor would the facts support it. The jurisdiction-destroying transfer of assets between the RTC and New Rock was an arms length transaction independent of the jurisdictional issue. Without the possibility of manipulative behavior, the primary policy behind the time of filing rule is not implicated.

Our rejection of an absolute time of filing requirement breaks no new ground. Courts that have considered the rule more fully have not hesitated to abandon it where appropriate. In Boelens v. Redman Homes, Inc., 759 F.2d 504 (5th Cir.1985), the Fifth Circuit discussed the policies behind the time of filing rule and held that in a federal question case, where the plaintiff's amended complaint omitted federal counts included in the original complaint on which jurisdiction could be based, the court would look to the amended complaint and decline jurisdiction. Id. at 508. The Fifth Circuit interpreted this rule as consistent with the general principle that the amended complaint "supersedes the original and renders it of no legal effect, unless the amended complaint specifically refers to or adopts the earlier pleading." Id. at 508.

We were equally quick to reject the time of filing rule in Lovell Mfg. v. Export-Import Bank, 843 F.2d 725 (3d Cir.1988):

Lovell ... cites several older Third Circuit cases for the proposition that our determination of jurisdiction should be based solely on the basis of the pleadings, and not on subsequent events.... We are uncertain that these cases stand for the broad proposition for which Lovell cites them. However, regardless of what they once might have stood for, and regardless of the merit of these principles elsewhere, plainly they do not reflect recent Third Circuit jurisprudence. As Lovell itself concedes, later cases clearly hold that once all federal claims have been dropped from a case, the case simply does not belong in federal court.

Id. at 734 (citations omitted). We concluded by observing "that to the extent a black-letter rule ever existed, precluding a court from relying on post-removal events ..., the Supreme Court clearly did not feel bound by it in Carnegie-Mellon Univ. v. Cohill, 484 U.S. 343, 108 S. Ct. 614, 98 L. Ed. 2d 720 (1988)." Id. at 735. Although the time of filing rule certainly retains a large measure of persuasive efficacy, we read Lovell as a clear rejection of any iron-clad time of filing requirement. Cf. Carr v. American Red Cross, 17 F.3d 671, 683-84 (3d Cir.1994) (federal jurisdiction arising from the involvement of the American Red Cross in a case will cease on the dismissal of the Red Cross from the case).

As a result, merely reciting the time of filing rule is not enough to support jurisdiction in this case. Although invoking this general principle has some surface appeal, the rule rests on policies that are not served by its application to these facts. There is also significant authority that supports our decision to diverge from it. New Rock's black letter maxim will not give the federal courts the power to hear this state law claim.

We have concluded that once the RTC was dismissed from the case, jurisdiction in the district court could no longer rest on § 1441a(l)(1). We now consider whether the district court had supplemental jurisdiction over the claims pursuant to 28 U.S.C. § 1367 after the RTC, the jurisdiction-conferring party, was dismissed and after the district court had invested considerable judicial resources and had resolved the case on its merits. We conclude that § 1367 provided supplemental jurisdiction under the circumstances of this case. Our holding means only that the district court had the discretion to retain jurisdiction after the RTC was dismissed; it does not suggest that the district court was obligated to resolve the case after the RTC dismissed, nor does it even suggest that district courts should retain jurisdiction in similar situations.

The question before us has two components. First, did Congress intend with 28 U.S.C. § 1367 to provide the federal courts with the discretion to exercise supplemental jurisdiction in the situation that we face here? Second, if this was Congress' intent, does this grant of jurisdiction exceed the scope of Article III of the United States Constitution? Because it best introduces the issues involved in this case, we begin with the second question.

The district court had jurisdiction over this action at the outset of the litigation pursuant to 12 U.S.C. § 1441a, which provides that any suit to which the RTC is a party "shall be deemed to arise under the laws of the United States." This jurisdictional grant did not expand the jurisdiction of the federal courts beyond that permissible under Article III. Brockman v. Merabank, 40 F.3d 1013, 1017 (9th Cir.1994); see also Osborn v. Bank of the United States, 22 U.S. (9 Wheat.) 738, 6 L. Ed. 204 (1824); American National Red Cross v. S.G., 505 U.S. 247, 264, 112 S. Ct. 2465, 2475-6, 120 L. Ed. 2d 201 (1992). The question before us now is whether Congress could extend this jurisdiction to include cases to which the RTC was party, but is no longer, without exceeding the bounds of Article III.

The Supreme Court delineated the modern constitutional bounds of pendent jurisdiction7 in United Mine Workers v. Gibbs, 383 U.S. 715, 86 S. Ct. 1130, 16 L. Ed. 2d 218 (1966). In that opinion, the Court considered when federal courts have jurisdiction over state claims which are related to federal claims between the same parties, but over which the federal courts have no independent basis for subject matter jurisdiction. The Court concluded that federal courts have the power to hear a state claim if the federal claim has "substance sufficient to confer subject matter jurisdiction on the court," and the state and federal claims "derive from a common nucleus of operative facts." 383 U.S. at 725, 86 S.Ct. at 1138. This question is ordinarily resolved on the pleadings. The decision to exercise this power, on the other hand, remains open, and should be based on considerations of "judicial economy, convenience and fairness to the litigants." 383 U.S. at 726-27, 86 S.Ct. at 1139. Once a court has decided to exercise jurisdiction over the state claim, however, elimination of the federal claim does not deprive the court of the constitutional power to adjudicate the pendent claim. Lentino v. Fringe Emp. Plans, Inc., 611 F.2d 474, 479 (3d Cir.1979).

Can this continuing jurisdiction over a state claim exist when, rather than the federal claim being eliminated, the federal claim itself becomes a state claim? The Supreme Court recently cited to Gibbs in a context analogous to the one with which we are faced. In Gutierrez de Martinez v. Lamagno, 515 U.S. 417, 115 S. Ct. 2227, 132 L. Ed. 2d 375 (1995) a bare majority of the Court held that under the Federal Employees Liability Reform and Tort Compensation Act (the "Westfall Act"), the federal courts could review certification by the Attorney General that an employee sued for a wrongful or negligent act was acting within the scope of his or her employment or office at the time of incident. If the Attorney General so certifies, the employee is dismissed from the action, the United States is substituted as the defendant, the case falls within the Federal Tort Claims Act. If the case is not already in federal court, it is then removed from state court by the Attorney General. 515 U.S. at ----, ----, 115 S. Ct. at 2229, 2235. Once in federal court the certification decision by the Attorney General is reviewable.

Article III comes into play in this situation because if the district court concludes that the employee acted outside the scope of employment and thus rejects the certification, the individual defendant must be resubstituted and the United States will no longer be a party. If the parties are not diverse, the federal court could be left "with a case without a federal question to support the court's subject matter jurisdiction." Id. at ----, 115 S.Ct. at 2236. Under the statute, the Attorney General's certification is conclusive for the purposes of removal, suggesting that even if the certification is rejected by the federal court and the United States is no longer a party, remand is not permissible. Although Lamagno itself did not raise this issue because the parties were diverse, the Court considered whether the statute should be interpreted to avoid this potential constitutional problem.

A four justice plurality cited to Gibbs in concluding that any Article III problem was not "a grave one."8 515 U.S. at ----, 115 S. Ct. at 2236. At the outset of the litigation, the Court reasoned, there was a significant federal question--whether the employee acted within the scope of his federal employment. Resolving this question, the Court concluded, involved consideration of the facts relevant to the underlying tort claims, thus " 'considerations of judicial economy, convenience and fairness' [citation to Gibbs omitted] make it reasonable and proper for the federal court to proceed beyond the federal question to final judgment once it has invested time and resources." Id. at ----, 115 S.Ct. at 2237. Therefore, even if the United States were replaced by a non-diverse party, and there was no other basis for federal jurisdiction, the federal court nonetheless retained jurisdiction over the case without running afoul of Article III.

Justice Souter dissented, joined by three other Justices, reasoning in part that "requiring" a federal court to keep jurisdiction over the case after the United States was no longer a party "must at least approach the limit [of 'arising under' jurisdiction under Article III], if it does not cross the line." Id. at ----, 115 S.Ct. at 2239. The certification question could not provide the basis for jurisdiction, according to the dissent, because that question itself was jurisdictional. In part to avoid testing the limits of Article III, the dissent read the statute as prohibiting review of the Attorney General's certification by the district court.

In our case there was federal jurisdiction at the outset of the litigation--the presence of the RTC--which does not implicate the concerns of the dissent in Lamagno. Jurisdiction springs not from the question of jurisdiction itself, but from the involvement of the RTC. Moreover, unlike the dissent's reading of the statute at issue in Lamagno, the supplemental jurisdiction statute at issue in this case does not "require" the district court to keep jurisdiction, it merely permits it to do so. The considerations of "judicial economy, convenience and fairness" cited by the Lamagno plurality are particularly compelling in this case where the district court has completely resolved the dispute on its merits. Requiring the parties to re-try the case in state court would needlessly duplicate the resources expended by the federal courts. We thus conclude that in such a case, where the jurisdiction-conferring party drops out and the federal court retains jurisdiction over what becomes a state law claim between non-diverse parties, the bounds of Article III have not been crossed.9 See Garcia v. U.S., 88 F.3d 318, 324 (5th Cir.1996) (agreeing with Lamagno plurality that even after the United States was dismissed as party, Article III did not prevent federal court from resolving case on the merits); see also Aliota v. Graham, 984 F.2d 1350 (3d Cir.1993), cert. denied, 510 U.S. 817, 114 S. Ct. 68, 126 L. Ed. 2d 37 (1993) (holding Attorney General's certification reviewable and that the case could not be remanded if the certification was invalidated and the United States was no longer a party).

An analogy to diversity jurisdiction supports our conclusion. In diversity cases a change in domicile that destroys the original basis for federal subject matter jurisdiction does not divest the federal courts of jurisdiction to continue to hear the case. Mollan v. Torrance, 22 U.S. (9 Wheat.) 537, 6 L. Ed. 154 (1824); Clarke v. Mathewson, 37 U.S. (12 Pet.) 164, 9 L. Ed. 1041 (1838).10 We have rejected an iron-clad "time of filing rule" that federal jurisdiction at the outset of a case is dispositive as to jurisdiction throughout the case. In rejecting the rule as dispositive, however, we do not suggest that jurisdiction at the outset of the case is irrelevant as to whether the district court continued to have jurisdiction after the RTC was dismissed. Indeed it is the initial jurisdiction over the case that provides the constitutional power for the court to continue to hear it.11

Finally, we note that a contrary conclusion would seriously limit the ability of Congress to provide a federal forum for litigation initiated by federally-created entities. For example, it would prevent Congress from deciding, after initially putting the case in federal court, that judicial economy required that that court have the discretion to keep the case. Congress would not even have the power to give continuing jurisdiction over the case for reasons related to the interests of the jurisdiction-conferring party. In this case, for example, the RTC may have more difficulty disposing of its assets if the purchaser knows that any litigation concerning those assets must be started anew in state court. Cf. Freeport-McMoRan, Inc. v. K.N. Energy, Inc., 498 U.S. 426, 428, 111 S. Ct. 858, 860, 112 L. Ed. 2d 951 (1991) (using this reasoning to support the rule that diversity jurisdiction, once established, is not defeated by the addition of a non-diverse party). We conclude that Article III does not require such a significant limitation on Congress' power to give jurisdiction to the federal courts of cases involving federally-created entities.

We now go to the second part of the inquiry. We have determined that Congress had the power to permit this case to continue to be heard in federal court. But is this what Congress intended under 28 U.S.C. § 1367?12 We turn first to the language of section 1367(a) which provides:

(a) Except as provided in subsections (b) and (c) or as expressly provided otherwise by Federal statute, in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under Article III of the United States Constitution. Such supplemental jurisdiction shall include claims that involve the joinder of additional parties.

28 U.S.C. § 1367. Under subsection (c),

the district courts may decline to exercise supplemental jurisdiction over a claim under subsection (a) if--

(1) the claim raises a novel or complex issue of state law,

(2) the claim substantially predominates over the claim or claims over which the district court has original jurisdiction,

(3) the district court has dismissed all claims over which it has original jurisdiction, or

(4) in exceptional circumstances, there are other compelling reasons for declining jurisdiction.

Id. (emphasis supplied).

Applying the statute to this case, we have pointed out that original jurisdiction pursuant to § 1441a(l)(1) existed when the RTC first filed this action. This satisfies § 1367(a). We have held that when the district court dismissed the RTC from the case, it no longer had jurisdiction under § 1441a(l)(1). This transition triggers a discretionary decision on whether jurisdiction over a state law claim should be declined pursuant to § 1367(c)(3). The plain language of the statute makes declination permissive, not mandatory.

We recognize that Congress may not have contemplated the precise issue raised by this case when it drafted the supplemental jurisdiction statute. Section 1367 appears to address multiple state and federal claims growing out of the same case and controversy or existing between the parties. Here, by contrast, we have a single claim that due to the accidents of circumstance has shifted mid-action from a claim arising under federal law to a claim existing under state law. We believe, however, that the supplemental jurisdiction statute, particularly through its underlying purpose of judicial efficiency, extends to this case. See Mill Investments Inc. v. Brooks Woolen Co. Inc., 797 F. Supp. 49, 51-53 (D.Me.1992) (considering § 1367); see also Kansas Pub. Employees Retirement Sys. v. Reimer & Koger Assoc. Inc., 61 F.3d 608, 611 (8th Cir.1995) (noting that after RTC removed case and was then dismissed, district court denied motion to remand to state court and asserted supplemental jurisdiction under § 1367). We do not suggest, of course, that the district court would have erred had it dismissed the case after the RTC was no longer a party. We decide only that this case comes within § 1367(a) and the district court therefore had the discretion to exercise supplemental jurisdiction after dismissal of the RTC.

Several factors support our conclusion. First, the statute specifically provides that the federal court may decline (and by implication, may chose to exercise) jurisdiction over supplemental claims even when the "district court has dismissed all claims over which it has original jurisdiction." (emphasis supplied). § 1367(c)(3). The situation which the statute specifically contemplates, in which the jurisdiction-conferring claims are "dismissed" and the court retains jurisdiction over the other claims, is difficult to distinguish from this case in which the jurisdiction-conferring party is dismissed. In both situations the jurisdiction-conferring element of the case drops out, and the federal court is left with a state law claim.

Moreover, this case is not one in which the dismissal of the RTC meant that the case was never properly in federal court. Congress put the case in federal court under § 1441a, and in doing so acted well within the scope of Article III. In applying § 1367 we have suggested a distinction between dismissal of a case for lack of subject matter jurisdiction, which means that § 1367(a) may not apply at all, and dismissal for failure to state a claim, which does not preclude application of § 1367(a). See Growth Horizons, Inc. v. Delaware Co., 983 F.2d 1277, 1284 (3d Cir.1993). But Growth Horizons did not address the issue here--whether § 1367 applies where the court had subject matter jurisdiction at the outset of the case, but after the dismissal of a jurisdiction-conferring party no longer does. In this case, we distinguish between jurisdictional defects that deprive the court of jurisdiction altogether, and those that arise only after the court has jurisdiction. Cf. State Farm Mutual Automobile Insurance Company v. Powell, 87 F.3d 93, 97 (3d Cir.1996) (making this distinction in diversity cases); IMFC Professional, Etc. v. Latin Am. Home Health, 676 F.2d 152, 159 n. 12 (5th Cir.1982) (making this distinction in removal cases).

Second, the language and legislative history of § 1367(a) support its extension to the limits that Article III permits. By its language § 1367(a) confers jurisdiction in mandatory terms to include those cases "which form part of the same case or controversy under Article III of the United States Constitution" (except as expressly excluded by statute or as provided for in subsections (b) and (c)). As one commentator explained:

the conferral is in very broad terms, and by using the "case or controversy" standard found in the Constitution's own statement of the outer limits of federal jurisdiction, Congress indicates that it wants supplemental jurisdiction, at least in the first instance--subject to its rejection as a matter of judicial discretion under subdivision (c)--to go the constitutional limit, where it appeared to be carried in the Gibbs case.

28 U.S.C. § 1367 (1993), David Siegal, Practice Commentary, "The 1990 Adoption of § 1367, Codifying 'Supplemental Jurisdiction' " at 831. We have consistently read § 1367(a) as codifying, (or in the area of pendant parties, expanding) the jurisdictional standard established in Gibbs. Borough of West Mifflin v. Lancaster, 45 F.3d 780, 788 (3d Cir.1995); Sinclair v. Soniform, Inc., 935 F.2d 599, 603 (3d Cir.1991); Lyon v. Whisman, 45 F.3d 758, 760 (3d Cir.1995).

The legislative history of § 1367 supports the conclusion that subsection (a) was intended to grant supplemental jurisdiction to the limits of Article III. See Arthur D. Wolf, Codification of Supplemental Jurisdiction: Anatomy of a Legislative Proposal, 14 W. New Eng. L.Rev. 1, 23 (1992) (concluding that under § 1367 "the test of the reach of subsection (a) will be the scope of Article III.") This history makes clear that the statute was passed in reaction to Finley v. United States, 490 U.S. 545, 109 S. Ct. 2003, 104 L. Ed. 2d 593 (1989), which held, in the context of pendent party jurisdiction, that the Court would not assume that Congress had intended the full constitutional power to hear the claim had been given to the federal courts, unless Congress passed specific legislation to that effect. H.R.Rep. No. 101-734, 101st Congress, 2d Sess., reprinted in 1990 U.S.C.C.A.N. 6802, 6874. The House Report states that the purpose of the legislation was to restore jurisdiction in cases like Finley and "essentially restore the pre-Finely [Finley] understandings of the authorization for and limits on other forms of supplemental jurisdiction." Id. The "pre-Finley " understanding of the authorization for supplemental jurisdiction was that "where Congress has not spoken to the contrary or where we cannot find Congressional intent to the contrary, jurisdictional statutes give federal courts the power to exercise ancillary and pendent jurisdiction to the constitutional limit." Ambromovage v. United Mine Workers, 726 F.2d 972, 991 n. 57 (3d Cir.1984). This was the approach taken in United Mine Workers v. Gibbs, 383 U.S. 715, 725, 86 S. Ct. 1130, 1138, 16 L. Ed. 2d 218 (1966): the Court looked to the constitutional limits on pendent jurisdiction and assumed that the federal courts' power extended to those limits, without looking for a specific grant of statutory authority. The language of § 1367(a) stating that it applies "except as expressly provided otherwise by federal statute" (emphasis supplied) and that it shall "include claims that involve the joinder or intervention of additional parties," confirms this reading of the statute.13

We express no opinion as to whether § 1367 should be read broadly in cases other than the one before us, but here it is particularly appropriate because the district court resolved this case on its merits and the risk of needless expenditures of judicial resources is accordingly a very real one. Lentino v. Fringe Emp. Plans, Inc., 611 F.2d 474, 479 (3d Cir.1979). The Supreme Court relied on similar reasoning when it rejected the argument that because an original constitutional claim was dismissed as moot, the district court should not have resolved the pendent claim:

We are not willing to defeat the common sense policy of pendent jurisdiction--the conservation of judicial energy and the avoidance of multiplicity of litigation--by a conceptual approach that would require jurisdiction over the primary claim at all stages as a prerequisite to the resolution of the pendent claim.

Rosado v. Wyman, 397 U.S. 397, 405, 90 S. Ct. 1207, 1214, 25 L. Ed. 2d 442 (1970).14

Accordingly, although we are mindful that jurisdictional statutes must be construed narrowly, Healy v. Ratta, 292 U.S. 263, 270, 54 S. Ct. 700, 703, 78 L. Ed. 1248 (1934), we are also mindful that it is our obligation to effectuate the intentions of Congress in interpreting those statutes. Here, those intentions are clear from the face of the statute and the legislative history, and they require that the statute be read broadly to retain jurisdiction in this case in which substantial judicial resources have produced a final decision on the merits.

Finally, § 1367 has been read by the Fourth Circuit to provide jurisdiction in an analogous situation arising in diversity. Shanaghan v. Cahill, 58 F.3d 106 (4th Cir.1995). In that case, after the court granted summary judgment on one of the plaintiff's claims, the amount in controversy fell below the $50,000 required for federal jurisdiction under 28 U.S.C. § 1332. The district court concluded that it no longer had subject matter jurisdiction. The Fourth Circuit disagreed, concluding that pursuant to § 1367 the court had supplemental jurisdiction over the remaining claims. Id. at 109. In that case, like this one, the jurisdictional basis for the claim dropped out; there was no claim to which the state claims were supplemental because jurisdiction was premised on the aggregate of amount in controversy involved in all the claims that were asserted at the outset. When one claim dropped out, this basis for federal jurisdiction no longer existed. The court reasoned, however, that the statute applies when the amount in controversy falls below $50,000 just as it does when a federal question drops out of the case:

... the same basic pattern of circumstances exists in both contexts: the jurisdictional basis of the action fades away and the court is left with what would otherwise be a state law case. There is no way to distinguish a reduction of the amount in controversy from the disappearance of a federal claim as contemplated under 1367(c). Indeed the factors applicable in the typical pendent jurisdiction case are equally applicable here--comity, the existence of a state limitations bar, and considerations of judicial economy. Whenever the basis for federal jurisdiction evaporates, Congress has provided for discretion.

Shanaghan, 58 F.3d at 110; see also Clarke v. Whitney, 934 F. Supp. 148, 151 n. 1 (E.D.Pa.1996) (following Shanaghan ).

Based on these considerations, we conclude that the district court had supplemental jurisdiction over this case after it dismissed the RTC. When the RTC was dismissed from the action, the district court should have considered supplemental jurisdiction as the statutory basis to decide issues which had been fully presented to the court. Given the district court's failure to consider § 1367, we would generally remand the case to the district court for further proceedings on the supplemental jurisdiction issue. See Growth Horizons, Inc. v. Delaware County, 983 F.2d 1277, 1284-85 (3d Cir.1993) (remanding for consideration of supplemental jurisdiction after holding that jurisdiction existed over primary federal claim). In light of the posture of the current dispute, however, we will dispense with remand, hold that supplemental jurisdiction exists, and affirm the district court's grant of summary judgment. See Sinclair v. Soniform, Inc., 935 F.2d 599, 604 (3d Cir.1991) (reversing district court's finding of no federal jurisdiction over maritime personal injury action; holding that supplemental jurisdiction existed over state law claims); Sparks v. Hershey, 661 F.2d 30, 34 (3d Cir.1981) (reversing district court's dismissal of state law claims as abuse of discretion and directing court to exercise jurisdiction over them).

In the instant case, the district court has already entered summary judgment for New Rock. That record is before us on appeal. We find no merit in Preferred Entity's challenges to the grant of summary judgment. Preferred Entity has never contested the validity of the mortgage, the existence of a default, or the acceleration of the indebtedness. Preferred Entity has never even contested the information in New Rock's affidavits or the specific amount due. Preferred Entity objects only to the legal sufficiency of the records to support a summary judgment motion, contending that New Rock's affidavits contain only inadmissible hearsay.

We have reviewed the contours of the personal knowledge requirement of Fed.R.Civ.P. 56(e) and the business records exception of Fed. R. Ev. 803(6). See, e.g., Colgan v. Fisher Scientific Co., 935 F.2d 1407, 1423 (3d Cir.) (discussing Rule 56(e)), cert. denied, 502 U.S. 941, 112 S. Ct. 379, 116 L. Ed. 2d 330 (1991); In re Japanese Elec. Products. Antitrust Litig., 723 F.2d 238, 288-89 (3d Cir.1983) (discussing Fed. R. Ev. 803(6)), rev'd on other grounds, 475 U.S. 574, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986). We believe the documents were properly considered. We find no genuine issue of material fact. Absent the jurisdictional hurdle, we would not hesitate to affirm the district court.

Given the fact that the district court has effectively resolved the case, we feel a rejection of supplemental jurisdiction would be inappropriate. See Growth Horizons, 983 F.2d at 1284-85 (emphasizing need to consider "judicial economy, convenience, and fairness to the litigants" in remanding to district court for consideration of supplemental jurisdiction over state law claim after trial on the merits); Sparks v. Hershey, 661 F.2d at 33 ("[W]e are inclined to believe that the dictates of judicial economy, convenience, fairness to the parties, and comity are better served by recognizing [supplemental] jurisdiction.") (citations omitted). Indeed, we might reverse any such rejection as an abuse of discretion. See Development Fin. Corp. v. Alpha Housing & Health Care, Inc., 54 F.3d 156, 161 (3d Cir.1995); but see Lyon v. Whisman, 45 F.3d 758, 760 (3d Cir.1995) (declining supplemental jurisdiction and dismissing case after trial on merits). We also recognize that New Rock has made substantial expenditures of time and effort in foreclosing on the property. We see no need to remand this case for a pro forma exercise of supplemental jurisdiction and a renewed appeal on the summary judgment order. We will ground jurisdiction on 28 U.S.C. § 1367 and affirm the district court. See Sinclair, 935 F.2d at 602-03; Sparks v. Hershey, 661 F.2d at 34.

For these reasons, the portion of the district court's order exercising jurisdiction pursuant to 12 U.S.C. § 1441a(l)(1) will be reversed. The district court's entry of summary judgment will be affirmed.

McKEE, Circuit Judge, dissenting.

I agree that the foreclosure sale has not mooted this appeal, and I therefore join in Part III.A of the majority opinion. However, even assuming arguendo that the majority correctly concludes that the RTC's dismissal defeated federal jurisdiction, I cannot agree that we can uphold the exercise of supplemental jurisdiction over this state law foreclosure action. Although Congress may well be able to provide for a federal court to retain jurisdiction over state law claims once the RTC was dismissed here, that is not the issue. The majority states that "[t]he question before us now is whether Congress could extend this jurisdiction to include cases to which the RTC was a party, but is no longer, without exceeding the bounds of Article III." Maj. Op. at 1505. Having gone in at the wrong place, it is perhaps not surprising that the majority has come out at the wrong place. The question posed is not the issue. Rather, the question is whether Congress did so extend this jurisdiction. Our inquiry, though simply stated, is not so simply resolved. We need to determine if Congress intended to so extend our jurisdiction when it enacted 28 U.S.C. § 1367. For the reasons that follow, I conclude that it did not, and I therefore must respectfully dissent from the majority opinion.

In relevant part, 28 U.S.C. § 1367 provides as follows:

(a) Except as provided in subsections (b) and (c) or as expressly provided otherwise by Federal statute, in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under ... the Constitution.

28 U.S.C § 1367(a) (emphasis added). Under the statute as drafted a court can exercise supplemental jurisdiction only if it has original jurisdiction. If it does not have jurisdiction over the underlying claim there is nothing to supplement. Here, the majority first holds that the "Time of Filing" rule does not apply and concludes that the district court lost subject matter jurisdiction once the RTC was dismissed. However, it then proceeds to declare that the court nevertheless has original jurisdiction, and can therefore exercise supplemental jurisdiction because jurisdiction originally attached when the RTC was a party. It can not be both ways. If the "Time of Filing" rule does apply, the district court had jurisdiction to enter summary judgment. That jurisdiction, however, must arise not from the supplemental jurisdiction conferred in 28 U.S.C. § 1367, but from the substantive grant of jurisdiction contained in the Financial Institutions Reform, Recovery, and Enforcement Act of 1989, 12 U.S.C. § 1441a(1)(1) ("FIRREA"). If the "Time of Filing" rule does not apply, there is no federal jurisdiction for any jurisdiction under 28 U.S.C. § 1367 to supplement.

In holding to the contrary, the majority relies heavily upon Gutierrez de Martinez v. Lamagno, 515 U.S. 417, 115 S. Ct. 2227, 132 L. Ed. 2d 375 (1995), however that case did not involve supplemental jurisdiction under section 1367, and is therefore inapposite to the inquiry here. The issue in Lamagno was the scope of the jurisdictional grant that Congress intended under the Westfall Act. Under that Act if a federal employee is sued, the Attorney General certifies whether or not the employee was acting within the scope of his or her federal employment at the time of the injury that forms the basis of the plaintiff's claim. If the Attorney General certifies that the individual was within the scope of his or her employment, the United States is substituted as a party, and the action is removed to federal court. If the Attorney General certifies to the contrary, the United States is not substituted as a party, and the action remains in state court, absent some other basis for federal jurisdiction. In Lamagno, an issue arose as to whether a federal court can review the certification of the Attorney General. Although the Court did discuss the parameters of the judicial power contained in Article III, the decision had nothing to do with supplemental jurisdiction. Indeed, the statute was never cited. The same is true of Garcia v. U.S., 88 F.3d 318 (5th Cir.1996) which the majority also relies upon. That case was also decided under the Westfall Act, however, it had the further dimension of determining whether a federal court lacked subject matter jurisdiction over the underlying dispute if a court reversed the Attorney General's certification and concluded that the United States could not be substituted as a party. The court stated

[w]e agree with the Lamagno plurality that, because of the Attorney General's certification, there is an initial colorable federal question. Accordingly, we agree likewise that there is no 'grave' Article III problem in a district court retaining jurisdiction after rejecting the Attorney General's certification.

Garcia, 88 F.3d at 325. However, that conclusion was based upon a very technical parsing of the precise language of the statute. Thus, resolution of jurisdictional issues under the statutory framework of the Westfall Act does not advance our inquiry under 28 U.S.C. § 1367 nearly so far as the majority concludes. The same is true of nearly all of the authority that the majority relies upon for its holding as to supplemental jurisdiction.

Surprisingly, the majority cites Mollan v. Torrance, 22 U.S. (9 Wheat.) 537, 6 L. Ed. 154 (1824) and Clarke v. Mathewson, 37 U.S. (12 Pet.) 164, 9 L. Ed. 1041 (1838) to suggest that section 1367 provides supplemental jurisdiction because the court had jurisdiction at the outset of the litigation. See Maj. Op. at 1507. Those cases strongly suggest that the majority is in error in not relying upon the "Time of Filing" rule in determining if jurisdiction was lost when the RTC was dismissed. In Mollan, the Supreme Court stated "[i]t is quite clear, that the jurisdiction of the Court depends upon the state of things at the time of the action brought, and that after vesting, it cannot be ousted by subsequent events." 22 U.S. at 539. Similarly, in Clarke, the Court noted

It has not been the course of the courts of the United States to consider their jurisdiction, after it has once attached, as taken away by the subsequent change of residence of the party. A suit properly commenced between citizens of different states still proceeds; although the parties may, before its termination, become citizens of the same state.

37 U.S. at 167. In St. Paul Mercury Indemnity Co. v. Red Cab, 303 U.S. 283, 58 S. Ct. 586, 82 L. Ed. 845 (1938) the Supreme Court reiterated the "Time of Filing" rule as first pronounced in Mollan. In Red Cab, the Court held that jurisdiction must be established from the good faith averments on the face of the complaint, and once established it is not defeated by subsequent events. "Events occurring subsequent to the institution of suit which reduce the amount recoverable below the statutory limit do not oust jurisdiction." 303 U.S. at 289-90, 58 S.Ct. at 590-91.1 This precedent does not support the majority's holding under section 1367, and it undermines the majority's reasoning for rejecting the "Time of Filing" rule. Nevertheless, assuming arguendo that this precedent can properly be limited to diversity jurisdiction as the majority suggests, the majority's reasoning still leaves a court free to exercise supplemental jurisdiction when it has no jurisdiction to supplement.

More recently in Lujan v. Defenders of Wildlife, 504 U.S. 555, 112 S. Ct. 2130, 119 L. Ed. 2d 351 (1992) (See Maj. op. at 1507, n. 10) the Supreme Court again upheld the "Time of Filing" rule in rejecting the idea that the standing of parties who were added after litigation was begun could create a sufficient case or controversy for Art III jurisdiction even if the original plaintiffs lacked standing.

The existence of federal jurisdiction ordinarily depends on the facts as they exist when the complaint is filed.... It cannot be that, by later participating in the suit, the State Department and AID retroactively created the redressability (and hence jurisdiction) that did not exist at the outset ... There is no support for the dissent's novel contention, ... that Rule 19 of the Federal Rules of Civil Procedure, ... somehow alters our longstanding rule that jurisdiction is to be assessed under the facts existing when the complaint is filed.

504 U.S. at 569, 112 S. Ct. at 2141.

However, if the "Time of Filing" rule does not apply here, it is nevertheless clear that 28 U.S.C. § 1367 does not intend that a federal court exercise jurisdiction over a state law claim unless the federal claim that the state claim supplements is properly before the court. I do not think that the Congress intended to allow the exercise of federal jurisdiction to resolve a uniquely state claim where, as here, the federal court concludes that it has no original federal jurisdiction. We are, after all, interpreting a jurisdictional statute.

The policy of the statute calls for its strict construction.... Due regard for the rightful independence of state governments, which should actuate federal courts requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.

Healy v. Ratta, 292 U.S. 263, 270, 54 S. Ct. 700, 703, 78 L. Ed. 1248 (1934).

This is not a case where there are federal claims and state claims that "derive from a common nucleus of operative facts ... such that [a plaintiff] would ordinarily be expected to try them all in one judicial proceeding." United Mine Workers of America v. Gibbs, 383 U.S. 715, 725, 86 S. Ct. 1130, 1138, 16 L. Ed. 2d 218 (1966). Gibbs is the source of the "modern doctrine of pendent jurisdiction", Carnegie-Mellon University v. Cohill, 484 U.S. 343, 348, 108 S. Ct. 614, 618, 98 L. Ed. 2d 720 (1988), and it teaches that, under the appropriate circumstances, a federal district court possesses the discretionary power to decide claims based on state law. Gibbs, 383 U.S. at 725-26, 86 S.Ct. at 1138-39. As the majority notes, the doctrine of pendent jurisdiction is now codified in the supplemental jurisdiction statute. 28 U.S.C. § 1367. However, as noted above, that jurisdiction only exists in civil actions "of which the district courts have original jurisdiction." Thus, the only way the majority can justify its contrary result is by relying upon the very "Time of Filing" rule that it rejects in the first instance.

Nor is this a question of abuse of discretion where a court properly has discretion to exercise its supplemental jurisdiction. We are confronted with an error of law arising from what I believe is an erroneous application of a legal principle occasioned by an incorrect reading of the supplemental jurisdiction statute. I think it clear that the district court's action can not survive the plenary review we must give it. Accordingly, the majority's emphasis on the fact that the district court has already invested a great deal of time and resource in adjudicating this matter is irrelevant. It is axiomatic that a federal court cannot give itself jurisdiction where none exists. Although the resources that the district court has invested in resolving this case would most certainly be relevant to an exercise of the court's discretion to exercise supplemental jurisdiction of a case properly before it, it is irrelevant to the question of whether the court had jurisdiction over the state law claims to begin with. The original subject matter jurisdiction that is the condition precedent for the exercise of supplemental jurisdiction can not be conferred by considerations of convenience and efficiency.

The majority's interpretation of 28 U.S.C. § 1367 is even more perplexing given the recognition of the unique state interests that are implicated in a foreclosure action, and the limited scope of federal jurisdiction.

our reading of the statute is consistent with the purpose of FIRREA as expressed in the statute and its legislative history. We also note that our reading is consistent with general policies underlying federal jurisdiction. These principles include the limited nature of federal jurisdiction and the goal of not interfering in the business of the states.

The limited nature of federal jurisdiction needs little discussion. This principle marks a fundamental precept of the American court system. Owen Equip. & Erection Co. v. Kroger, 437 U.S. 365, 374, 98 S. Ct. 2396, 2402-03, 57 L. Ed. 2d 274 (1978). Interpreting [FIRREA] narrowly comports with this general rule.

Maj. op. at 1502. Here again, the majority attempts to have its analytical pie and eat it too. However, I fail to see how the majority can justify giving a narrow interpretation to the jurisdictional inquiry under FIRREA (and thereby departing from Mollan and its progeny) while giving such an expansive interpretation to section 1367.

Here, the RTC had no interest in the New Jersey real estate that was the subject of the mortgage lien, nor was the RTC a promisee under any instruments that created the disputed indebtedness. The RTC is merely a successor in interest to some of the parties to the original lawsuit. That suit was actually initiated in state court by private entities that had an interest in the litigation. The RTC then removed it to state court, but is no longer a party. Now, through dubious reasoning, we allow the federal court to which this state action was removed to adjudicate purely state claims that are unconnected to the RTC or any other federal agency.

3. Plaintiff, New Rock, is the sole owner and holder of all right, title and interest in the Indebtedness ... pursuant to the Mortgage Assignments ...

31. New Rock has succeeded to all of RTC's right title and interest....

See Maj. op. at 1495. If that is so, I do not understand how 28 U.S.C. § 1367 can be interpreted to allow a federal court to adjudicate New Rock's dispute.2 Indeed, the analytical sleight of hand that allows the majority to rely upon the very principles it rejects has the end result of giving us the "jurisdiction by contract" it tried to avoid in refusing to apply the "Time of Filing" rule in the first instance. Id. at 1498. ("[T]he RTC cannot pass its jurisdictional rights by contract.")

Accordingly, I cannot join in the majority opinion.