No. 94-7117.United States Court of Appeals, District of Columbia Circuit.Argued May 9, 1995.
Decided August 22, 1995.
Mark F. Rosenberg argued the cause for appellant. With him on the briefs were Richard H. Sauer an Bernard A. Joseph.
John R. Fornaciari appeared pro hac vice
and argued the cause for appellee. John J. Vecchione was on the brief for appellee.
Appeal from the United States District Court for the District of Columbia.
Before: Williams, Ginsburg, and Randolph, Circuit Judges.
Opinion for the Court filed by Circuit Judge
GINSBURG, Circuit Judge:
 A.I. Trade Finance, Inc. (AITF) appeals the district court’s grant of summary judgment in favor of Petra International Banking Corporation (PIBC), an Edge Act corporation. AITF brought suit in order to hold PIBC liable on six notes guaranteed by Petra Bank of Jordan (Petra Bank), upon the theory that PIBC is the alter-ego of Petra Bank. For the reasons set out below, we affirm the judgment of the district court.
 I. Background
 The facts as alleged in AITF’s complaint (or as otherwise reflected in the record and not disputed) are as follows. On January 15, 1989 Nissilios Shipping of Piraeus, Greece executed six negotiable instruments with a face value of $15 million in favor of Welfin, S.A., a Swiss investment bank, ostensibly in order to finance the purchase of some electronic equipment to be used in the construction of the ship M.V. Nissilios.
Each note is guaranteed by Petra Bank with the notation “per aval” and promises a payment of $2.5 million dollars; two notes were due on October 17, 1989, and four were due on January 17, 1990.
 This transaction is an example of “forfaiting,” an increasingly common method of trade financing in which the exporter receives in payment a negotiable obligation guaranteed by the importer’s bank with terms that give it a present value equal to the purchase price of the goods sold. The forfaiter, in this case Welfin, may either hold the note to maturity or sell it on the secondary market. Forfaiters are more willing to deal in such obligations because they are guaranteed by a bank “per aval” which, unlike an ordinary guaranty that is triggered only upon the default of the original maker, renders the guarantor bank liable directly upon the instrument. In exchange for the discount on the obligation, the forfaiter assumes only interest rate risk and the credit risk associated with the guarantor bank. In addition, these instruments are relatively liquid because they can be sold on the secondary market “without recourse” to the seller. See generally Elnora Uzzelle, Forfaiting Should Not Be Overlooked As An Innovative Means of Export Finance, Business America, Feb. 1995 at 20.
 To continue, the six notes guaranteed by Petra Bank were endorsed by Welfin, without recourse, to AITF, a secondary forfaiter, for just over $13.5 million. Shortly thereafter, AITF sold three of the notes, also “without recourse,” to another secondary forfaiter, Centro Internationale Handelsbank, A.G. for a bit under $6.75 million. Centro, in turn, sold the three notes to ABN Amro, a Dutch bank, upon similar terms. The secondary forfaiters left holding the notes, AITF and Amro, had thus each purchased debt with a face value of $7.5 million at a discount that presumably reflected prevailing interest rates and their confidence in Petra Bank as guarantor of the notes.
 Soon after the transactions described above Petra Bank began to suffer large losses, apparently due to misconduct on the part of some of its officers. The Jordanian government eventually put the bank into receivership. Prior to that, however, in August 1989 the failing Petra Bank declared a moratorium upon the payment of all guaranties. By January 1990 it had refused payment of the six notes involved in this case, which set off a chain reaction of lawsuits. First, AITF sued Centro in the federal court in New York for a declaratory judgment that AITF was not liable on the three notes that it had sold to Centro; inevitably Centro counterclaimed, alleging that AITF had committed various wrongs in connection with that sale. See A.I. Trade Finance, Inc. v. Centro Internationale Handelsbank, A.G., Dkt. No. 89 Civ. 7664 (PNL) (S.D.N.Y. filed Nov. 16, 1989). Meanwhile, Amro sued Centro in Vienna, and AITF sued Petra Bank, again in New York, for failing to honor its guaranty on the three notes that AITF still held. See A.I. Trade Finance, Inc. v. Petra Bank, 89 Civ. 7987 (JFK) (S.D.N.Y. filed Nov. 30, 1989).
 AITF’s litigation with Centro soon bogged down in discovery disputes, and its litigation against Petra Bank was in jeopardy of being dismissed in favor of the bankruptcy proceedings in Jordan. So in August 1993 AITF filed this suit against PIBC in the United States District Court for the District of Columbia.
 PIBC is incorporated under the Edge Act, a federal law that authorizes the chartering of a corporation “for the purpose of engaging in international or foreign banking or other international or foreign financial operations.” 12 U.S.C. § 611. PIBC’s only office is located in the District of Columbia. With the approval of the Federal Reserve Board, Petra Bank owns approximately 70 percent of PIBC’s stock. See 12 U.S.C. § 619; 12 C.F.R. § 211.4(b)(2) (requiring majority of shares in Edge Act corporation be owned by
U.S. citizens or firms unless Board approves foreign ownership).
 AITF alleges that PIBC is the “mere agent, instrumentality and alter ego of Petra [Bank],” and that PIBC is therefore liable on the $7.5 million worth of notes that Petra Bank guaranteed and AITF still holds. AITF also seeks a declaration that PIBC is likewise liable for any losses that AITF may incur in connection with the $7.5 million worth of notes that AITF sold to Centro, which are the subject of litigation in both New York and Vienna.
 PIBC moved for dismissal or, in the alternative, summary judgment upon the grounds that AITF’s claim is barred by the District of Columbia’s three-year statute of limitations for contract actions and that, in any event, PIBC is neither the alter-ego of Petra Bank nor otherwise responsible for Petra Bank’s guaranties. AITF opposed, arguing that: (1) the statute of limitations has not even begun to run on its claim involving the Centro notes because AITF has not yet been held liable to Centro in New York; (2) under federal choice-of-law rules New York’s six-year statute of limitations applies to AITF’s claim based upon the notes that it holds; (3) even if a D.C. statute of limitations applies, it was tolled in 1989 when AITF filed suit against Petra Bank in New York; (4) if there was no tolling, the action is still timely because the applicable limitation period under D.C. law is 12 years for contracts under seal; and (5) PIBC’s assertions concerning the merits of the case are inadequate and, because AITF has not yet been able to conduct any discovery, premature.
 The district court granted PIBC’s motion for summary judgment. The court held first that its jurisdiction rested both upon the parties’ diversity of citizenship and upon the specific grant of jurisdiction over suits involving foreign banking transactions of Edge Act corporations in 28 U.S.C. § 632. The court reasoned that because its subject-matter jurisdiction was based in part upon diversity of citizenship, it would look to the law of the forum to determine the applicable statute of limitations; the court also noted, however, that it would reach the same result if it were to decide that issue under federal law because the District of Columbia has a superior interest in a suit seeking to “pierce the veil of a District of Columbia corporation.” The court then held that the District’s three-year limitation upon contract actions applies and, because the suit was filed some four-and-one-half years after Petra Bank had dishonored its guaranties, entered judgment for the defendant.
 Since the district court’s decision, Amro has lost its bid in the Austrian court of first instance to hold Centro liable on the three notes that Centro sold to Amro, which prompted AITF to seek dismissal of its claims and of Centro’s counterclaims in their New York litigation upon the ground that there is no longer a case or controversy over which the court may exercise jurisdiction. Centro has opposed that motion upon the grounds that not all of its claims against AITF are rendered moot by the decision of the Vienna court and that the case is as live as ever in light of Amro’s pending appeal. AITF’s suit against Petra Bank in New York also remains pending; the bankruptcy court (to which the district court referred the case) recently denied Petra Bank’s motion to dismiss in favor of the proceedings in Jordan, which it considered inadequate to protect AITF’s interests. Petra Bank is currently appealing that decision to the district court.
 On appeal here, AITF presses four statute of limitations arguments. First, however, we take up its argument that federal choice-of-law rules dictate that we apply the appropriate New York statute of limitations to its cause of action against PIBC. Only after we have determined what law applies will we address AITF’s statute of limitations arguments, starting with its contention that the statute of limitations has not even begun to run on its claim against PIBC based upon the three notes that AITF sold to Centro.
 II. Choice of Law
 AITF argues that this is a “federal question” case, and that we should therefore apply federal choice-of-law principles in order to determine which state’s law supplies the applicable statute of limitations. AITF, of course, argues for New York’s six-year period for contract actions. See N.Y. Civ. Prac.
L. R. § 213. PIBC counters that this case presents no reason to depart from the general rule that a federal court applies the law (including the choice-of-law rules) of the state in which it sits, and that the district court properly applied D.C.’s three-year limitation upon contract actions. See D.C. Code §12-301(7). The resolution of the choice-of-law issue turns upon the precise nature of the district court’s jurisdiction over this case.
 A. Diversity Jurisdiction
 The district court having first determined that it was sitting in diversity then applied the choice-of-law rules of the forum, viz., the District of Columbia. We would surely affirm that result if indeed the sole basis of the district court’s jurisdiction in this case were diversity of citizenship.
 A federal court sitting in diversity must apply state law to the substantive issues before it. Erie Railroad Co. v. Tompkins, 304 U.S. 64 (1938); Rules of Decision Act, 28 U.S.C. § 1652. For this purpose, the statute of limitations is substantive; therefore, a federal court sitting in diversity looks to state law to determine whether a cause of action based upon state law has expired. Guaranty Trust Co. v. York, 326 U.S. 99 (1945). In determining which state’s limitation period applies, the federal court looks to the choice-of-law rules of the state in which it sits. Klaxon Co. v. Stentor Electric Manufacturing Co., Inc., 313 U.S. 487 (1941). (Although the District of Columbia is not a state, the principle unquestionably applies. See Lee v. Flintkote Co., 593 F.2d 1275, 1278-80 (D.C. Cir. 1979).) Looking to the D.C. choice-of-law rules, we see that they treat statutes of limitations as procedural, and therefore almost always mandate application of the District’s own statute of limitations Namerdy v. Generalcar, 217 A.2d 109, 113 (D.C.App. 1966); cf. Sun Oil Co. v. Wortman, 486 U.S. 717, 726 (1988) (“Guaranty Trust itself rejects the notion that there is an equivalence between what is substantive under the Erie doctrine and what is substantive for purposes of conflict of laws”).
 There is some question, however, whether the parties to this case are indeed diverse. While a state-chartered corporation is a citizen of the chartering state for the purpose of diversity jurisdiction, St. Louis San Francisco Railway Co. v. James, 161 U.S. 545, 562 (1896), the Supreme Court has also held that a corporation chartered pursuant to an Act of Congress is not a citizen of any state for this purpose Bankers Trust Co. v. Texas Pacific Railway Co., 241 U.S. 295, 309-10 (1916). In 1958, however, the Congress specifically sought to restrict the diversity jurisdiction by providing that a corporation shall be “deemed to be a citizen of any State by which it has been incorporated and of the State where it has its principal place of business.”28 U.S.C. § 1332(c)(1), thus combating “the evil whereby a local institution, engaged in a local business and in many cases locally owned, is enabled to bring its litigation into the Federal courts simply because it has obtained a corporate charter from another State.” H.R. Rep. No. 1706, 85th Cong., 2d Sess. 4 (1958).
 Whether Bankers Trust remains good law in light of the 1958 amendment has not been determined by any federal court of appeals, though a number of district courts have stuck by the rule excluding federally chartered corporations altogether from diversity jurisdiction. See, e.g., Rice v. Disabled American Veterans, 295 F. Supp. 131, 133 (D.D.C. 1968). According to those courts, it is a corollary of the Bankers Trust rule that a federally chartered corporation the activities of which are limited to a single state is to be deemed a citizen of that state, see Bankers Trust, 241 U.S. at 309 (noting corporation’s activities “were not to be confined to a single State, but to be carried on, as in fact they are, in different States”), though the courts have disagreed over whether such a limitation depends upon the corporation’s charter limiting its activities to a single state or upon the actual extent of the corporation’s activities. Compare, e.g., Rice, 295 F. Supp. at 133, and Monsanto Co. v. TVA, 448 F. Supp. 648, 650 (N.D. Ala. 1978), with Little League Baseball, Inc. v. Welsh Publishing
Group, Inc., 874 F. Supp. 648, 651 (M.D. Pa. 1995).
 The parties have not addressed this issue, and we are therefore reluctant to rule upon it. Nor need we decide whether the district court properly exercised diversity jurisdiction over this case, for even if it did we would still have to inquire whether there is any alternative head of jurisdiction that overrides the rules that apply when the sole basis of a district court’s jurisdiction is the parties’ diversity of citizenship. We therefore turn to the district court’s alternative assertion of jurisdiction under 12 U.S.C. § 632.
 B. Federal Question Jurisdiction
 The only possible alternative source of jurisdiction of which we are aware is 12 U.S.C. § 632, upon which the district court relied in part. Section 632 provides that:
Notwithstanding any other provision of law, all suits of a civil nature at common law or in equity to which any corporation organized under the laws of the United States shall be a party, arising out of transactions involving international or foreign banking . . . shall be deemed to arise under the laws of the United States, and the district courts of the United States shall have original jurisdiction of all such suits. . . .
 AITF claims that by “deeming” its cause of action against PIBC to arise “under the laws of the United States,” § 632 makes this dispute an issue of federal law within the “federal question” jurisdiction of 28 U.S.C. § 1331. That assertion does not capture the precise nature of the federal courts’ jurisdiction over this case, however.
 1. Jurisdiction under 28 U.S.C. § 1331
 In a “federal question” case within the scope of § 1331, there is by definition some substantive federal law to govern the case from the outset. See Louisville Nashville Railway Co. v. Mottley, 211 U.S. 149 (1908) (question of federal law must appear on face of well-pleaded complaint). There is nothing in AITF’s complaint, however, that appears to raise an issue of federal law. AITF nevertheless argues that there is a federal question here because § 632 incorporates and makes a matter of federal law the substantive commercial law principles governing the disputed issues in this case. Under AITF’s theory, the effect of a bank’s guaranty on a negotiable instrument, for example, would be decided as a matter of federal law, despite the absence of any federal legislation on the subject.
 If we were to accept AITF’s argument, it would necessarily follow that the federal courts would develop a federal common law that applies to all the international banking transactions of federally chartered institutions similar, we presume, to the federal common law developed under cases such as Clearfield Trust Co. v. United States, 318 U.S. 363, 366 (1943) (holding federal common law governs “rights and duties of the United States on commercial paper which it issues”), an United States v. Kimball Foods, Inc., 440 U.S. 715, 726 (1979) (holding “the priority of liens stemming from federal lending programs must be determined with reference to federal law”). In fashioning this federal common law, AITF urges us to look to the law of New York, including that state’s statute of limitations for contract actions, on the ground that New York has the greatest interest in the present controversy; regardless of the merit of this last contention, cf. Lampf, Pleva, Lipkind, Prupis Petigrow v. Gilbertson, 501 U.S. 350, 355-58 (1991) (summarizing proper analysis for borrowing state statutes of limitations), we must reject AITF’s argument for a more fundamental reason.
 Simply put, we have no warrant to expand what little federal common law of necessity fills the interstices of our federated legal system especially in an area where the Congress has legislated to some extent yet stopped short of supplanting a well-developed body of state law. The Supreme Court has been quite clear on this issue in recent years. In O’Melveny Myers v. FDIC, 114 S.Ct. 2048, 2054 (1994), the Court refused to go beyond the terms of the Financial Institutions Reform, Recovery, and Enforcement
Act to displace state law, stating that: “matters left unaddressed in such a [“comprehensive and detailed’] scheme are presumably left subject to disposition provided by state law.”
 Turning to the arena of international banking specifically, we see that the Congress and the Board of Governors of the Federal Reserve System have prescribed in detail the corporate law applicable to an Edge Act corporation, including its powers and limitations. See 12 U.S.C. §§ 611–633; 12 C.F.R. part 211. In this circumstance, O’Melveny Myers suggests that the lawmakers’ silence upon a particular subject relative to Edge Act corporations should be taken to reflect their intention to leave state law undisturbed in that regard — not as an invitation to the federal courts to conjure up a “federal common law untethered to a genuinely identifiable (as opposed to a judicially constructed) federal policy.” 114 S.Ct. at 2055. The Court’s admonition has even greater force where the proposal is to develop a federal commercial law, for the Court has already made clear that the “federal courts should incorporate state law as the federal rule of decision . . . particularly . . . in areas in which private parties have entered into legal relationships with the expectation that their rights and obligations would be governed by state-law standards.” Kamen v. Kemper Financial Services, Inc., 500 U.S. 90, 98 (1991).
 For these reasons, we are reluctant to conclude that the Congress meant to federalize a good portion of commercial law without passing any substantive legislation upon the subject. Therefore, we cannot accept AITF’s argument that the deeming clause of § 632 brings this dispute within the federal question jurisdiction conferred by § 1331.
 2. Jurisdiction under Article III
 Notwithstanding our scruples about fashioning a federal commercial law out of common law cloth, the fact remains that, on its face, § 632 clearly grants the federal district courts jurisdiction over many a matter otherwise unregulated by federal law, such as this suit to hold an Edge Act corporation liable upon the controlling bank’s guaranty of negotiable instruments. Perhaps the Congress enacted § 632 in order to ensure that just in case an issue of federal law should arise in litigation involving an international or foreign banking transaction of an Edge Act corporation, it could be heard in federal court. Such a grant of jurisdiction would be outside of the federal question jurisdiction of § 1331; the federal court’s power (if any) to hear such a case would therefore have to come directly from Article III, § 2 of the Constitution. But there are limits upon the power of the Congress to provide a federal forum for cases that raise only the possibility of a federal question. In view of our obligation to inquire into our own jurisdiction see FW/PBS, Inc. v. City of Dallas, 493 U.S. 215, 231 (1990), and because the applicable choice-of-law rules in such a case arising directly under Article III must depend upon the reason for the exercise of federal jurisdiction, we must first inquire whether the jurisdiction contemplated by § 632 is within “the judicial Power of the United States,” and thus within the constitutional power of the Congress to confer upon the federal courts.
 a. Constitutionality
 In Osborn v. Bank of the United States, 22 U.S. 738
(1824), the Court held that a statute granting the second Bank of the United States the right “to sue and be sued . . . in any circuit court in the United States” was a permissible exercise of the “arising under” jurisdiction authorized in Article III, § 2. Id. at 817. Chief Justice Marshall wrote:
When a question to which the judicial power of the Union is extended by the constitution, forms an ingredient of the original cause, it is in the power of congress to give the circuit courts jurisdiction of that cause, although other questions of fact or of law may be involved in it.
 Id. at 822. Although the issues in dispute i Osborn were matters of contract and thus “unconnected” to federal law, id., the Court made clear that it was enough for the purpose of Article III jurisdiction that the anterior question of the Bank’s authority in federal
law to enter into such a contract must always be presupposed, even if not litigated:
[The Bank] is not only itself the mere creature of a law [of the United States], but all its actions and all its rights are dependent on the same law [of the United States]. . . . The [federal] question forms an original ingredient in every cause.
 Id. at 824.
 A merely potential issue of federal law was said to support the exercise of federal jurisdiction because the legislative, executive, and judicial powers delineated in the Constitution are necessarily coextensive: “The executive department may constitutionally execute every law which the legislature may constitutionally make, and the judicial department may receive from the legislature the power of construing every such law.”Osborn, 22 U.S. at 818. Therefore, if the Congress has validly exercised its power under Article I, § 8 to create federal law, then it may likewise grant the district courts jurisdiction over suits that may implicate the interpretation or application of that law.
 Thus it was that in Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480 (1983), the Court upheld the grant of federal jurisdiction in the Foreign Sovereign Immunities Act: that law “codifies the standards governing foreign sovereign immunity as an aspect of substantive federal law and applying those standards will generally require interpretation of numerous points of federal law.” Id. at 497. By parity of reasoning in Mesa v. California, 489 U.S. 121, 136-39
(1989), the Court rejected an interpretation of 28 U.S.C. § 1442(a) under which a defendant federal officer could remove a case to federal court without alleging a defense based upon federal law. Such a construction would render the law “a pure jurisdictional statute, seeking to do nothing more than grant district court jurisdiction over cases in which a federal officer is a defendant,” and therefore likely impermissible under Article III. Mesa, 489 U.S. at 136. As Mesa
illustrates, the Court has rejected the notion of a “protective jurisdiction” that goes beyond the reach of any substantive federal law. Id. at 137-39; see also John T. Cross, Congressional Power to Extend Federal Jurisdiction to Disputes Outside Article III: A Critical Analysis from the Perspective of Bankruptcy, 87 Nw. U.L. Rev. 1188, 1210-14 (1993) (summarizing various theories of protective jurisdiction).
 While the broad implication of Osborn — that the Congress may constitutionally provide a federal forum for any case presenting the mere possibility that a question of federal law will arise — has been seriously questioned, see, e.g., Verlinden, 461 U.S. at 492-93, and its practical effect narrowed both by judicial decision, see Gully v. First National Bank, 299 U.S. 109, 112-14 (1936) (tracing history), and by legislation, see 28 U.S.C. § 1349
(foreclosing federal jurisdiction based solely upon federal incorporation); Murphy v. Colonial Federal Savings and Loan Association, 338 F.2d 609, 611-12 (2d Cir. 1967) (Friendly, J.), those developments prevent only the “sue and be sued” provision of the Edge Act, 12 U.S.C. § 614, from serving as an unrestricted visa for any Edge Act corporation to get into federal court at will. Thus the narrow holding o Osborn is still good law: the Congress may constitutionally make a specific grant of federal jurisdiction over cases involving a federally created entity so long as the legislature does more than “merely confer[ ] a new jurisdiction on the district courts,” Verlinden, 461 U.S. at 496
(quoting The Propeller Genesee Chief v. Fitzhugh, 53 U.S. 443, 451 (1851)); what the Congress may not do is grant a federal forum to a case devoid of federal law. See, e.g., Mesa, 489 U.S. at 136. The Court recently confirmed this point in American National Red Cross v. S.G., 112 S.Ct. 2465, 2475-76 (1992), stating that it was “loathe to repudiate such a long-standing and settled rule.”
 The implication of all of this is that for § 632 constitutionally to confer upon a federal court jurisdiction over a suit by or against a federally chartered corporation “arising out of transactions involving international or foreign banking,” it must do so for the general
purpose of ensuring the proper administration of some federal law (although the disputed issues in any specific case may be confined to matters of state law). It is not enough simply to decide whether “federal interests” are at stake; that would be to adopt the very notion of a protective jurisdiction that the Supreme Court has consistently disavowed. Instead, we must look to the substantive federal law anchoring the federal jurisdiction invoked by § 632 and ask whether the potential application of that law provides a sufficient predicate for the exercise of the federal judicial power — that is, whether “the title or right set up by a party, may be defeated by one construction of the . . . laws of the United States, and sustained by the opposite construction.” Verlinden, 462 U.S. at 497 (quoting Osborn, 22 U.S. at 822).
 The substantive law implicated in the grant of jurisdiction found in § 632 is the Edge Act, which was added to the Federal Reserve Act in 1919 to provide for “[c]orporations to be organized for the purpose of engaging in international or foreign banking or other international or foreign financial operations.” 41 Stat. 378, 66th Cong., 2d Sess (Dec. 24, 1919); codified as amended at 12 U.S.C. § 611 et seq. The Edge Act addressed issues of corporate governance and gave the Federal Reserve Board broad powers to set specific rules of operation Id.
 Section 632 was in turn added to the Federal Reserve Act by the Banking Act of 1933, a/k/a the Glass-Steagall Act. 48 Stat. 162, 184, 73d Cong., 1st Sess. § 15 (June 16, 1933). The legislative history of § 632 is of little help in divining its purpose. Indeed, its subtleties appear to have been lost on at least some legislators. The eminent Senator Glass disavowed any appreciation of its import. 75 Cong. Rec. at 9889. Representative Luce, summarizing the bill in the House, said of what eventually became § 632:
Then there is a page or more that only a lawyer can understand, and not having been actively engaged in the practice of law for a long time, I dare not try to explain it to you. It is something about taking Federal Reserve bank business into the United States district courts. I suppose it is desirable. You will have to take that for granted so far as I go.
 77 Cong. Rec. at 3916.
 Looking back to the Edge Act itself, however, one can divine the likely reasons for the grant of federal jurisdiction that would follow 14 years later. Crafted in the wake of the turmoil that the World War had caused in international financial markets, the Edge Act called forth a new type of federally controlled institution intended to increase the stability of, and the public’s confidence in, international markets. See 59 Cong. Rec. at 50 (Act “furnishes . . . machinery for financing foreign trade, a trade more essential to our prosperity than ever before, and more essential also to the war-torn countries of Europe”) (Rep. Platt). Federal supervision of these financial institutions was seen as essential if they were ever to succeed in the international marketplace. Thus a Governor of the Federal Reserve Board would tell the Senate Committee on Banking and Currency that:
The time will probably come when the conflict of the dual control exercised by the Federal Reserve Board and by the banking department of a State may be a matter of embarrassment or operate to restrict the activities of the banking corporation[, and] the benefits and protection of a Federal charter . . . would be of great value in competing for business in foreign countries.
 S. Rep. No. 108, 66th Cong., 1st Sess. 2 (July 25, 1919) see also 59 Cong. Rec. at 622 (remarks of Senator McLean expressing hope that similar state-chartered institutions “would come in under this Federal law, where they would be supervised and examined”).
 We infer, therefore, that the substantive federal regulations that the Congress placed upon Edge Act corporations, to be supplemented by the oversight of the Federal Reserve Board, are intended to facilitate and stimulate international trade by providing the uniformity of federal law. The Edge Act regime is unquestionably a valid exercise of the Congress’s powers under Article I, § 8, and its substance lies close enough to the heart of any case involving an international
transaction with an Edge Act bank to sustain the assertion of federal subject-matter jurisdiction. As in Osborn, where it was enough for federal jurisdiction that the Bank’s ability to enter into a transaction depended upon the terms of its federal charter, 22 U.S. at 823, an issue of federal law might well arise in a suit involving a foreign or international banking transaction of an Edge Act corporation. As detailed below (at 17-18), an Edge Act corporation’s powers and limitations are governed by specific provisions of federal law, see 12 U.S.C. §§ 612–31, and divers interpretations of those substantive provisions might very well lead to conflicting results vexing to the very commerce that the Edge Act was enacted to promote See, e.g., Republic of Panama v. Republic National Bank of New York, 681 F. Supp. 1066 (S.D.N.Y. 1988) (resolving case in part upon interpretation of Edge Act provision giving presumptive validity to acts of certified representatives of foreign states) Rose Hall, Ltd. v. Chase Manhattan Overseas Banking Corp., 576 F. Supp. 107 (D. Del. 1983) (last in series of decisions resolving case in part upon interpretation of Edge Act provision concerning limitations on stock ownership). The prospect of divergent outcomes in the 50 state courts of last resort is something that the Congress has the power to avoid by bringing all such disputes within the unifying jurisdiction of the federal courts, regardless whether the issues in dispute in a particular case present that problem. There may have been other reasons for the enactment of § 632 — the unavailability of diversity jurisdiction under the rule of Bankers Trust comes to mind — but be that as it may, we think there is enough substantive federal law underlying the grant of jurisdiction in § 632 to render it constitutional.
 b. Application
 We now consider whether this particular brand of federal jurisdiction calls for the application of a federal or of a local choice-of-law rule. We conclude that where the “federal question” giving rise to federal jurisdiction need not appear upon the face of a well-pleaded complaint, there is no reason for the federal court to conduct any different choice-of-law inquiry than would a court of the forum state in deciding the same issue.
 In other settings in which a federal court must rule upon an issue regulated only by state law, it applies the forum state’s choice-of-law rules and the state statute of limitations indicated thereby. It does so when exercising its supplemental jurisdiction under 28 U.S.C. § 1367, see United Mine Workers of America v. Gibbs, 383 U.S. 715, 726 (1966) Erie, 304 U.S. at 78 (“Except in matters governed by the Federal Constitution or by Acts of Congress, the law to be applied in any case is the law of the State”); when sitting in diversity under 28 U.S.C. § 1332, Klaxon; and when hearing a claim under the Foreign Sovereign Immunities Act Gilson v. Republic of Ireland, 682 F.2d 1022
(D.C. Cir. 1982); Barkanic v. General Administration of Civil Aviation of the People’s Republic of China, 923 F.2d 957 (2d Cir. 1991). Cf. Richards v. United States, 369 U.S. 1 (1962) (Congress specified choice-of-law rule in Federal Tort Claims Act).
 We see no reason to remove the Edge Act from what appears to be the general rule: a federal court applies state law when it decides an issue not addressed by federal law, regardless of the source from which the cause of action is deemed to have arisen for the purpose of establishing federal jurisdiction. The Edge Act contains many specifics in the way of corporate governance. It provides for the particulars of incorporation, 12 U.S.C. §§ 612–614; lists the powers of, as well as the limitations upon, such corporations, §§ 615-617; lays down capital requirements and restricts stock ownership both in and by the corporation, §§ 618-21; makes provision for the dissolution of the corporation, §§ 622-623; and for receivership, § 624; requires stockholder meetings, § 625; governs the payment of certain dividends, § 626; allows for state taxation, § 626; and provides criminal penalties for offenses by corporate officers, §§ 630-631. Clearly, the Congress wished to avoid the variation in these matters that would surely occur if like institutions were incorporated under the laws of the several states.
 All indications are that an Edge Act corporation is subject to state law, however, outside the areas of corporate concern listed above. Neither the Congress nor the Federal Reserve Board has purported to regulate such commercial law questions as the liability of an Edge Act corporation upon a guaranteed note or when such a corporation may be held liable for the acts of its corporate parent. The Congress apparently intended that day-to-day transactions be governed by whatever law would apply if the institution involved were not chartered under the Edge Act. Of course, when § 632 was first enacted, such questions were (in federal courts) decided according to “the general principles and doctrines of commercial jurisprudence,”Swift v. Tyson, 41 U.S. 1, 19 (1842), but soon thereafter (in Erie) the Supreme Court made it quite clear that “[t]here is no federal general common law.”304 U.S. at 78. Though the Congress and the Board have had more than fifty years to leap into the breach with positive law, there is still no federal statute or regulation governing the commercial transactions of financial institutions chartered under the Edge Act, and we simply may not craft such rules ourselves.
 Nor, under the rule of Klaxon, may we craft a choice-of-law rule out of whole cloth. The Erie
doctrine is not simply a rule of convenience for diversity cases but an acknowledgment of the powers of the several states and of the limited nature of the federal government. See
Friendly, In Praise of Erie — And of the New Federal Common Law, 39 N.Y.U. L. Rev. 383, 408 n. 122 (1964) (characterizing as a “heresy” the notion tha Erie is applicable only to diversity cases; rather, “th Erie doctrine applies, whatever the ground for federal jurisdiction, to any issue or claim which has its source in state law”). A choice-of-law rule is no less a rule of state law than any other; therefore, to adopt any rule other than that o Klaxon would, in the words of the Supreme Court, “do violence to the principle of uniformity within a state, upon which [Erie] is based.” Klaxon, 313 U.S. at 496. In view of the reasoning in Erie, and its explicit extension to choice-of-law rules in Klaxon, it would make little sense for us to conclude that the peculiar brand of federal jurisdiction provided for certain Edge Act transactions calls for a different rule, i.e., one that would undermine uniformity between the state and federal courts sitting within a given jurisdiction. We therefore conclude that the District of Columbia’s choice-of-law rules apply to AITF’s suit against PIBC on the promissory notes guaranteed by Petra Bank, and that those rules call for the application of a D.C. statute of limitations.
 A moment’s reflection upon the practical implications of this decision only confirms our confidence in it. The Supreme Court has decreed that in the absence of federal legislation there shall be what one might call “vertical uniformity”: a suit in federal court shall be handled as it would be in the courts of the state where that federal court sits. One can easily imagine a “horizontal uniformity” among the federal courts regardless of state boundaries. Not all banking disputes will be heard in federal court, however, so the attraction of such a regime is superficial, as the Supreme Court noted when it overrule Swift v. Tyson: “[T]hough doubtless intended to promote uniformity in the operation of business transactions, [that case’s] chief effect has been to render it difficult for business men to know in advance [what law applies].” Erie, 304 U.S. at 74 n. 8. That would be the effect also if some litigation involving an Edge Act corporation were subject to the horizontal “uniformity” of a federal choice-of-law rule while the vast majority of similar banking transactions are subject to the vertical uniformity of state choice-of-law rules unde Klaxon.
 Finally, any problem that may arise from holding international commercial transactions subject to the laws, including the choice-of-law rules, of the 50 states is easily remedied by the Congress (or perhaps by the Federal Reserve Board, see 12 U.S.C. § 611a). Thus far, as noted above, the legislature has not (nor has the Board) displaced the state except with regard to certain aspects of corporation law, presumably because it does not believe that it need do any more in order to give Edge Act corporations stability and credibility in the marketplace. Whether the needs of international commerce now demand
more federal law is not a matter for the judgment of the courts.
 3. Other circuit court decisions
 We are aware that two other circuits have reached the contrary conclusion. See, e.g., Edelman v. Chase Manhattan Bank, N.A., 861 F.2d 1291, 1293-1301 (1st Cir. 1988) Corporacion Venezolana de Fomento v. Vintero Sales Corp., 629 F.2d 786, 795 (2d Cir. 1980). Neither of those courts seems to have considered the precise character of the federal jurisdiction at work in a suit involving an Edge Act corporation, however. They also wrote without the benefit of some of the Supreme Court’s more recent pronouncements upon the limited domain of federal common law. Moreover, in many of their cases, the inquiry was purely academic because the choice-of-law rules of the forum state would have pointed to the law of the same state as did the court pursuant to its federal choice-of-law inquiry. See, e.g., Wells Fargo Asia, Ltd. v. Citibank, N.A., 936 F.2d 723, 726-27 (2d Cir. 1991) (holding New York law applies under either analysis).
 In Edelman the First Circuit did not discuss the choice-of-law issue in depth. It merely stated in a footnote that “[w]hen jurisdiction is not based upon diversity of citizenship, choice of law questions are appropriately resolved as matters of federal common law.” 861 F.2d at 1294 n. 14. Moreover, the authority cited in Edelman for this proposition is Vintero, which concerned what jurisdiction’s substantive law should govern the dispute, not which statute of limitations to apply; the court there proclaimed that it was deciding “a federal question case,” for which conclusion it cited Hinderlider v. LaPlata River Cherry Creek Ditch Co., 304 U.S. 92 (1938), an Clearfield Trust, 318 U.S. 363. Vintero, 629 F.2d at 795.
 Neither Hinderlider nor Clearfield Trust,
however, dictates the result reached by the First and Second Circuits. Both cases deal with the application of substantive federal common law, not with which choice-of-law principles to apply when state law governs the substantive issues in a federal case. Thus, Hinderlider concerned the apportionment between two states of the water of an interstate stream, presenting the Court with a situation in which “neither the statutes nor the decisions of the states can be conclusive.”304 U.S. at 110. Likewise, Clearfield Trust concerned the law that applies to the exercise of the United States’s constitutional power to disburse funds; not surprisingly, the Court concluded that “[t]he rights and duties of the United States on commercial paper which it issues are governed by federal rather than local law.” 318 U.S. at 366. There are no such strong federal interests in this case, where AITF’s rights and duties under Petra Bank’s guaranty are governed by state contract law.
 In any event, we think the better rule is the one that the Second Circuit has itself endorsed in a somewhat different context: “It is the source of the right, not the basis of federal jurisdiction, which determines the controlling law.”Van Gemert v. Boeing Co., 553 F.2d 812, 813 (2d Cir. 1977).
 III. The Statute of Limitations
 We return now to AITF’s first claim. It argues that whatever statute of limitations properly applies to its claim against PIBC for any liability that AITF might incur in its litigation with Centro, that statute has not yet begun to run, let alone run its course, because AITF has not yet been held liable to Centro in the New York litigation. (AITF admits that its cause of action on the three remaining notes accrued, at the latest, upon their dishonor by Petra Bank in January 1989.)
 A. Mootness
 At the outset, we must determine whether that part of AITF’s claim based upon the Centro notes is moot, lest we stray beyond the limits of our jurisdiction under Article III, § 2 of the Constitution. AITF itself claims in its litigation with Centro that Centro’s claim based upon the notes is moot in light of the Vienna court’s ruling that Centro is not liable to Amro. If this holds true, it would seem to follow that any suit to declare PIBC liable to AITF on the notes that AITF resold to Centro would also be
moot. That is a big “if,” however. As AITF points out, Centro is contesting its motion to dismiss and at least until the New York court rules upon that motion, there is a possibility that AITF will be held liable to Centro. Moreover, even assuming that the New York court eventually grants AITF’s motion, AITF’s claim against PIBC would still not be moot because AITF has incurred attorneys fees in litigation with Centro for which it claims that Petra Bank (and through it PIBC) is responsible.
 Ordinarily “[a] request for attorney’s fees does not preserve a case which is otherwise moot.” Friends of Keesville, Inc. v. FERC, 859 F.2d 230, 233 n. 7 (D.C. Cir. 1988). As we pointed out in Washington Hospital Center National Rehabilitation Hospital v. Collier, 947 F.2d 1498, 1502
(D.C. Cir. 1991), however, this observation has no place where the attorney’s fees are part and parcel of the damages stemming from the defendant’s alleged breach of contract. I Collier, as here, the plaintiff sought attorney’s fees not in a collateral proceeding predicated upon the defendant’s having forced it to sue but “as an element of damages stemming from [the defendant’s] alleged breach of the contract now in suit.” Id. AITF’s claim that PIBC is responsible for the expenses it incurred in its suit with Centro, on the theory that PIBC’s breach of contract caused this loss, is therefore live without regard to the fate on appeal of AITF’s underlying claim.
 B. Accrual
 The claim for attorney’s fees is itself subject to the applicable D.C. statute of limitations, of course. AITF continues to pile up damages in the form of attorney’s fees, but that does not toll the running of the statute any more than does the continuing accrual of lost wages or continued pain and suffering in a tort case. We must therefore decide whether AITF’s cause of action against PIBC (via Petra Bank) on the promissory notes that AITF sold to Centro accrued upon Petra Bank’s failure to pay the notes as they came due (in October of 1989 and January of 1990) or, as AITF claims, will accrue only if and when AITF is held liable to Centro.
 The answer to this question depends upon whether AITF is trying to enforce a guaranty or is claiming indemnity. For “under District of Columbia law the statute of limitations on [indemnification] claims begins to run only after a judgment has been paid,” Long v. District of Columbia, 820 F.2d 409, 417 (D.C. Cir. 1987), while “a cause of action against a maker or an acceptor accrues in the case of a time instrument on the day after maturity.” D.C. Code § 28:3-122(1)(a). A guarantor is not precisely the same as either a maker or an acceptor, compare D.C. Code § 28:3-413, with D.C. Code § 28:3-416, but in this case AITF claims that Petra Bank’s “aval” effectively renders PIBC liable “not only as a guarantor, but as co-obligor [on the notes] as well.” By AITF’s own theory of liability, therefore, PIBC should be treated as would the maker of the notes; as such, under D.C. law AITF’s cause of action has long since accrued.
 AITF tries nevertheless to characterize its claim against PIBC on the Centro notes as a claim for indemnity. (In its complaint it asked the court to “[d]eclare that [PIBC] must indemnify AITF for any settlement or judgment entered by or against AITF in the Centro action, as well as for all costs, expenses, and attorneys’ fees incurred by AITF in connection with that litigation.”) There is no mention in the complaint, however, of any facts that could possibly give rise to a contract of indemnity, as opposed to a guaranty. To the contrary, the key provision of the complaint linking Petra Bank (and arguably PIBC) to the notes states that “Petra [Bank] was unconditionally obligated to honor the Notes at their maturity.” Indeed the presence of Petra Bank’s aval denotes “an unconditional guarantee under which the “avalor’ is obliged to pay the debt obligation as if it were the primary debtor.” FR. Winkler KG v. Stoller, 839 F.2d 1002, 1009 (4th Cir. 1988). It defies common sense, as well as the more sophisticated rationales for having a statute of limitations, for AITF to assert at the same time that PIBC is liable on the notes from the time they come due and that the statute of limitations on a claim to enforce this obligation does not begin to run until some later date.
 Surely AITF’s sale of three notes to Centro cannot make Petra Bank liable on those instruments for longer than it is liable on the three notes that AITF still owns. We therefore conclude that the D.C. statute of limitations that governs the second count of AITF’s complaint in this case began to run the day after the notes matured.
 C. Tolling
 AITF further argues that even if the D.C. statute of limitations applies and that statute began to run as of Petra Bank’s refusal to pay the notes, the statute was tolled when AITF filed its complaint against Petra Bank in the Southern District of New York in November 1989. The law of the District of Columbia is quite clear on that score: filing an action in another jurisdiction does not toll the statute Namerdy, 217 A.2d at 113. AITF counters that under federal law the answer is otherwise, citing, e.g., Atkins v. Schmutz Mfg. Co., 435 F.2d 527, 527-28 (4th Cir. 1970) (holding federal law governs effect of identical federal case filed in another jurisdiction), but we expressly disagreed with that position in Walko Corporation v. Burger Chef Systems, Inc., 554 F.2d 1165, 1171 n. 51 (D.C. Cir. 1977) (holding limitation on state-created cause of action determined according to state law), and we see no reason to change our view now. For the same reasons that we declined above to invent any federal common law with regard to the applicable statute of limitations itself, we shall not engage in a similar exercise now in order to determine whether the filing of a case in another jurisdiction operates to toll that statute. We therefore hold (under D.C. law) that AITF’s filing against Petra Bank in New York did not toll the limitation period for filing its claim against PIBC in the District of Columbia.
 D. Contracts under Seal
 Finally, AITF argues that even if the D.C. statute of limitation applies, has begun to run, and has not been tolled, the applicable period of limitation is the 12 years provided to sue upon a contract under seal rather than the three-year limit upon all other contract actions. While the notes giving rise to this action were stamped with Petra Bank’s corporate seal, that does not automatically render them obligations made “under seal.” A signature over the word “seal” may operate to place a contract under seal, see Phillips v. A C Adjusters, Inc., 213 A.2d 586, 586-87 (D.C.App. 1965) (relying upon Wells v. Aropa Corp., 82 F.2d 887 (D.C. Cir. 1936)), but the use of a corporate seal is only a mark of identification and genuineness absent any indication, in the document itself or elsewhere, that the parties intended to create an obligation under seal Fox-Greenwald Sheet Metal Co., Inc. v. Markowitz Bros., Inc., 452 F.2d 1346, 1357 n. 68 (D.C. Cir. 1971) (fashioning District of Columbia law, see Steorts v. American Airlines, Inc., 647 F.2d 194, 196 n. 19 (D.C. Cir. 1981)). AITF points to no such indication. We therefore conclude that the three-year limitation period for contracts not under seal applies in the present case.
 IV. Conclusion
 Regardless whether the district court had jurisdiction pursuant to 28 U.S.C. § 1331, it did have jurisdiction pursuant to 12 U.S.C. § 632 over AITF’s claim that PIBC is liable on the six promissory notes guaranteed by Petra Bank. Under D.C. law the period of limitation applicable to such a claim is the three-year period in which to bring an action based upon a contract. The statute having run before AITF filed this suit against PIBC, its claim is barred. Accordingly, the judgment of the district court is