IRVING R. KAUFMAN, Circuit Judge:
These four appeals grow out of one of the most enormous commercial disputes in history, and present questions which strike to the very heart of the modern international economic order. An African nation, developing at breakneck speed by virtue of huge exports of high-grade oil, contracted to buy huge quantities of Portland cement, a commodity crucial to the construction of its infrastructure. It overbought, and the country’s docks and harbors became clogged with ships waiting to unload. Imports of other goods ground to a halt. More vessels carrying cement arrived daily; still others were steaming toward the port. Unable to accept delivery of the cement it had bought, the nation repudiated its contracts. In response to suits brought by disgruntled suppliers, it now seeks to invoke an ancient maxim of sovereign immunity — par in parent imperium non habet1 — to insulate itself from liability. But Latin phrases speak with a hoary simplicity inappropriate to the modern financial world. For the ruling principles here, we must look instead to a new and vaguely-worded statute, the Foreign Sovereign Immunities Act of 1976 (“FSIA” or “Act”)2 — a law described by its draftsmen as providing only “very modest guidance” on issues of preeminent importance.3 For answers to those most difficult questions, the authors of the law “decided [303]*303to put [their] faith in the U.S. courts.”4 Guided by reason, precedent, and equity, we have attempted to give form and substance to the legislative intent. Accordingly, we find that the defense of sovereign immunity is not available in any of these four cases.5
I.
The facts of the four appeals are remarkably parallel, and can be stated in somewhat consolidated form.6 Early in 1975, the Federal Military Government of the Federal Republic of Nigeria (“Nigeria”) embarked on an ambitious program to purchase immense amounts of cement. We have already had occasion in another case to call the program “incredible,” see National American Corp. v. Federal Republic of Nigeria, 597 F.2d 314, 316 (2d Cir. 1979), but the statistics speak for themselves. Nigeria executed 109 contracts, with 68 suppliers. It purchased, in all, over sixteen million metric tons of cement. The price was close to one billion dollars.
A.
Four of the 109 contracts were made with American companies that were plaintiffs below in the cases now before us: Texas Trading & Milling Corp. (“Texas Trading”), Decor by Nikkei International, Inc. (“Nikkei”), East Europe Import-Export, Inc. (“East Europe”), and Chenax Majesty, Inc. (“Chenax”). The four plaintiffs are not industrial corporations; they are, instead, “trading companies,” which buy from one person and sell to another in hopes of making a profit on the differential. Each of the plaintiffs is a New York corporation.
The contracts at issue were signed early in 1975. Each is substantially similar; indeed, Nigeria seems to have mimeographed them in blank, and filled in details with individual suppliers. Overall, each contract called for the sale by the supplier to Nigeria of 240,000 metric tons of Portland cement.7 Specifically, the contracts required Nigeria, within a time certain after execution,8 to establish in the seller’s favor “an Irrevocable, Transferable abroad, Divisible and Confirmed letter of credit” for the total amount due under the particular contract, slightly over $14 million in each case.9 The contract also named the bank through which the letter of credit was to be made payable. Nikkei and East Europe named First National City Bank in New York, and Texas Trading specified Fidelity International Bank, also in New York. Chenax denominated Schroeder, Muenchmeyer, Hengst & Co. of Hamburg, West Germany. Drafts under the letters of credit were to be “payable at sight, on presentation” of certain documents to the specified bank.
Within a time certain after establishment and receipt of the letter of credit,10 each [304]*304seller was to start shipping cement to Nigeria. The cement was to be bagged, and was to meet certain chemical specifications. Shipments were to be from ports named in the contracts, mostly Spanish, and were to proceed at approximately 20,000 tons per month. Delivery was to the port of Lagos/Apapa, Nigeria, and the seller was obligated to insure the freight to the Nigerian quay. Each contract also provided for de-murrage.11 The Nikkei and East Europe contracts provided they were to be governed by the laws of the United States. The Chenax contract specified the law of Switzerland, and the Texas Trading contract named the law of Nigeria.
In short, performance under the contracts was to proceed as follows. Nigeria was to establish letters of credit. The suppliers were to ship cement. Each time a supplier had loaded a ship and insured its cargo to Lagos/Apapa, the supplier could take documents so proving to the bank named in the contract and, “at sight,” be paid for the amount of cement it shipped. The ship might sink on the way to Nigeria, or it might never leave the Spanish port at all, but — on presentation of proper documents showing a loaded ship and an insured cargo — the supplier had a right to be paid. Demurrage was to operate in the same manner: if a ship was detained in Nigerian waters, the supplier would receive certain documents. It could present the documents to the bank, and receive payment.
The actual financial arrangements differed from those set forth in the cement contracts. Instead of establishing “confirmed” letters of credit with the banks named, Nigeria established what it called “irrevocable” letters of credit with the Central Bank of Nigeria (“Central Bank”), an instrumentality of the Nigerian government,12 and advised those letters of credit through the Morgan Guaranty Trust Company (“Morgan”) of New York. That is, under the letters of credit as established, each seller was to present appropriate documents not to the named bank, but to Morgan. And, since the letters were not “confirmed,” Morgan did not promise to pay “on sight”; it assumed no independent liability.13 Each of the letters of credit provided it was to be governed by the Uniform Customs and Practice for Documentary Credits (“UCP”) (1962 Revision), as set forth in Brochure No. 222 of the International Chamber of Commerce.
The principal objects of the Bank shall be to issue legal tender currency in Nigeria, to maintain external reserves to safeguard the international value of that currency, to promote monetary stability and a sound financial structure in Nigeria and to act as banker and financial adviser to the Federal Government.
Nigeria’s choice of Morgan as the bank to which suppliers presented documents and from which suppliers secured payments came in the course of a longstanding relationship between Nigeria and Morgan. Central Bank used Morgan as its correspondent bank in the United States, and Morgan conducted myriad transactions on Nigeria’s behalf. Employees of Central Bank regularly came to Morgan for training seminars. On Nigeria’s request, Morgan made payments to Nigerian students in the United States, to American corporations to which Nigeria owed money, and to the Nigerian embassy and consulates in the United States. Indeed, Nigeria used Morgan to make payments (for salaries, operating expenses, and the like) to Nigerian embassies in other countries as well. Until 1974, Morgan had the right to draw up to $1 million per day from Nigeria’s account at the Federal Reserve Bank of New York to satisfy Nigeria’s obligations. Nigeria raised the limit to $3 million per day in 1974, and [305]*305Morgan enjoyed unlimited drawing rights on Nigeria’s funds beginning in November 1975. Central Bank kept over $200 million of securities in a custody account at Morgan. Morgan advised as much as $200 million in letters of credit established by Nigeria, and confirmed, in addition, letters of credit totalling at least $70 million more.
After receiving notice that the letters of credit had been established, the suppliers set out to secure subcontracts to procure the cemént, and shipping contracts to transport it.14 They, through their subcontractors, began to bag the cement and load it on ships, as suppliers across the globe were doing the same. Hundreds of ships arrived in Lagos/Apapa in the summer of 1975, and most were carrying cement. Nigeria’s port facilities could accept only one to five million tons of cement per year; at any rate, they could not begin to unload the over sixteen million tons Nigeria had slated for delivery in eighteen short months. Based on prior experience, Nigeria had made the contracts expecting only twenty percent of the suppliers to be able to perform. By July, when the harbor held over 400 ships waiting to unload — 260 of them carrying cement — Nigeria realized it had misjudged the market considerably.
C.
With demurrage piling up at astronomical rates, and suppliers, hiring, loading, and dispatching more ships daily, Nigeria decided to act. On August 9, 1975, Nigeria caused its Ports Authority to issue Government Notice No. 1434, a regulation which stated that, effective August 18, all ships destined for Lagos/Apapa would be required to convey to the Ports Authority, two months before sailing, certain information concerning their time of arrival in the port. The regulation also stated vaguely that the Ports Authority would “co-ordinate all sailing,” and that it would “refus[e] service” to vessels which did not comply with the regulation. Then, on August 18, Nigeria cabled its suppliers and asked them to stop sending cement, and to cease loading or even chartering ships. In late September, Nigeria took the crucial step: Central Bank instructed Morgan not to pay under the letters of credit unless the supplier submitted — in addition to the documents required by the letter of credit as written — a statement from Central Bank that payment ought to be made. Morgan notified each supplier of Nigeria’s instructions, and Morgan commenced refusing to make payment under the letters of credit as written. Almost three months later, on December 19, 1975, Nigeria promulgated Decree No. 40, a law prohibiting entry into a Nigerian port to any ship which had not secured two months’ prior approval, and imposing criminal penalties for unauthorized entry.
Nigeria’s unilateral alteration of the letters of credit took place on a scale previously unknown to international commerce. Officers of Morgan explained the potential consequences of Nigeria’s action to representatives of Central Bank; Central Bank was adamant that Morgan not pay. After a meeting with Central Bank personnel, one Morgan officer stated that Central Bank’s Deputy Governor “responded that the [Nigerian] Government was willing to go to court if we did pay.” Within weeks of Nigeria’s instructions to Morgan not to pay without the additional documentation, Morgan warned Central Bank in a telex: “We believe that there is an increasing possibility that litigation against you may be instituted in New York.”
Nigeria’s next step was to invite its suppliers to cancel the contracts. As part of the program, Nigeria convened a meeting at Morgan’s offices in New York, to discuss [306]*306Nigeria’s position with members of the American financial community. Over forty suppliers eventually did settle. See National American Corp. v. Federal Republic of Nigeria, supra, 597 F.2d at 316. Nigeria asked Morgan to effect several of the settlement payments; some were for settling suppliers not located in the United States.
Cement suppliers who did not settle sued in courts all over the world.15 The four suppliers at issue here — Texas Trading, Nikkei, East Europe, and Chenax — sued in the Southern District of New York. Named as defendants were both Nigeria and Central Bank. The complaints alleged that Central Bank’s September instructions to Morgan, changing the terms of payment under the letters of credit, constituted anticipatory breaches of both the cement contracts (requiring Nigeria to establish “Irrevocable” letters of credit with certain terms of payment) and the letters of credit (requiring Central Bank to authorize payment when certain documents were presented to Morgan). Defendants do not seriously dispute that their actions constitute such anticipatory breaches;16 their defenses go more to the propriety of jurisdiction under the FSIA. Judge Cannella in Texas Trading, D.C., 500 F.Supp. 320, found jurisdiction lacking; Judge Pierce in the consolidated Nikkei, East Europe, and Che-nax actions, D.C., 497 F.Supp. 893, held it present. Judge Pierce proceeded to a trial on the merits, and awarded $1.857 million to Nikkei, $1.986 million to East Europe, and nothing to Chenax. These appeals followed.17
II.
The law before us is complex and largely unconstrued, and has introduced sweeping changes in some areas of prior law. See House Judiciary Committee, Jurisdiction of United States Courts in Suits Against Foreign States, H.R.Rep. No. 1487, 94th Cong., 2d Sess. 7-8, reprinted in [1976] U.S.Code Cong. & Admin.News 6605, 6604-6 (“House Report ”).18 In structure, the FSIA is a marvel of compression. Within the bounds of a few tersely-worded sections, it purports to provide answers to three crucial questions in a suit against a foreign state: the availability of sovereign immunity as a defense, the presence of subject matter jurisdiction over the claim, and the propriety of personal jurisdiction over the defendant. See House Report at 6611-12. Through a series of intricately coordinated provisions, the FSIA seems at first glance to make the answer to one of the questions, subject matter jurisdiction, dispositive of all three. See Hearings on H.R. 11315 Before the Subcommittee on Administrative Law and Governmental Relations of the House Committee on the Judiciary, 94th Cong., 2d Sess. 28 (1976) (“1976 Hearings”) (testimony of [307]*307Monroe Leigh, Legal Adviser, Department of State).19 This economy of decision has come, however, at the price of considerable confusion in the district courts. In fact, Congress intended the sovereign immunity and subject matter jurisdiction decisions to remain slightly distinct, and it drafted the Act accordingly. Moreover, Congress has only an incomplete power to tie personal jurisdiction to subject matter jurisdiction; its prerogatives are constrained by the due process clause. These cases present an opportunity to untie the FSIA’s Gordian knot, and to vindicate the Congressional purposes behind the Act.
Turning to the specific provisions of the law, a description of the FSIA’s analytic structure is helpful. The jurisdiction-conferring provision of the Act, 28 U.S.C. § 1330(a), creates in the district courts:
original jurisdiction without regard to amount in controversy of any nonjury civil action against a foreign state as defined in section 1603(a) of this title as to any claim for relief in personam with respect to which the foreign state is not entitled to immunity either under sections 1605-1607 of this title or under any applicable international agreement.
Although § 1330(a) refers to sections 1605-1607, the section most frequently relevant,20 and the one applicable here, is § 1605. It provides, in part:
(a) A foreign state shall not be immune from the jurisdiction of courts of
the United States or of the States in any case—
******
(2) in which the action is based upon a commercial activity carried on in the United States by the foreign state; or upon an act performed in the United States in connection with a commercial activity of the foreign state elsewhere; or upon an act outside the territory of the United States in connection with a commercial activity of the foreign state elsewhere and that act causes a direct effect in the United States.
Crucial to each of the three clauses of § 1605(a)(2) is the phrase “commercial activity.” In it is lodged centuries of Anglo-American and civil law precedent construing the term “sovereign immunity.” If the activity is not “commercial,” but, rather, is “governmental,” then the foreign state is entitled to immunity under section 1605, and “original jurisdiction” is not present under § 1330(a).21
For the definition of “commercial activity,” we turn to subsection 1603(d), which provides:
(d) A “commercial activity” means either a regular course of commercial conduct or a particular commercial transaction or act. The commercial character of an activity shall be determined by reference to the nature of the course of conduct or particular transaction or act, rather than by reference to its purpose.
[308]*308If “commercial activity” under § 1603(d) is present, and if it bears the relation to the United States required by § 1605(a)(2), then the foreign state is “not entitled to immunity,” and the district court has statutory subject matter jurisdiction over the claim through § 1330(a). And, if the exercise of that jurisdiction falls within the judicial power set forth by Article III of the Constitution, subject matter jurisdiction over the claim exists.22
Our analysis next proceeds to the question of personal jurisdiction; we come again to § 1330. Subsection (b) of section 1330 provides: “Personal jurisdiction over a foreign state shall exist as to every claim for relief over which the district courts have jurisdiction under subsection (a) where service has been made under section 1608 of this title.” The Act, therefore, makes the statutory aspect of personal jurisdiction simple: subject matter jurisdiction plus service of process equals personal jurisdiction. See House Report at 6622. But, the Act cannot create personal jurisdiction where the Constitution forbids it. Accordingly, each finding of personal jurisdiction under the FSIA requires, in addition, a due process scrutiny of the court’s power to exercise its authority over a particular defendant.
In short, a “§ 1605(a)(2) case” calls for the resolution of a series of five questions:
1) Does the conduct the action is based upon or related to qualify as “commercial activity”?
2) Does that commercial activity bear the relation to the cause of action and to the United States described by one of the three phrases of § 1605(a)(2), warranting the Court’s exercise of subject matter jurisdiction under § 1330(a)?
3) Does the exercise of this congressional subject matter jurisdiction lie within the permissible limits of the “judicial power” set forth in Article III?
4) Do subject matter jurisdiction under § 1330(a) and service under § 1608 exist, thereby making personal jurisdiction proper under § 1330(b)?
5) Does the exercise of personal jurisdiction under § 1330(b) comply with the due process clause, thus making personal jurisdiction proper?
It is to those questions we now turn.
B.
Before undertaking the threshold “commercial activity” analysis, our first task is to identify what particular conduct in this case is relevant. Subsection 1603(d) states that “commercial activity” might consist of either “a regular course of commercial conduct” or “a particular commercial transaction or act.” The words “regular course of ... conduct” seem to authorize courts to cast the net wide, and to identify a broad series of acts as the relevant set of activities. See House Report at 6615. Here, the relevant “course of ... conduct” is undoubtedly Nigeria’s massive cement purchase program. Alternatively, each of its contracts or letters of credit with these four plaintiffs would qualify as “a particular ... transaction.”
The determination of whether particular behavior is “commercial” is perhaps the most important decision a court faces in an FSIA suit. This problem is significant because the primary purpose of the Act is to “restrict” the immunity of a foreign state to suits involving a foreign state’s public acts. House Report at 6605. If the activity is not “commercial,” it satisfies none of the three clauses of § 1605(a)(2), and the foreign state is (at least under that subsection) immune from suit. Unfortunately, the definition of “commercial” is the one issue on which the Act provides almost no guidance at all. Subsection 1603(d) advances the inquiry somewhat, for it provides: “The commercial character of an activity shall be determined by reference to the nature of the course of conduct or particular transaction or act, rather than by reference to its purpose.” No provision of the Act, however, defines “commercial.” Congress de[309]*309liberately left the meaning open and, as noted above, “put [its] faith in the U.S. courts to work out progressively, on a case-by-case basis . . . the distinction between commercial and governmental.” 1976 Hearings at 53 (testimony of Monroe Leigh). Accord, House Report at 6615; Hearings on HR. 3493 Before Subcommittee on Claims and Governmental Relations of the House Committee on the Judiciary, 93d Cong., 1st Sess. 16 (1973) (“1973 Hearings”) (testimony of Charles N. Brower, Legal Adviser, Department of State).23 See also Kahale & Vega, Immunity and Jurisdiction: Toward a Uniform Body of Law in Actions Against Foreign States, 18 ColumJ. Transnat’l L. 211, 236 (1979). We are referred to no less than three separate sources of authority to resolve this fundamental definitional question.
The first source is statements contained in the legislative history itself. Perhaps the clearest of them was made by Bruno Ris-tau, then Chief of the Foreign Litigation Section of the Civil Division, Department of Justice. Ristau stated: “[I]f a government enters into a contract to purchase goods and services, that is considered a commercial activity. It avails itself of the ordinary contract machinery. It bargains and negotiates. It accepts an offer. It enters into a written contract and the contract is to be performed.” 1976 Hearings at 51. The House Report seems to conclude that a contract or series of contracts for the purchase of goods would be per se a “commercial activity,” see House Report at 6615, and the illustrations cited by experts who testified on the bill — contracts, for example, for the sale of army boots24 or grain25 — support such a rule. Or, put another way, if the activity is one in which a private person could engage, it is not entitled to immunity. See 1976 Hearings at 24 (testimony of Monroe Leigh), 53 (same); 1973 Hearings at 15 (testimony of Charles N. Brower).
The second source for interpreting the phrase “commercial activity” is the “very large body of case law which exist[ed]” in American law upon passage of the Act in 1976. See 1976 Hearings at 53 (testimony of Monroe Leigh). See also id. at 94 (testimony of Michael M. Cohen, Chairman, Committee on Maritime Legislation, Maritime Law Association of the United States). Testifying on an earlier version of the bill, Charles N. Brower, then Legal Adviser of the Department of State, stated:
[T]he restrictive theory of sovereign immunity from jurisdiction, which has been followed by the Department of State and the courts since it was articulated in the familiar letter of Acting Legal Adviser Jack B. Tate of May 29, 1952, would be incorporated into statutory law. This theory limits immunity to public acts, leaving so-called private acts subject to suit. The proposed legislation would make it clear that immunity cannot be claimed with respect to acts or transactions that are commercial in nature, regardless of their underlying purpose.
1973 Hearings at 15.26 See generally Petrol Shipping Corp. v. Kingdom of Greece, 360 F.2d 103 (2d Cir.), cert. denied, 385 U.S. 931, 87 S.Ct. 291, 17 L.Ed.2d 213 (1966); Victory Transport, Inc. v. Comisaria General de Abastecimientos y Transportes, 336 F.2d 354 (2d Cir. 1964), cert. denied, 381 U.S. 934, [310]*31085 S.Ct. 1763, 14 L.Ed.2d 698 (1965).27 But see Isbrandtsen Tankers, Inc. v. President of India, 446 F.2d 1198 (2d Cir.) (State Dep’t suggestion of immunity supersedes application of restrictive theory), cert. denied, 404 U.S. 985, 92 S.Ct. 452, 30 L.Ed.2d 369 (1971).
Finally, current standards of international law concerning sovereign immunity add content to the “commercial activity” phrase of the FSIA. Section 1602 of the Act, entitled “Findings and declaration of purpose,” contains a cryptic reference to international law, but fails wholly to adopt it.28 The legislative history states that the Act “incorporates standards recognized under international law,” House Report at 6613, and the drafters seem to have intended rather generally to bring American sovereign immunity practice into line with that of other nations. See 1976 Hearings at 25 (testimony of Monroe Leigh), 32 (testimony of Bruno A. Ristau); 1973 Hearings at 18 (testimony of Charles N. Brower). At this point, there can be little doubt that international law follows the restrictive theory of sovereign immunity. House Report at 6613. See, e. g., State Immunity Act, 1978, s. 3 (United Kingdom); Council of Europe, European Convention on State Immunity, art. 4 (1972), reprinted in 1976 Hearings at 37, 38; Empire of Iran, 45 I.L.R. 57 (1963) (West Germany).
Under each of these three standards, Nigeria’s cement contracts and letters of credit qualify as “commercial activity.” Lord Denning, writing in Trendtex Trading Corp. v. Central Bank of Nigeria, [1977] 2 W.L.R. 356, 369, 1 All E.R. 881, with his usual erudition and clarity, stated: “If a government department goes into the market places of the world and buys boots or cement — as a commercial transaction — that government department should be subject to all the rules of the marketplace.” Nigeria’s activity here is in the nature of a private contract for the purchase of goods. Its purpose — to build roads, army barracks, whatever — is irrelevant. Accordingly, courts in other nations have uniformly held Nigeria's 1975 cement purchase program and appurtenant letters of credit to be “commercial activity,” and have denied the defense of sovereign immunity.29 We find defendants’ activity here to constitute “commercial activity,” and we move on to the next step of analysis.
C. 1.
We need look no further than the third clause of § 1605(a)(2) to find statutory subject matter jurisdiction here. That clause provides: “A foreign state shall not be immune ... in any case ... in which the [311]*311action is based ... upon an act outside the territory of the United States in connection with a commercial activity of the foreign state elsewhere and that act causes a direct effect in the United States.”
The focus of our analysis is to determine whether “the act cause[d] a direct effect in the United States” within the meaning of the FSIA.30 The “direct effect” clause has been the subject of considerable commentary31 but remains somewhat abstruse. The House Report, for example, states only that the direct effect clause “would subject [commercial] conduct [abroad] to the exercise of jurisdiction by the United States consistent with principles set forth in section 18, Restatement of the Law, Second, Foreign Relations Law of the United States (1965).” House Report at 6618. The reference is a bit of a non sequitur, since § 18 concerns the extent to which substantive American law may be applied to conduct overseas, not the proper extraterritorial jurisdictional reach of American courts n’importe quelle substantive law.32 Nor is the House Report’s vague reference to the District of Columbia’s long-arm statute, D.C.Code Ann. § 13-423(a), especially helpful; that provision looks to personal jurisdiction, not subject matter jurisdiction, and in any event is concerned in its “effects” provision only with torts. We are left with the words, “direct effect in the United States,” and with Congress’s broad mandate in passing the FSIA: “Under section 1605(a)(2), no act of a foreign state, tortious or not, which is connected with the commercial activities of a foreign state would give rise to immunity if the act takes place in the United States or has a direct effect within the United States.” 1973 Hearings at 42 (State Department Section-by-Section Analysis of 1973 bill).
Fortunately, a certain amount of case law interprets both components of the proble[312]*312matic phrase: “direct” and “in the United States.” For a paradigm of “direct,” we look to the vivid examples of Harris v. VAO Intourist, Moscow, 481 F.Supp. 1056 (E.D.N. Y.1979), and Upton v. Empire of Iran, 459 F.Supp. 264 (D.D.C.1978), aff’d mem., 607 F.2d 494 (D.C.Cir.1979). In Harris, a man lost his life in a hotel fire; in Upton, a man was injured when the roof of a building collapsed on him. Both men undoubtedly suffered “direct” effects.
Applying the term to a corporation is not so simple. Unlike a natural person, a corporate entity is intangible; it cannot be burned or crushed. It can only suffer financial loss. Accordingly, the relevant inquiry under the direct effect clause when plaintiff is a corporation is whether the corporation has suffered a “direct” financial loss. To discover whether breach of a contract causes this type of loss, we look to Carey v. National Oil Corp., 592 F.2d 673, 676-77 (2d Cir. 1979) (per curiam). In Carey, we decided that a direct effect can arise not only from a tort, e. g., Harris and Upton, supra, but from cancellation of a contract for the sale of oil as well. Here, under either theory of recovery, breach of the cement contracts or breach of the letters of credit, the effect of the suppliers was “direct.” They were beneficiaries of the contracts that were breached. See Note, Effects Jurisdiction Under the Foreign Sovereign Immunities Act and the Due Process Clause, 55 N.Y.U.L.Rev. 474, 500-10 & n.192 (1980) (“NYU Note”).
Finally, the most difficult aspect of the direct effect clause concerns its phrase, “in the United States.” State law abounds with decisions locating “effects” for personal jurisdiction purposes. But those cases are not precisely on point, for they are concerned more with federalism, and less with international relations, than was Congress in passing the FSIA.33 Reliance on state cases is not necessary here because the financial loss in these cases occurred “in the United States” for two much simpler reasons. First, the cement suppliers were to present documents and collect money in the United States, and the breaches precluded their doing so. Second, each of the plaintiffs is an American corporation. Whether a failure to pay a foreign corporation in the United States34 or to pay an American corporation overseas35 creates an effect “in the United States” under § 1605(a)(2) is not before us. Both factors are present here, and the subsection is clearly satisfied. See NYU Note at 510-13.
The foregoing analysis demonstrates that neither “direct” nor “in the United States” is a term susceptible of easy definition. A corporation is no more than a series of conduits, filtering profit — or loss— through each stage from the company’s customers to its shareholders, who may themselves be fictional entities as well. Harm to any component is somewhat “indirect,” and locating the site of the injury, especially when the harm consists in an omission, is an enterprise fraught with artifice. Courts construing either term should be mindful more of Congress’s concern with providing “access to the courts” to those aggrieved by the commercial acts of a foreign sovereign, House Report at 6605, than with cases defining “direct” or locating effects under [313]*313state statutes passed for dissimilar purposes. Before the FSIA, plaintiffs enjoyed a broad right to bring suits against foreign states, subject only to State Department intervention and the presence of attachable assets. Congress in the FSIA certainly did not intend significantly to constrict jurisdiction; it intended to regularize it. See House Report at 6605-06. The question is, was the effect sufficiently “direct” and sufficiently “in the United States” that Congress would have wanted an American court to hear the case? No rigid parsing of § 1605(a)(2) should lose sight of that purpose. We have no doubt that Congress intended to bring suits like these into American courts, see House Report at 6618, and we hold that statutory subject matter jurisdiction here exists.
2.
The final step in establishing the court’s right to hear the claim is to find a constitutional basis for the statutory exercise of subject matter jurisdiction. See Kline v. Burke Construction Co., 260 U.S. 226, 234, 43 S.Ct. 79, 82, 67 L.Ed. 226 (1922). Each of these four suits is “between a State, or the Citizens thereof, and foreign States,” U.S. Const., art. Ill, § 2, cl. 1, and therefore comes within the judicial power by way of the diversity grant. Indeed, it is to that clause that the drafters of the FSIA looked in securing a constitutional basis for FSIA suits generally. See House Report at 6611, 6632. Cf. Verlinden B. V. v. Central Bank of Nigeria, 647 F.2d 320 (2d Cir. 1981) (suit by alien against foreign state not supported by constitutional diversity grant). The district court, we conclude, had the power to hear these claims.
D. 1.
Subsequent to the determination of subject matter jurisdiction is the issue of personal jurisdiction. The statutory aspects of the analysis are quite simple. Subsection 1330(b) provides for personal jurisdiction over the foreign state as to any claim the district court has power to hear under § 1330(a), so long as service has been made under § 1608.36 Service here has been made, or at least not objected to. See Fed.R.Civ.P. 12(h)(1). Subject matter jurisdiction exists, so statutory personal jurisdiction exists as well.
Turning to the constitutional constraints on personal jurisdiction, our first inquiry must be whether the safeguards of due process, which otherwise regulate every exercise of personal jurisdiction, apply to FSIA cases at all. Specifically, is a foreign state a “person” within the meaning of the due process clause? Cases on the point are rare, since pre-FSIA suits against foreign states were generally brought quasi in rem, and the due process clause was not uniformly applied to quasi in rem suits until 1977, a year after the FSIA was passed. See Shaffer v. Heitner, 433 U.S. 186, 97 S.Ct. 2569, 53 L.Ed.2d 683 (1977). Nonetheless, in Amoco Overseas Oil Co. v. Compagnie Nation ale Algerienne de Navigation, 605 F.2d 648 (2d Cir. 1979), a quasi in rem suit filed before the FSIA and decided after Shaffer, we applied constitutional due process analysis to a suit against a foreign state. We affirm that holding today. Thos. P. Gonzalez Corp. v. Consejo Nacional de Producción de Costa Rica, 614 F.2d 1247 (9th Cir. 1980) (applying due process clause in suit against foreign state in personam); Petrol Shipping Corp. v. Kingdom of Greece, supra, 360 F.2d at 110 (same); Purdy Co. v. Argentina, 333 F.2d 95, 98 (7th Cir. 1964) (same); Rovin Sales Co. v. Socialist Republic of Romania, 403 F.Supp. 1298, 1302 (N.D.Ill.1975) (same); T.J. Stevenson & Co. v. 81,193 Bags of Wheat Flour, 399 F.Supp. 936, 938 (S.D.Ala.1975) (same) (counterclaim).
[314]*314Since the constitutional constraints apply here, our next concern must be to delineate the contacts that are relevant. The inquiry has two aspects: whose contacts, and with what? To bring any defendant before the court, of course, the due process analysis must be satisfied as to him. Nonetheless, it is not only defendant’s activities in the forum, but also actions relevant to the transaction by an agent on defendant’s behalf, which support personal jurisdiction. Gelfand v. Tanner Motor Tours, Ltd., 385 F.2d 116, 120-21 (2d Cir. 1967), cert. denied, 390 U.S. 996, 88 S.Ct. 1198, 20 L.Ed.2d 95 (1968); Frummer v. Hilton Hotels International, Inc., 19 N.Y.2d 533, 538, 281 N.Y.S.2d 41, 44-45, 227 N.E.2d 851, 854, cert. denied, 389 U.S. 923, 88 S.Ct. 241, 19 L.Ed.2d 266 (1967). Under the Gelfand standard, Central Bank’s activities with respect to the “commercial activity” are chargeable to Nigeria, and Morgan’s activities are chargeable to both. If Morgan had not performed for Central Bank, and Central Bank for Nigeria, the entire payment mechanism supporting the cement contracts, Nigeria would have been required to make the payments directly. See Gelfand, supra, 385 F.2d at 121. Since service was made under § 1608, the relevant area in delineating contacts is the entire United States, not merely New York. Compare 28 U.S.C. § 1608 (service of process provision for FSIA) with 15 U.S.C. §§ 21(f) (service of process provision for antitrust laws), 77v (same for securities laws). See Bersch v. Drexel Firestone, Inc., 519 F.2d 974, 998-1000 (2d Cir.), cert. denied, 423 U.S. 1018, 96 S.Ct. 453, 46 L.Ed.2d 389 (1975); Mariash v. Morrill, 496 F.2d 1138, 1143 (2d Cir. 1974); Leasco Data Processing Equipment Corp. v. Maxwell, 468 F.2d 1326, 3,340 (2d Cir. 1972).
Whether a defendant’s contacts with the forum are so numerous that they reach the “minimum contacts” required by International Shoe Co. v. Washington, 326 U.S. 310, 66 S.Ct. 154, 90 L.Ed. 95 (1945), depends broadly on whether “maintenance of the suit does not offend ‘traditional notions of fair play and substantial justice.’ ” Id. at 316, 66 S.Ct. at 158, citing Milliken v. Meyer, 311 U.S. 457, 463, 61 S.Ct. 339, 343, 85 L.Ed. 278 (1940). That standard, in turn, involves at least four separate inquiries. Under the cases from International Shoe to World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 100 S.Ct. 559, 62 L.Ed.2d 490 (1980), the court must examine the extent to which defendants availed themselves of the privileges of American law, the extent to which litigation in the United States would be foreseeable to them, the inconvenience to defendants of litigating in the United States, and the countervailing interest of the United States in hearing the suit. See World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S. at 292, 297, 100 S.Ct. at 564, 567; Kulko v. California Superior Court, 436 U.S. 84, 97-98, 98 S.Ct. 1690, 1699-1700, 56 L.Ed.2d 132 (1978); Hanson v. Denckla, 357 U.S. 235, 253, 78 S.Ct. 1228, 1239, 2 L.Ed.2d 1283 (1958); McGee v. International Life Insurance Co., 355 U.S. 220, 223, 78 S.Ct. 199, 201, 2 L.Ed.2d 223 (1957); International Shoe Co., supra, 326 U.S. at 316, 66 S.Ct. at 158.
The facts of these cases establish that Central Bank, and through Central Bank Nigeria, repeatedly and “purposefully avail[ed themselves]” of the privilege of conducting activities in the United States. Hanson v. Denckla, supra, 357 U.S. at 253, 78 S.Ct. at 1240, citing International Shoe Co., supra, 326 U.S. at 319, 66 S.Ct. at 159. (Central Bank alone sent its employees to New York for training, kept large cash balances here, and maintained a custody : account as well. If Morgan had converted ' Central Bank’s funds from either account, New York law would have protected Central Bank, and allowed it to sue. Central Bank made it a regular practice to advise letters of credit through Morgan, and to use Morgan as its means of paying bills throughout the world. New York law protected Central Bank in each of its instructions, transfers, and withdrawals. Central Bank’s activities with respect to the cement contracts and the letters of credit, directly chargeable to Nigeria under Gelfand, show the same pattern. In Nigeria’s behalf and on Nigeria’s instructions, Central Bank advised each of the letters of credit through [315]*315Morgan, in the United States, regardless of the individual supplier’s wishes. Having chosen American law and process as their protectors, Nigeria and Central Bank were not hesitant to invoke them; at the mere hint Morgan was reluctant to honor defendants’ amendments to the letters of credit, an officer of Central Bank threatened to “go to court” to enforce them.
Having so thoroughly “invok[ed] the benefits and protections of [American] laws,” Hanson v. Denekla, supra, 357 U.S. at 253, 78 S.Ct. at 1240, Nigeria and Central Bank would have every “reason to expect to be haled before a ... court” here. Shaffer v. Heitner, supra, 433 U.S. at 216, 97 S.Ct. at 2586. Having threatened litigation on their own, and having been notified of its likelihood by Morgan, defendants cannot now assert they could not have expected it. Moreover, the very function of the due process clause is to give “a degree of predictability to the legal system that allows potential defendants to structure their primary conduct with some minimum assurance as to where that conduct will and will not render them liable to suit.” World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S. at 297, 100 S.Ct. at 567. In light of defendants’ intentional activities in the United States, litigation was clearly foreseeable here.
Although the United States is certainly distant from Nigeria, litigation here is not unduly inconvenient for defendants. Every modern transnational commercial contract presents problems of adjudicatory cost; in the cement purchase program, where Nigeria bargained with corporations from a plethora of nations, required in the contracts that the goods came mostly from Europe, and provided in some letters of credit that payment was to be made in the United States, the inconvenience was at least expected. Moreover, Nigeria in the cement contracts agreed to submit to arbitration by the International Chamber of Commerce (“ICC”). The ICC’s headquarters are in Paris, but its arbitrations can take place anywhere in the world. See ICC, ICC Arbitration — The International Solution to Business Disputes, art. 12 (undated). Further, any assertion of inconvenience here is belied by defendants’ constant resort to the United States for a myriad of services, and by the frequent visits of Central Bank officials to Morgan in New York. Cf. World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S. at 297, 100 S.Ct. at 567.
In McGee v. International Life Insurance Co., the Supreme Court depended in part on the forum’s “manifest interest in providing effective means of redress for its residents ... . ” 355 U.S. at 223, 78 S.Ct. at 201. Here, we should not be unmindful that Congress has passed the FSIA specifically to provide “access to the courts.” House Report at 6605. Similarly, the plaintiff has an “interest in obtaining convenient and effective relief.” World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S. at 292,100 S.Ct. at 564, citing Kulko v. California Superior Court, supra, 436 U.S. at 92, 98 S.Ct. at 1696. See generally Plant Food Co-op v. Wolfkill Feed & Fertilizer Corp., 633 F.2d 155, 158-60 (9th Cir. 1980); Pedi Bares, Inc. v. P & C Food Markets, Inc., 567 F.2d 933, 936-38 (10th Cir. 1977); Product Promotions, Inc. v. Cousteau, 495 F.2d 483, 494-98 (5th Cir. 1974). Accordingly, we hold that defendants’ relation to the forum here satisfies the “minimum contacts” requirement of International Shoe Sterling National Bank & Trust Co. of New York v. Fidelity Mortgage Investors, 510 F.2d 870 (2d Cir. 1975).37
III.
Our rulings today vindicate more than Congressional intent. They affirm the [316]*316right of all participants in the marketplace of the world to be treated as equals, and to ascribe to principles of trade which found their birth in the law merchant, centuries ago. Corporations can enter contracts without fear that the defense of sovereign immunity will be inequitably interposed, and foreign states can bargain without paying a premium required by a trader in anticipation of a judgment-proof client. Commerce is fostered, and all interests are advanced.
In Nikkei, East Europe, and Chenax, the district court held jurisdiction to be present, and proceeded to the trial of plaintiffs’ claims. We find no error in its rulings on the merits, and affirm those three judgments in full.38 In Texas Trading, the district court ordered the complaint dismissed for lack of jurisdiction. That order is reversed, and the case is remanded for proceedings consistent with this opinion.