delivered the opinion of the Court with respect to Part I, and delivered a dissenting opinion with respect to Part II.*

With respect to the petition in No. 91-1111, I join the Court’s judgment and Parts I and II-A of its opinion. I write separately because I do not agree with Justice Souter’s analysis, set forth in Part II-B of his opinion, of what constitutes a “boycott” for purposes of §3(b) of the McCarran-Ferguson Act, 15 U. S. C. § 1013(b). With respect to the petition in No. 91-1128, I dissent from the Court’s ruling concerning the extraterritorial application of the Sherman Act. Part I below discusses the boycott issue; Part II extraterritoriality.

I

Determining proper application of § 3(b) of the McCarranFerguson Act to the present cases requires precise definition of the word “boycott.”1 It is a relatively new word, little more than a century old. It was first used in 1880, to describe the collective action taken against Captain Charles Boycott, an English agent managing various estates in Ireland. The Land League, an Irish organization formed the previous year, had demanded that landlords reduce their rents and had urged tenants to avoid dealing with those who failed to do so. Boycott did not bend to the demand and instead ordered evictions. In retaliation, the tenants “sen[t] Captain Boycott to Coventry in a very thorough manner.” J. McCarthy, England Under Gladstone 108 (1886). “The population of the region for miles round resolved not to have anything to do with him, and, as far as they could prevent *803it, not to allow any one else to have anything to do with him. . . . [T]he awful sentence of excommunication could hardly have rendered him more helplessly alone for a time. No one would work for him; no one would supply him with food.” Id., at 108-109; see also H. Laidler, Boycotts and the Labor Struggle 23-27 (1968). Thus, the verb made from the unfortunate Captain’s name has had from the outset the meaning it continues to carry today. To “boycott” means “[tjo combine in refusing to hold relations of any kind, social or commercial, public or private, with (a neighbour), on account of political or other differences, so as to punish him for the position he has taken up, or coerce him into abandoning it.” 2 Oxford English Dictionary 468 (2d ed. 1989).

Petitioners have suggested that a boycott ordinarily requires “an absolute refusal to deal on any terms,” which was concededly not the case here. Brief for Petitioners in No. 91-1111, p. 31; see also Reply Brief for Petitioners in No. 91-1111, pp. 12-13. We think not. As the definition just recited provides, the refusal may be imposed “to punish [the target] for the position he has taken up, or coerce him into abandoning it.” The refusal to deal may, in other words, be conditional, offering its target the incentive of renewed dealing if and when he mends his ways. This is often the case — and indeed seems to have been the case with the original Boycott boycott. Cf. McCarthy, supra, at 109 (noting that the Captain later lived “at peace” with his neighbors). Furthermore, other dictionary definitions extend the term to include a partial boycott — a refusal to engage in some, but not all, transactions with the target. See Webster’s New International Dictionary 321 (2d ed. 1950) (defining “boycott” as “to withhold, wholly or in part, social or business intercourse from, as an expression of disapproval or means of coercion” (emphasis added)).

It is, however, important — and crucial in the present cases — to distinguish between a conditional boycott and a concerted agreement to seek particular terms in particular *804transactions. A concerted agreement to terms (a “cartelization”) is “a way of obtaining and exercising market power by concertedly exacting terms like those which a monopolist might exact.” L. Sullivan, Law of Antitrust 257 (1977). The parties to such an agreement (the members of a cartel) are not engaging in a boycott, because:

“They are not coercing anyone, at least in the usual sense of that word; they are merely (though concertedly) saying ‘we will deal with you only on the following trade terms.’

“. . . Indeed, if a concerted agreement, say, to include a security deposit in all contracts is a ‘boycott’ because it excludes all buyers who won’t agree to it, then by parity of reasoning every price fixing agreement would be a boycott also. The use of the single concept, boycott, to cover agreements so varied in nature can only add to confusion.” Ibid, (emphasis added).

Thus, if Captain Boycott’s tenants had agreed among themselves that they would refuse to renew their leases unless he reduced his rents, that would have been a concerted agreement on the terms of the leases, but not a boycott.2 The tenants, of course, did more than that; they refused to engage in other, unrelated transactions with Boycott — e. g., selling him food — unless he agreed to their terms on rents. It is *805this expansion of the refusal to deal beyond the targeted transaction that gives great coercive force to a commercial boycott: unrelated transactions are used as leverage to achieve the terms desired.

The proper definition of “boycott” is evident from the Court’s opinion in Eastern States Retail Lumber Dealers’ Assn. v. United States, 234 U. S. 600 (1914), which is recognized in the antitrust field as one of the “leading case[s] involving commercial boycotts.” Barber, Refusals to Deal under the Federal Antitrust Laws, 103 U. Pa. L. Rev. 847, 873 (1955). The associations of retail lumber dealers in that ease refused to buy lumber from wholesale lumber dealers who sold directly to consumers. The boycott attempted “to impose as a condition ... on [the wholesale dealers’] trade that they shall not sell in such manner that a local retailer may regard such sale as an infringement of his exclusive right to trade.” 234 U. S., at 611. We held that to be an “‘artificial conditio[n],’” since “the trade of the wholesaler with strangers was directly affected, not because of any supposed wrong which he had done to them, but because of the grievance of a member of one of the associations.” Id., at 611-612. In other words, the associations’ activities were a boycott because they sought an objective — the wholesale dealers’ forbearance from retail trade — that was collateral to their transactions with the wholesalers.

Of course as far as the Sherman Act (outside the exempted insurance field) is concerned, concerted agreements on contract terms are as unlawful as boycotts. For example, in Paramount Famous Lasky Corp. v. United States, 282 U. S. 30 (1930), and United States v. First Nat. Pictures, Inc., 282 U. S. 44 (1930), we held unreasonable an agreement among competing motion picture distributors under which they refused to license films to exhibitors except on standardized terms. We also found unreasonable the restraint of trade in Anderson v. Shipowners Assn. of Pacific Coast, 272 U. S. 359 (1926), which involved an attempt by an association of *806employers to establish industry-wide terms of employment. These sorts of concerted actions, similar to what is alleged to have occurred here, are not properly characterized as “boycotts,” and the word does not appear in the opinions.3 In fact, in the 65 years between the coining of the word and enactment of the McCarran-Ferguson Act in 1945, “boycott” appears in only seven opinions of this Court involving commercial (nonlabor) antitrust matters, and not once is it used as Justice Souter uses it — to describe a concerted refusal to engage in particular transactions until the terms of those transactions are agreeable.4

In addition to its use in the antitrust field, the concept of “boycott” frequently appears in labor law, and in this context as well there is a clear distinction between boycotts and concerted agreements seeking terms. The ordinary strike *807seeking better contract terms is a “refusal to deal” — i. e., union members refuse to sell their labor until the employer capitulates to their contract demands. But no one would call this a boycott, because the conditions of the “refusal to deal” relate directly to the terms of the refused transaction (the employment contract). A refusal to work changes from strike to boycott only when it seeks to obtain action from the employer unrelated to the employment contract. This distinction is well illustrated by the famous boycott of Pullman cars by Eugene Debs’ American Railway Union in 1894. The incident began when workers at the Pullman Palace Car Company called a strike, but the “boycott” occurred only when other members of the American Railway Union, not Pullman employees, supported the strikers by refusing to work on any train drawing a Pullman car. See In re Debs, 158 U. S. 564, 566-567 (1895) (statement of the case); H. Laidler, Boycotts and the Labor Struggle 100-108 (1968). The refusal to handle Pullman cars had nothing to do with Pullman cars themselves (working on Pullman cars was no more difficult or dangerous than working on other cars); rather, it was in furtherance of the collateral objective of obtaining better employment terms for the Pullman workers. In other labor cases as well, the term “boycott” invariably holds the meaning that we ascribe to it: Its goal is to alter, not the terms of the refused transaction, but the terms of workers’ employment.5

*808The one case in which we have found an activity to constitute a “boycott” within the meaning of the McCarranFerguson Act is St. Paul Fire & Marine Ins. Co. v. Barry, 438 U. S. 531 (1978). There the plaintiffs were licensed physicians and their patients, and the defendant (St. Paul) was a malpractice insurer that had refused to renew the physicians’ policies on an “occurrence” basis, but insisted upon a “claims made” basis. The allegation was that, at the instance of St. Paul, the three other malpractice insurers in the State had collectively refused to write insurance for St. Paul’s customers, thus forcing them to accept St. Paul’s renewal terms. Unsurprisingly, we held the allegation sufficient to state a cause of action. The insisted-upon condition of the boycott (not being a former St. Paul policyholder) was “artificial”: it bore no relationship (or an “artificial” relationship) to the proposed contracts of insurance that the physicians wished to conclude with St. Paul’s competitors.

Under the standard described, it is obviously not a “boycott” for the reinsurers to “refus[e] to reinsure coverages written on the ISO CGL forms until the desired changes were made,” ante, at 788, because the terms of the primary coverages are central elements of the reinsurance contract— they are what is reinsured. See App. 16-17 (Cal. Complaint ¶¶ 26-27). The “primary policies are . . . the basis of the losses that are shared in the reinsurance agreements.” 1 B. Webb, H. Anderson, J. Cookman, & P. Kensicki, Principles of Reinsurance 87 (1990); see also id., at 55; Gurley, Regulation of Reinsurance in the United States, 19 Forum 72, 73 (1983). Indeed, reinsurance is so closely tied to the terms of the primary insurance contract that one of the two categories of reinsurance (assumption reinsurance) substitutes the re-insurer for the primary or “ceding” insurer and places the reinsurer into contractual privity with the primary insurer’s policyholders. See id., at 73-74; Colonial American Life Ins. Co. v. Commissioner, 491 U. S. 244, 247 (1989); B. Ostrager & T. Newman, Handbook on Insurance Coverage *809Disputes chs. 15-16 (5th ed. 1992). And in the other category of reinsurance (indemnity reinsurance), either the terms of the underlying insurance policy are incorporated by reference (if the reinsurance is written under a facultative agreement), see J. Butler & R. Merkin, Reinsurance Law B.l.1-04 (1992); R. Carter, Reinsurance 235 (1979), or (if the reinsurance is conducted on a treaty basis) the reinsurer will require full disclosure of the terms of the underlying insurance policies and usually require that the primary insurer not vary those terms without prior approval, see id., at 256, 297.

Justice Souter simply disregards this integral relationship between the terms of the primary insurance form and the contract of reinsurance. He describes the reinsurers as “individuals and entities who were not members of ISO, and who would not ordinarily be parties to an agreement setting the terms of primary insurance, not being in the business of selling it.” Ante, at 788. While this factual assumption is crucial to Justice Souter’s reasoning (because otherwise he would not be able to distinguish permissible agreements among primary insurers), he offers no support for the statement. But even if it happens to be true, he does not explain why it must be true — that is, why the law must exclude reinsurers from full membership and participation. The realities of the industry may make explanation difficult:

“Reinsurers also benefit from the services by ISO and other rating or service organizations. The underlying rates and policy forms are the basis for many reinsurance contracts. Reinsurers may also subscribe to various services. For example, a facultative reinsurer may subscribe to the rating service, so that they have the rating manuals available, or purchase optional services, such as a sprinkler report for a specific property location.” 2 R. Reinarz, J. Schloss, G. Patrik, & P. Kensicki, Reinsurance Practices 18 (1990).

*810Justice Souter also describes reinsurers as being “outside the primary insurance industry.” Ante, at 789. That is technically true (to the extent the two symbiotic industries can be separated) but quite irrelevant. What matters is that the scope and predictability of the risks assumed in a reinsurance contract depend entirely upon the terms of the primary policies that are reinsured. The terms of the primary policies are the “subject-matter insured” by reinsurance, Carter, supra, at 4, so that to insist upon certain primary-insurance terms as a condition of writing reinsurance is in no way “artificial”; and hence for a number of reinsurers to insist upon such terms jointly is in no way a “boycott.”6

Justice Souter seems to believe that a nonboycott is converted into a boycott by the fact that it occurs “at the behest of,” ante, at 789, or is “solicited” by, ibid., competitors of the target. He purports to find support for this implausible proposition in United States v. South-Eastern Underwriters Assn., 322 U. S. 533 (1944), which involved a classic boycott, by primary insurers, of competitors who refused to join their price-fixing conspiracy, the South-Eastern Underwriters Association (S. E. U. A.). The conspirators would not deal with independent agents who wrote for such companies, and would not write policies for customers who insured with them. See id., at 535-536. Moreover, Justice Black’s opinion for the Court noted cryptically, “Companies not members of S. E. U. A. were cut off from the opportunity to rein-sure their risks.” Id., at 535. Justice Souter speculates *811that “the [S. E. U. A.] defendants could have [managed to cut the targets off from reinsurance] by prompting reinsurance companies to refuse to deal with nonmembers.” Ante, at 789. Even assuming that is what happened, all that can be derived from S. E. U. A. is the proposition that one who prompts a boycott is a co-conspirator with the boycotters. For with or without the defendants’ prompting, the reinsurers’ refusal to deal in S. E. U. A. was a boycott, membership in the association having no discernible bearing upon the terms of the refused reinsurance contracts.

Justice Souter suggests that we have somehow mistakenly “posit[ed] . . . autonomy on the part of the reinsurers.” Ante, at 792. We do not understand this. Nothing in the complaints alleges that the reinsurers were deprived of their “autonomy,” which we take to mean that they were coerced by the primary insurers. (Given the sheer size of the Lloyd’s market, such an allegation would be laughable.) That is not to say that we disagree with Justice Souter’s contention that, according to the allegations, the reinsurers would not “have taken exactly the same course of action without the intense efforts of the four primary insurers.” Ibid. But the same could be said of the participants in virtually all conspiracies: If they had not been enlisted by the “intense efforts” of the leaders, their actions would not have been the same. If this factor renders otherwise lawful conspiracies (under McCarran-Ferguson) illegal, then the Act would have a narrow scope indeed.

Perhaps Justice Souter feels that it is undesirable, as a policy matter, to allow insurers to “prompt” reinsurers not to deal with the insurers’ competitors — whether or not that refusal to deal is a boycott. That feeling is certainly understandable, since under the normal application of the Sherman Act the reinsurers’ concerted refusal to deal would be an unlawful conspiracy, and the insurers’ “prompting” could make them part of that conspiracy. The McCarran*812Ferguson Act, however, makes that conspiracy lawful (assuming reinsurance is state regulated), unless the refusal to deal is a “boycott.”

Under the test set forth above, there are sufficient allegations of a “boycott” to sustain the relevant counts of complaint against a motion to dismiss. For example, the complaints allege that some of the defendant reinsurers threatened to “withdra[w] entirely from the business of reinsuring primary U. S. insurers who wrote on the occurrence form.” App. 31 (Cal. Complaint ¶ 89), id., at 83 (Conn. Complaint ¶ 93). Construed most favorably to respondents, that allegation claims that primary insurers who wrote insurance on disfavored forms would be refused all reinsurance, even as to risks written on other forms. If that were the case, the reinsurers might have been engaging in a boycott — they would, that is, unless the primary insurers’ other business were relevant to the proposed reinsurance contract (for example, if the reinsurer bears greater risk where the primary insurer engages in riskier businesses). Cf. Gonye, Underwriting the Reinsured, in Reinsurance 439, 463-466 (R. Strain ed. 1980); 2 R. Reinarz, J. Schloss, G. Patrik, & P. Kensicki, Reinsurance Practices 21-23 (1990) (same). Other allegations in the complaints could be similarly construed. For example, the complaints also allege that the reinsurers “threatened a boycott of North American CGL risks,” not just CGL risks containing dissatisfactory terms, App. 26 (Cal. Complaint ¶ 74), id., at 79 (Conn. Complaint ¶ 78); that “the foreign and domestic reinsurer representatives presented their agreed upon positions that there would be changes in the CGL forms or no reinsurance,” id., at 29 (Cal. Complaint ¶82), id., at 81-82 (Conn. Complaint ¶ 86); that some of the defendant insurers and reinsurers told “groups of insurance brokers and agents .. . that a reinsurance boycott, and thus loss of income to the agents and brokers who would be unable to find available markets for their customers, would ensue if the [revised] ISO forms were not ap*813proved,” id., at 29 (Cal. Complaint ¶ 85), id., at 82 (Conn. Complaint ¶ 89).

Many other allegations in the complaints describe conduct that may amount to a boycott if the plaintiffs can prove certain additional facts. For example, General Re, the largest American reinsurer, is alleged to have “agreed to either coerce ISO to adopt [the defendants’] demands or, failing that, ‘derail’ the entire CGL forms program.” Id., at 24 (Cal. Complaint ¶ 64), id., at 77 (Conn. Complaint ¶ 68). If this means that General Re intended to withhold all reinsurance on all CGL forms — even forms having no objectionable terms — that might amount to a “boycott.” Also, General Re and several other domestic reinsurers are alleged to have “agreed to boycott the 1984 ISO forms unless a retroactive date was added to the claims-made form, and a pollution exclusion and a defense cost cap were added to both [the occurrence and claims made] forms.” Id., at 25 (Cal. Complaint ¶ 66), id., at 78 (Conn. Complaint ¶ 70). Liberally construed, this allegation may mean that the defendants had linked their demands so that they would continue to refuse to do business on either form until both were changed to their liking. Again, that might amount to a boycott. “[A] complaint should not be dismissed unless ‘it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief.’ ” McLain v. Real Estate Bd. of New Orleans, Inc., 444 U. S. 232, 246 (1980) (quoting Conley v. Gibson, 355 U. S. 41, 45-46 (1957)). Under that standard, these allegations are sufficient to sustain the First, Second, Third, and Fourth Claims for Relief in the California Complaint and the First and Second Claims for Relief in the Connecticut Complaint.7

*814II

Petitioners in No. 91-1128, various British corporations and other British subjects, argue that certain of the claims against them constitute an inappropriate extraterritorial application of the Sherman Act.8 It is important to distinguish two distinct questions raised by this petition: whether the District Court had jurisdiction, and whether the Sherman Act reaches the extraterritorial conduct alleged here. On the first question, I believe that the District Court had subject-matter jurisdiction over the Sherman Act claims against all the defendants (personal jurisdiction is not contested). Respondents asserted nonfrivolous claims under the Sherman Act, and 28 U. S. C. § 1331 vests district courts with subject-matter jurisdiction over cases “arising under” federal statutes. As precedents such as Lauritzen v. Larsen, 345 U. S. 571 (1953), make clear, that is sufficient to establish the District Court’s jurisdiction over these claims. Lauritzen involved a Jones Act claim brought by a foreign sailor against a foreign shipowner. The shipowner contested the District Court’s jurisdiction, see id., at 573, apparently on the grounds that the Jones Act did not govern the dispute between the foreign parties to the action. Though ultimately agreeing with the shipowner that the Jones Act did not apply, see discussion infra, at 816, the Court held that the District Court had jurisdiction.

“As frequently happens, a contention that there is some barrier to granting plaintiff’s claim is cast in terms of an exception to jurisdiction of subject matter. A cause of action under our law was asserted here, and the court had power to determine whether it was or was not well founded in law and in fact.” 345 U. S., at 575.

*815See also Romero v. International Terminal Operating Co., 358 U. S. 354, 359 (1959).

The second question — the extraterritorial reach of the Sherman Act — has nothing to do with the jurisdiction of the courts. It is a question of substantive law turning on whether, in enacting the Sherman Act, Congress asserted regulatory power over the challenged conduct. See EEOC v. Arabian American Oil Co., 499 U. S. 244, 248 (1991) (Aramco) (“It is our task to determine whether Congress intended the protections of Title VII to apply to United States citizens employed by American employers outside of the United States”). If a plaintiff fails to prevail on this issue, the court does not dismiss the claim for want of subject-matter jurisdiction — want of power to adjudicate; rather, it decides the claim, ruling on the merits that the plaintiff has failed to state a cause of action under the relevant statute. See Romero, supra, at 384 (holding no claim available under the Jones Act); American Banana Co. v. United Fruit Co., 213 U. S. 347, 359 (1909) (holding that complaint based upon foreign conduct “alleges no case under the [Sherman Act]”).

There is, however, a type of “jurisdiction” relevant to determining the extraterritorial reach of a statute; it is known as “legislative jurisdiction,” Aramco, supra, at 253; Restatement (First) Conflict of Laws § 60 (1934), or “jurisdiction to prescribe,” 1 Restatement (Third) of Foreign Relations Law of the United States 235 (1987) (hereinafter Restatement (Third)). This refers to “the authority of a state to make its law applicable to persons or activities,” and is quite a separate matter from “jurisdiction to adjudicate,” see id., at 231. There is no doubt, of course, that Congress possesses legislative jurisdiction over the acts alleged in this complaint: Congress has broad power under Article I, § 8, cl. 3, “[t]o regulate Commerce with foreign Nations,” and this Court has repeatedly upheld its power to make laws applicable to persons or activities beyond our territorial boundaries where United *816States interests are affected. See Ford v. United States, 273 U. S. 593, 621-623 (1927); United States v. Bowman, 260 U. S. 94, 98-99 (1922); American Banana, supra, at 356. But the question in this litigation is whether, and to what extent, Congress has exercised that undoubted legislative jurisdiction in enacting the Sherman Act.

Two canons of statutory construction are relevant in this inquiry. The first is the “longstanding principle of American law ‘that legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States.’ ” Aramco, supra, at 248 (quoting Foley Bros., Inc. v. Filardo, 336 U. S. 281, 285 (1949)). Applying that canon in Aramco, we held that the version of Title VII of the Civil Rights Act of 1964 then in force, 42 U. S. C. §§2000e to 2000e-17 (1988 ed.), did not extend outside the territory of the United States even though the statute contained broad provisions extending its prohibitions to, for example, “‘any activity, business, or industry in commerce.’” Id., at 249 (quoting 42 U. S. C. §2000e(h)). We held such “boilerplate language” to be an insufficient indication to override the presumption against extraterritoriality. Id., at 251; see also id., at 251-253. The Sherman Act contains similar “boilerplate language,” and if the question were not governed by precedent, it would be worth considering whether that presumption controls the outcome here. We have, however, found the presumption to be overcome with respect to our antitrust laws; it is now well established that the Sherman Act applies extraterritorially. See Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 U. S. 574, 582, n. 6 (1986); Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U. S. 690, 704 (1962); see also United States v. Aluminum Co. of America, 148 P. 2d 416 (CA2 1945).

But if the presumption against extraterritoriality has been overcome or is otherwise inapplicable, a second canon of statutory construction becomes relevant: “[A]n act of congress *817ought never to be construed to violate the law of nations if any other possible construction remains.” Murray v. Schooner Charming Betsy, 2 Cranch 64, 118 (1804) (Marshall, C. J.). This canon is “wholly independent” of the presumption against extraterritoriality. Aramco, supra, at 264 (Marshall, J., dissenting). It is relevant to determining the substantive reach of a statute because “the law of nations,” or customary international law, includes limitations on a nation’s exercise of its jurisdiction to prescribe. See Restatement (Third) §§401-416. Though it clearly has constitutional authority to do so, Congress is generally presumed not to have exceeded those customary international-law limits on jurisdiction to prescribe.

Consistent with that presumption, this and other courts have frequently recognized that, even where the presumption against extraterritoriality does not apply, statutes should not be interpreted to regulate foreign persons or conduct if that regulation would conflict with principles of international law. For example, in Romero v. International Terminal Operating Co., 358 U. S. 354 (1959), the plaintiff, a Spanish sailor who had been injured while working aboard a Spanish-flag and Spanish-owned vessel, filed a Jones Act claim against his Spanish employer. The presumption against extraterritorial application of federal statutes was inapplicable to the case, as the actionable tort had occurred in American waters. See id,., at 383. The Court nonetheless stated that, “in the absence of a contrary congressional direction,” it would apply “principles of choice of law that are consonant with the needs of a general federal maritime law and with due recognition of our self-regarding respect for the relevant interests of foreign nations in the regulation of maritime commerce as part of the legitimate concern of the international community.” Id., at 382-383. “The controlling considerations” in this choice-of-law analysis were “the interacting interests of the United States and of foreign countries.” Id., at 383.

*818Romero referred to, and followed, the choice-of-law analysis set forth in Lauritzen v. Larsen, 345 U. S. 571 (1953). As previously mentioned, Lauritzen also involved a Jones Act claim brought by a foreign sailor against a foreign employer. The Lauritzen Court recognized the basic problem: “If [the Jones Act were] read literally, Congress has conferred an American right of action which requires nothing more than that plaintiff be ‘any seaman who shall suffer personal injury in the course of his employment.’” Id., at 576. The solution it adopted was to construe the statute “to apply only to areas and transactions in which American law would be considered operative under prevalent doctrines of international law.” Id., at 577 (emphasis added). To support application of international law to limit the facial breadth of the statute, the Court relied upon — of course — Chief Justice Marshall’s statement in Schooner Charming Betsy, quoted supra, at 814-815. It then set forth “several factors which, alone or in combination, are generally conceded to influence choice of law to govern a tort claim.” 345 U. S., at 583; see id., at 583-593 (discussing factors). See also McCulloch v. Sociedad Nacional de Marineros de Honduras, 372 U. S. 10, 21-22 (1963) (applying Schooner Charming Betsy principle to restrict application of National Labor Relations Act to foreign-flag vessels).

Lauritzen, Romero, and McCulloch were maritime cases, but we have recognized the principle that the scope of generally worded statutes must be construed in light of international law in other areas as well. See, e. g., Sale v. Haitian Centers Council, Inc., ante, at 178, n. 35; Weinberger v. Rossi, 456 U. S. 25, 32 (1982). More specifically, the principle was expressed in United States v. Aluminum Co. of America, 148 F. 2d 416 (CA2 1945), the decision that established the extraterritorial reach of the Sherman Act. In his opinion for the court, Judge Learned Hand cautioned “we are not to read general words, such as those in [the Sherman] *819Act, without regard to the limitations customarily observed by nations upon the exercise of their powers; limitations which generally correspond to those fixed by the ‘Conflict of Laws.’ ” Id., at 443.

More recent lower court precedent has also tempered the extraterritorial application of the Sherman Act with considerations of “international comity.” See Timberlane Lumber Co. v. Bank of America, N. T. & S. A., 549 F. 2d 597, 608-615 (CA9 1976); Mannington Mills, Inc. v. Congoleum Corp., 595 F. 2d 1287, 1294-1298 (CA3 1979); Montreal Trading Ltd. v. Amax Inc., 661 F. 2d 864, 869-871 (CA10 1981); Laker Airways Limited v. Sabena, Belgian World Airlines, 235 U. S. App. D. C. 207, 236, and n. 109, 731 F. 2d 909, 938, and n. 109 (1984); see also Pacific Seafarers, Inc. v. Pacific Far East Line, Inc., 131 U. S. App. D. C. 226, 236, and n. 31, 404 F. 2d 804, 814, and n. 31 (1968). The “comity” they refer to is not the comity of courts, whereby judges decline to exercise jurisdiction over matters more appropriately adjudged elsewhere, but rather what might be termed “prescriptive comity”: the respect sovereign nations afford each other by limiting the reach of their laws. That comity is exercised by legislatures when they enact laws, and courts assume it has been exercised when they come to interpreting the scope of laws their legislatures have enacted. It is a traditional component of choice-of-law theory. See J. Story, Commentaries on the Conflict of Laws § 38 (1834) (distinguishing between the “comity of the courts” and the “comity of nations,” and defining the latter as “the true foundation and extent of the obligation of the laws of one nation within the territories of another”). Comity in this sense includes the choice-of-law principles that, “in the absence of contrary congressional direction,” are assumed to be incorporated into our substantive laws having extraterritorial reach. Romero, supra, at 382-383; see also Lauritzen, supra, at 578-579; Hilton v. Guyot, 159 U. S. 113, 162-166 (1895). Considering comity in *820this way is just part of determining whether the Sherman Act prohibits the conduct at issue.9

In sum, the practice of using international law to limit the extraterritorial reach of statutes is firmly established in our jurisprudence. In proceeding to apply that practice to the present cases, I shall rely on the Restatement (Third) for the relevant principles of international law. Its standards appear fairly supported in the decisions of this Court construing international choice-of-law principles (Lauritzen, Romero, and McCulloch) and in the decisions of other federal courts, especially Timberlane. Whether the Restatement precisely reflects international law in every detail matters little here, as I believe this litigation would be resolved the same way under virtually any conceivable test that takes account of foreign regulatory interests.

Under the Restatement, a nation having some “basis” for jurisdiction to prescribe law should nonetheless refrain from exercising that jurisdiction “with respect to a person or activity having connections with another state when the exercise of such jurisdiction is unreasonable.” Restatement (Third) §403(1). The “reasonableness” inquiry turns on a number of factors including, but not limited to: “the extent to which the activity takes place within the territory [of the regulating state],” id., §403(2)(a); “the connections, such as nationality, residence, or economic activity, between the regulating state and the person principally responsible for the *821activity to be regulated,” id., §403(2)(b); “the character of the activity to be regulated, the importance of regulation to the regulating state, the extent to which other states regulate such activities, and the degree to which the desirability of such regulation is generally accepted,” id., § 403(2)(c); “the extent to which another state may have an interest in regulating the activity,” id., §403(2)(g); and “the likelihood of conflict with regulation by another state,” id., §403(2)(h). Rarely would these factors point more clearly against application of United States law. The activity relevant to the counts at issue here took place primarily in the United Kingdom, and the defendants in these counts are British corporations and British subjects having their principal place of business or residence outside the United States.10 Great Britain has established a comprehensive regulatory scheme governing the London reinsurance markets, and clearly has a heavy “interest in regulating the activity,” id., § 403(2)(g). See 938 F. 2d, at 932-933; In re Insurance Antitrust Litigation, 723 F. Supp. 464, 487-488 (ND Cal. 1989); see also J. Butler & R. Merkin, Reinsurance Law A.1.1-02 (1992). Finally, § 2(b) of the McCarran-Ferguson Act allows state regulatory statutes to override the Sherman Act in the insurance field, subject only to the narrow “boycott” exception set forth in §3(b) — suggesting that “the importance of regulation to the [United States],” Restatement (Third) §403(2)(c), is slight. Considering these factors, I think it unimaginable that an assertion of legislative jurisdiction by the United States would be considered reasonable, and therefore it is inappropriate to assume, in the absence of statutory indication to the contrary, that Congress has made such an assertion.

*822It is evident from what I have said that the Court’s comity analysis, which proceeds as though the issue is whether the courts should “decline to exercise . . . jurisdiction,” ante, at 798, rather than whether the Sherman Act covers this conduct, is simply misdirected. I do not at all agree, moreover, with the Court’s conclusion that the issue of the substantive scope of the Sherman Act is not in the cases. See ante, at 796, n. 22; ante, at 797, n. 24. To be sure, the parties did not make a clear distinction between adjudicative jurisdiction and the scope of the statute. Parties often do not, as we have observed (and have declined to punish with procedural default) before. See the excerpt from Lauritzen quoted supra, at 812; see also Romero, 358 U. S., at 359. It is not realistic, and also not helpful, to pretend that the only really relevant issue in this litigation is not before us. In any event, if one erroneously chooses, as the Court does, to make adjudicative jurisdiction (or, more precisely, abstention) the vehicle for taking account of the needs of prescriptive comity, the Court still gets it wrong. It concludes that no “true conflict” counseling nonapplication of United States law (or rather, as it thinks, United States judicial jurisdiction) exists unless compliance with United States law would constitute a violation of another country’s law. Ante, at 798-799. That breathtakingly broad proposition, which contradicts the many cases discussed earlier, will bring the Sherman Act and other laws into sharp and unnecessary conflict with the legitimate interests of other countries — particularly our closest trading partners.

In the sense in which the term “conflictt]” was used in Lauritzen, 345 U. S., at 582, 592, and is generally understood in the field of conflicts of laws, there is clearly a conflict in this litigation. The petitioners here, like the defendant in Lauritzen, were not compelled by any foreign law to take their allegedly wrongful actions, but that no more precludes a conflict-of-laws analysis here than it did there. See id., at 575-576 (detailing the differences betwéen foreign and *823United States law). Where applicable foreign and domestic law provide different substantive rules of decision to govern the parties’ dispute, a conflict-of-laws analysis is necessary. See generally R. Weintraub, Commentary on Conflict of Laws 2-3 (1980); Restatement (First) of Conflict of Laws § 1, Comment c and Illustrations (1934).

Literally the only support that the Court adduces for its position is §403 of the Restatement (Third) — or more precisely Comment e to that provision, which states:

“Subsection (3) [which says that a State should defer to another state if that State’s interest is clearly greater] applies only when one state requires what another prohibits, or where compliance with the regulations of two states exercising jurisdiction consistently with this section is otherwise impossible. It does not apply where a person subject to regulation by two states can comply with the laws of both . . ..”

The Court has completely misinterpreted this provision. Subsection (3) of §403 (requiring one State to defer to another in the limited circumstances just described) comes into play only after subsection (1) of §403 has been complied with — i. e., after it has been determined that the exercise of jurisdiction by both of the two States is not “unreasonable.” That prior question is answered by applying the factors (inter alia) set forth in subsection (2) of §403, that is, precisely the factors that I have discussed in text and that the Court rejects.11

*824* * *

I would reverse the judgment of the Court of Appeals on this issue, and remand to the District Court with instructions to dismiss for failure to state a claim on the three counts at issue in No. 91-1128.

announced the judgment of the Court and delivered the opinion of the Court with respect to Parts I, II-A, III, and IV, and an opinion concurring in the judgment with respect to Part II-B.*

The Sherman Act makes every contract, combination, or conspiracy in unreasonable restraint of interstate or foreign commerce illegal. 26 Stat. 209, as amended, 15 U. S. C. § 1. These consolidated cases present questions about the application of that Act to the insurance industry, both here and abroad. The plaintiffs (respondents here) allege that both domestic and foreign defendants (petitioners here) violated the Sherman Act by engaging in various conspiracies to affect the American insurance market. A group of domestic defendants argues that the McCarran-Ferguson Act, 59 Stat. 33, as amended, 15 U. S. C. § 1011 et seq., precludes application of the Sherman Act to the conduct alleged; a group of foreign defendants argues that the principle of international comity requires the District Court to refrain from exercising jurisdiction over certain claims against it. We hold that most of the domestic defendants’ alleged conduct is not im*772munized from antitrust liability by the McCarran-Ferguson Act, and that, even assuming it applies, the principle of international comity does not preclude District Court jurisdiction over the foreign conduct alleged.

I

The two petitions before us stem from consolidated litigation comprising the complaints of 19 States and many private plaintiffs alleging that the defendants, members of the insurance industry, conspired in violation of § 1 of the Sherman Act to restrict the terms of coverage of commercial general liability (CGL) insurance1 available in the United States. Because the cases come to us on motions to dismiss, we take the allegations of the complaints as true.2

A

According to the complaints, the object of the conspiracies was to force certain primary insurers (insurers who sell insurance directly to consumers) to change the terms of their *773standard CGL insurance policies to conform with the policies the defendant insurers wanted to sell. The defendants wanted four changes.3

First, CGL insurance has traditionally been sold in the United States on an “occurrence” basis, through a policy obligating the insurer “to pay or defend claims, whenever made, resulting from an accident or ‘injurious exposure to conditions’ that occurred during the [specific time] period the policy was in effect.” App. 22 (Cal. Complaint ¶ 52). In place of this traditional “occurrence” trigger of coverage, the defendants wanted a “claims-made” trigger, obligating the insurer to pay or defend only those claims made during the policy period. Such a policy has the distinct advantage for the insurer that when the policy period ends without a claim having been made, the insurer can be certain that the policy will not expose it to any further liability. Second, the defendants wanted the “claims-made” policy to have a “retroactive date” provision, which would further restrict coverage to claims based on incidents that occurred after a certain date. Such a provision eliminates the risk that an insurer, by issuing a claims-made policy, would assume liability arising from incidents that occurred before the policy’s effective date, but remained undiscovered or caused no immediate harm. Third, CGL insurance has traditionally covered “sudden and accidental” pollution; the defendants wanted to eliminate that coverage. Finally, CGL insurance has traditionally provided that the insurer would bear the legal costs of defending covered claims against the insured without regard to the policy’s stated limits of coverage; the defendants *774wanted legal defense costs to be counted against the stated limits (providing a “legal defense cost cap”).

To understand how the defendants are alleged to have pressured the targeted primary insurers to make these changes, one must be aware of two important features of the insurance industry. First, most primary insurers rely on certain outside support services for the type of insurance coverage they wish to sell. Defendant Insurance Services Office, Inc. (ISO), an association of approximately 1,400 domestic property and casualty insurers (including the primary insurer defendants, Hartford Fire Insurance Company, Allstate Insurance Company, CIGNA Corporation, and Aetna Casualty and Surety Company), is the almost exclusive source of support services in this country for CGL insurance. See id., at 19 (Cal. Complaint ¶ 38). ISO develops standard policy forms and files or lodges them with each State’s insurance regulators; most CGL insurance written in the United States is written on these forms. Ibid. (Cal. Complaint ¶ 39); id., at 74 (Conn. Complaint ¶ 50). All of the “traditional” features of CGL insurance relevant to this litigation were embodied in the ISO standard CGL insurance form that had been in use since 1973 (1973 ISO CGL form). Id., at 22 (Cal. Complaint ¶¶ 51-54); id., at 75 (Conn. Complaint ¶¶ 56-58). For each of its standard policy forms, ISO also supplies actuarial and rating information: it collects, aggregates, interprets, and distributes data on the premiums charged, claims filed and paid, and defense costs expended with respect to each form, id., at 19 (Cal. Complaint ¶ 39); id., at 74 (Conn. Complaint ¶¶ 51-52), and on the basis of these data it predicts future loss trends and calculates advisory premium rates, id., at 19 (Cal. Complaint ¶ 39); id., at 74 (Conn. Complaint ¶ 53). Most ISO members cannot afford to continue to use a form if ISO withdraws these support services. See id., at 32-33 (Cal. Complaint ¶ ¶ 97, 99).

Second, primary insurers themselves usually purchase insurance to cover a portion of the risk they assume from the *775consumer. This so-called “reinsurance” may serve at least two purposes, protecting the primary insurer from catastrophic loss, and allowing the primary insurer to sell more insurance than its own financial capacity might otherwise permit. Id., at 17 (Cal. Complaint ¶ 29). Thus, “[t]he availability of reinsurance affects the ability and willingness of primary insurers to provide insurance to their customers.” Id., at 18 (Cal. Complaint ¶ 34); id., at 63 (Conn. Complaint ¶ 4(p)). Insurers who sell reinsurance themselves often purchase insurance to cover part of the risk they assume from the primary insurer; such “retrocessional reinsurance” does for reinsurers what reinsurance does for primary insurers. See ibid. (Conn. Complaint ¶ 4(r)). Many of the defendants here are reinsurers or reinsurance brokers, or play some other specialized role in the reinsurance business; defendant Reinsurance Association of America (RAA) is a trade association of domestic reinsurers.

B

The prehistory of events claimed to give rise to liability starts in 1977, when ISO began the process of revising its 1973 CGL form. Id., at 22 (Cal. Complaint ¶ 55). For the first time, it proposed two CGL forms (1984 ISO CGL forms), one the traditional “occurrence” type, the other “with a new ‘claims-made’ trigger.” Id., at 22-23 (Cal. Complaint ¶ 56). The “claims-made” form did not have a retroactive date provision, however, and both 1984 forms covered “ ‘sudden and accidental’ pollution” damage and provided for unlimited coverage of legal defense costs by the insurer. Id., at 23 (Cal. Complaint ¶¶ 59-60). Within the ISO, defendant Hartford Fire Insurance Company objected to the proposed 1984 forms; it desired elimination of the “occurrence” form, a retroactive date provision on the “claims-made” form, elimination of sudden and accidental pollution coverage, and a legal defense cost cap. Defendant Allstate Insurance Company also expressed its desire for a retroactive date provision on *776the “claims-made” form. Id., at 24 (Cal. Complaint ¶ 61). Majorities in the relevant ISO committees, however, supported the proposed 1984 CGL forms and rejected the changes proposed by Hartford and Allstate. In December 1983, the ISO Board of Directors approved the proposed 1984 forms, and ISO filed, or lodged the forms with state regulators in March 1984. Ibid. (Cal. Complaint ¶ 62).

Dissatisfied with this state of affairs, the defendants began to take other steps to force a change in the terms of coverage of CGL insurance generally available, steps that, the plaintiffs allege, implemented a series of conspiracies in violation of § 1 of the Sherman Act. The plaintiffs recount these steps as a number of separate episodes corresponding to different claims for relief in their complaints;4 because it will become important to distinguish among these counts and the acts and defendants associated with them, we will note these correspondences.

The first four Claims for Relief in the California Complaint, id., at 36-43 (¶¶ 111-130), and the Second Claim for Relief in the Connecticut Complaint, id., at 90-92 (¶ ¶ 120-124), charge the four domestic primary insurer defendants and varying groups of domestic and foreign reinsurers, brokers, and associations with conspiracies to manipulate the ISO CGL forms. In March 1984, primary insurer Hartford persuaded General Reinsurance Corporation (General Re), the largest American reinsurer, to take steps either to procure desired changes in the ISO CGL forms, or “failing that, [to] ‘derail’ the entire ISO CGL forms program.” Id., at 24 (Cal. Complaint ¶ 64). General Re took up the matter with its trade association, RAA, which created a special committee that met and agreed to “boycott” the 1984 ISO CGL forms unless a retroactive-date provision was added to the *777claims-made form, and a pollution exclusion and defense cost cap were added to both forms. Id., at 24-25 (Cal. Complaint ¶¶ 65-66). RAA then sent a letter to ISO “announcing] that its members would not provide reinsurance for coverages written on the 1984 CGL forms,” id., at 25 (Cal. Complaint ¶ 67), and Hartford and General Re enlisted a domestic reinsurance broker to give a speech to the ISO Board of Directors, in which he stated that no reinsurers would “break ranks” to reinsure the 1984 ISO CGL forms. Ibid. (Cal. Complaint ¶ 68).

The four primary insurer defendants (Hartford, Aetna, CIGNA, and Allstate) also encouraged key actors in the London reinsurance market, an important provider of reinsurance for North American risks, to withhold reinsurance for coverages written on the 1984 ISO CGL forms. Id., at 25-26 (Cal. Complaint ¶ ¶ 69-70). As a consequence, many London-based underwriters, syndicates, brokers, and reinsurance companies informed ISO of their intention to withhold reinsurance on the 1984 forms, id., at 26-27 (Cal. Complaint ¶¶ 71-75), and at least some of them told ISO that they would withhold reinsurance until ISO incorporated all four desired changes, see supra, at 771, and n. 3, into the ISO CGL forms. App. 26 (Cal. Complaint ¶ 74).

For the first time ever, ISO invited representatives of the domestic and foreign reinsurance markets to speak at an ISO Executive Committee meeting. Id., at 27-28 (Cal. Complaint ¶78). At that meeting, the reinsurers “presented their agreed upon positions that there would be changes in the CGL forms or no reinsurance.” Id., at 29 (Cal. Complaint ¶ 82). The ISO Executive Committee then voted to include a retroactive-date provision in the claims-made form, and to exclude all pollution coverage from both new forms. (But it neither eliminated the occurrence form, nor added a legal defense cost cap.) The 1984 ISO CGL forms were then withdrawn from the marketplace, and replaced with forms (1986 ISO CGL forms) containing the new provisions. Ibid. *778(Cal. Complaint ¶ 84). After ISO got regulatory approval of the 1986 forms in most States where approval was needed, it eliminated its support services for the 1973 CGL form, thus rendering it impossible for most ISO members to continue to use the form. Id,., at 32-33 (Cal. Complaint ¶¶97, 99).

The Fifth Claim for Relief in the California Complaint, id., at 43-44 (¶¶ 131-135), and the virtually identical Third Claim for Relief in the Connecticut Complaint, id., at 92-94 (¶¶ 125-129), charge a conspiracy among a group of London reinsurers and brokers to coerce primary insurers in the United States to offer CGL coverage only on a claims-made basis. The reinsurers collectively refused to write new reinsurance contracts for, or to renew longstanding contracts with, “primary ... insurers unless they were prepared to switch from the occurrence to the claims-made form,” id., at 30 (Cal. Complaint ¶ 88); they also amended their reinsurance contracts to cover only claims made before a “‘sunset date,’” thus eliminating reinsurance for claims made on occurrence policies after that date, id., at 31 (Cal. Complaint ¶¶ 90-92).

The Sixth Claim for Relief in the California Complaint, id., at 45-46 (¶¶ 136-140), and the nearly identical Fourth Claim for Relief in the Connecticut Complaint, id., at 94-95 (¶¶ 130-134), charge another conspiracy among a somewhat different group of London reinsurers to withhold reinsurance for pollution coverage. The London reinsurers met and agreed that all reinsurance contracts covering North American casualty risks, including CGL risks, would be written with a complete exclusion for pollution liability coverage. Id., at 32 (Cal. Complaint ¶¶ 94-95). In accordance with this agreement, the parties have in fact excluded pollution liability coverage from CGL reinsurance contracts since at least late 1985. Ibid. (Cal. Complaint ¶ 94).

*779The Seventh Claim for Relief in the California Complaint, id., at 46-47 (¶¶ 141-145), and the closely similar Sixth Claim for Relief in the Connecticut Complaint, id., at 97-98 (¶¶ 140-144), charge a group of domestic primary insurers, foreign reinsurers, and the ISO with conspiring to restrain trade in the markets for “excess” and “umbrella” insurance by drafting model forms and policy language for these types of insurance, which are not normally offered on a regulated basis. Id., at 33 (Cal. Complaint ¶ 101). The ISO Executive Committee eventually released standard language for both “occurrence” and “claims-made” umbrella and excess policies; that language included a retroactive date in the claims-made version, and an absolute pollution exclusion and a legal defense cost cap in both versions. Id., at 34 (Cal. Complaint ¶ 105).

Finally, the Eighth Claim for Relief in the California Complaint, id., at 47-49 (¶ ¶ 146-150), and its counterpart in the Fifth Claim for Relief in the Connecticut Complaint, id., at 95-97 (¶¶ 135-139), charge a group of London and domestic retrocessional reinsurers5 with conspiring to withhold retrocessional reinsurance for North American seepage, pollution, and property contamination risks. Those retrocessional re-insurers signed, and have implemented, an agreement to use their “ ‘best endeavors’ ” to ensure that they would provide such reinsurance for North American risks “ ‘only . .. where the original business includes a seepage and pollution exclu*780sion wherever legal and applicable.’ ” Id., at 35 (Cal. Complaint ¶ 108).6

C

Nineteen States and a number of private plaintiffs filed 36 complaints against the insurers involved in this course of events, charging that the conspiracies described above violated § 1 of the Sherman Act, 15 U. S. C. § 1. After the actions had been consolidated for litigation in the Northern District of California, the defendants moved to dismiss for failure to state a cause of action, or, in the alternative, for summary judgment. The District Court granted the motions to dismiss. In re Insurance Antitrust Litigation, 723 F. Supp. 464 (1989). It held that the conduct alleged fell within the grant of antitrust immunity contained in § 2(b) of the McCarran-Ferguson Act, 15 U. S. C. § 1012(b), because it amounted to “the business of insurance” and was “regulated by State Law” within the meaning of that section; none of the conduct, in the District Court’s view, amounted to a “boycott” within the meaning of the § 3(b) exception to that grant of immunity. 15 U. S. C. § 1013(b). The District Court also dismissed the three claims that named only certain London-based defendants,7 invoking international comity and applying the Ninth Circuit’s decision in Timberlane Lumber Co. v. Bank of America, N. T. & S. A., 549 F. 2d 597 (1976).

The Court of Appeals reversed. In re Insurance Antitrust Litigation, 938 F. 2d 919 (CA9 1991). Although it held the conduct involved to be “the business of insurance” within the meaning of § 2(b), it concluded that the defendants could *781not claim McCarran-Ferguson Act antitrust immunity for two independent reasons. First, it held, the foreign reinsurers were beyond the regulatory jurisdiction of the States; because their activities could not be “regulated by State Law” within the meaning of §2(b), they did not fall within that section’s grant of immunity. Although the domestic insurers were “regulated by State Law,” the court held, they forfeited their § 2(b) exemption when they conspired with the nonexempt foreign reinsurers. Second, the Court of Appeals held that, even if the conduct alleged fell within the scope of §2(b), it also fell within the § 3(b) exception for “act[s] of boycott, coercion, or intimidation.” Finally, as to the'three claims brought solely against foreign defendants, the court applied its Timberlane analysis, but concluded that the principle of international comity was no bar to exercising Sherman Act jurisdiction.

We granted certiorari in No. 91-1111 to address two narrow questions about the scope of McCarran-Ferguson Act antitrust immunity,8 and in No. 91-1128 to address the application of the Sherman Act to the foreign conduct at issue.9 506 U. S. 814 (1992). We now affirm in part, reverse in part, and remand.

*782II

The petition in No. 91 — 1111 touches on the interaction of two important pieces of economic legislation. The Sherman Act declares “[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, ... to be illegal.” 15 U. S. C. § 1. The McCarran-Perguson Act provides that regulation of the insurance industry is generally a matter for the States, 15 U. S. C. § 1012(a), and (again, generally) that “[n]o Act of Congress shall be construed to invalidate, impair, or supersede any law enacted by any State for the purpose of regulating the business of insurance,” § 1012(b). Section 2(b) of the McCarranFerguson Act makes it clear nonetheless that the Sherman Act applies “to the business of insurance to the extent that such business is not regulated by State Law,” § 1012(b), and § 3(b) provides that nothing in the McCarran-Ferguson Act “shall render the ... Sherman Act inapplicable to any agreement to boycott, coerce, or intimidate, or act of boycott, coercion, or intimidation,” § 1013(b).

Petitioners in No. 91-1111 are all of the domestic defendants in the consolidated cases: the four domestic primary insurers, the domestic reinsurers, the trade associations ISO and RAA, and the domestic reinsurance broker Thomas A. Greene & Company, Inc. They argue that the Court of Appeals erred in holding, first, that their conduct, otherwise immune from antitrust liability under § 2(b) of the McCarranFerguson Act, lost its immunity when they conspired with the foreign defendants, and, second, that their conduct amounted to “act[s] of boycott” falling within the exception to antitrust immunity set out in §3(b). We conclude that the Court of Appeals did err about the effect of conspiring with foreign defendants, but correctly decided that all but one of the complaints’ relevant Claims for Relief are fairly read to allege conduct falling within the “boycott” exception to McCarran-Ferguson Act antitrust immunity. We there*783fore affirm the Court of Appeals’s judgment that it was error for the District Court to dismiss the complaints on grounds of McCarran-Ferguson Act immunity, except as to the one claim for relief that the Court of Appeals correctly found to allege no boycott.

A

By its terms, the antitrust exemption of §2(b) of the McCarran-Ferguson Act applies to “the business of insurance” to the extent that such business is regulated by state law. While “business” may mean “[a] commercial or industrial establishment or enterprise,” Webster’s New International Dictionary 362 (2d ed. 1942), the definite article before “business” in § 2(b) shows that the word is not used in that sense, the phrase “the business of insurance” obviously not being meant to refer to a single entity. Rather, “business” as used in § 2(b) is most naturally read to refer to “Mercantile transactions; buying and selling; [and] traffic.” Ibid.

The cases confirm that “the business of insurance” should be read to single out one activity from others, not to distinguish one entity from another. In Group Life & Health Ins. Co. v. Royal Drug Co., 440 U. S. 205 (1979), for example, we held that §2(b) did not exempt an insurance company from antitrust liability for making an agreement fixing the price of prescription drugs to be sold to Blue Shield policyholders. Such activity, we said, “would be exempt from the antitrust laws if Congress had extended the coverage of the McCarran-Ferguson Act to the ‘business of insurance companies.’ But that is precisely what Congress did not do.” Id., at 233 (footnote omitted); see SEC v. National Securities, Inc., 393 U. S. 453, 459 (1969) (the McCarran-Ferguson Act’s “language refers not to the persons or companies who are subject to state regulation, but to laws ‘regulating the business of insurance’”) (emphasis in original). And in Union Labor Life Ins. Co. v. Pireno, 458 U. S. 119 (1982), we explicitly framed the question as whether “a particular practice is part of the ‘business of insurance’ exempted from the anti*784trust laws by §2(b),” id., at 129 (emphasis added), and each of the three criteria we identified concerned a quality of the practice in question: “first, whether the practice has the effect of transferring or spreading a policyholder’s risk; second, whether the practice is an integral part of the policy relationship between the insurer and the insured; and third, whether the practice is limited to entities within the insurance industry,” ibid, (emphasis in original).

The Court of Appeals did not hold that, under these criteria, the domestic defendants’ conduct fell outside “the business of insurance”; to the contrary, it held that that condition was met.10 See 938 F. 2d, at 927. Nor did it hold the domestic defendants’ conduct to be “ [unregulated by State Law.” Rather, it constructed an altogether different chain of reasoning, the middle link of which comes from a sentence in our opinion in Royal Drug Co. “[Regulation ... of foreign reinsurers,” the Court of Appeals explained, “is beyond the jurisdiction of the states,” 938 F. 2d, at 928, and hence § 2(b) does not exempt foreign reinsurers from antitrust liability, because their activities are not “regulated by State Law.” Under Royal Drug Co., “an exempt entity forfeits antitrust exemption by acting in concert with nonexempt parties.” 440 U. S., at 231. Therefore, the domestic insurers, by acting in concert with the nonexempt foreign insurers, lost their McCarran-Ferguson Act antitrust immunity. See 938 F. 2d, at 928. This reasoning fails, however, because even if we were to agree that foreign reinsurers were not subject to state regulation (a point on which we express no opinion), the quoted language from Royal Drug Co., read *785in context, does not state a proposition applicable to this litigation.

The full sentence from Royal Drug Co. places the quoted fragment in a different light. “In analogous contexts,” we stated, “the Court has held that an exempt entity forfeits antitrust exemption by acting in concert with nonexempt parties.” 440 U. S., at 231. We then cited two cases dealing with the Capper-Volstead Act, which immunizes from liability under § 1 of the Sherman Act particular activities of certain persons “engaged in the production of agricultural products.”11 Capper-Volstead Act, § 1, 42 Stat. 388, 7 U. S. C. §291; see Case-Swayne Co. v. Sunkist Growers, Inc., 389 U. S. 384 (1967); United States v. Borden Co., 308 U. S. 188 (1939). Because these cases relied on statutory language referring to certain “persons,” whereas we specifically acknowledged in Royal Drug Co. that the McCarranFerguson Act immunizes activities rather than entities, see 440 U. S., at 232-233, the analogy we were drawing was of course a loose one. The agreements that insurance companies made with “parties wholly outside the insurance industry,” id., at 231, we noted, such as the retail pharmacists involved in Royal Drug Co. itself, or “automobile body repair shops or landlords,” id., at 232 (footnote omitted), are un*786likely to be about anything that could be called “the business of insurance,” as distinct from the broader “ ‘business of insurance companies/” id., at 233. The alleged agreements at issue in the instant litigation, of course, are entirely different; the foreign reinsurers are hardly “wholly outside the insurance industry,” and respondents do not contest the Court of Appeals’s holding that the agreements concern “the business of insurance.” These facts neither support even the rough analogy we drew in Royal Drug Co. nor fall within the rule about acting in concert with nonexempt parties, which derived from a statute inapplicable here. Thus, we think it was error for the Court of Appeals to hold the domestic insurers bereft of their McCarran-Ferguson Act exemption simply because they agreed or acted with foreign reinsurers that, we assume for the sake of argument, were “not regulated by State Law.”12

B

That the domestic defendants did not lose their § 2(b) exemption by acting together with foreign reinsurers, however, is not enough reason to reinstate the District Court’s dismissal order, for the Court of Appeals reversed that order on two independent grounds. Even if the participation of foreign reinsurers did not affect the §2(b) exemption, the Court of Appeals held, the agreements and acts alleged by the plaintiffs constitute “agreement^] to boycott” and “act[s] of boycott [and] coercion” within the meaning of § 3(b) of the McCarran-Ferguson Act, which makes it clear that the Sherman Act applies to such agreements and acts regardless of the §2(b) exemption. See 938 F. 2d, at 928. I agree with *787the Court that, construed in favor of the plaintiffs, the First, Second, Third, and Fourth Claims for Relief in the California Complaint, and the First and Second Claims for Relief in the Connecticut Complaint, allege one or more §3(b) “act[s] of boycott,” and are thus sufficient to survive a motion to dismiss. See infra, at 789-790; post, at 811.

In reviewing the motions to dismiss, however, the Court has decided to use what I believe to be an overly narrow definition of the term “boycott” as used in §3(b), confining it to those refusals to deal that are “unrelated” or “collateral” to the objective sought by those refusing to deal. Post, at 803. I do not believe that the McCarran-Ferguson Act or our precedents warrant such a cramped reading of the term.

The majority and I find common ground in four propositions concerning § 3(b) boycotts, as established in our decisions in St. Paul Fire & Marine Ins. Co. v. Barry, 438 U. S. 531 (1978), and United States v. South-Eastern Underwriters Assn., 322 U. S. 533 (1944). First, as we noted in St. Paul, our only prior decision construing “boycott” as it appears in §3(b), only those refusals to deal involving the coordinated action of multiple actors constitute § 3(b) boycotts: “conduct by individual actors falling short of concerted activity is simply not a ‘boycott’ within [the meaning of] §3(b).” 438 U. S., at 555; see post, at 800 (“‘boycott’” used “to describe . . . collective action”); post, at 801 (“To ‘boycott’ means ‘[t]o combine in refusing to hold relations’” (citation omitted)).

Second, a § 3(b) boycott need not involve an absolute refusal to deal.13 A primary goal of the alleged conspirators in South-Eastern Underwriters, as we described it, was “to force nonmember insurance companies into the conspiracies.”14 322 U. S., at 535; cf. Joint Hearing on S. 1362, H. R. *7883269, and H. R. 3270 before the Subcommittees of the Senate Committee on the Judiciary, 78th Cong., 1st Sess., pt. 2, p. 335 (1943) (statement of Edward L. Williams, President, Insurance Executives Association) (“[T]he companies that want to come into the Interstate Underwriters Board can come in there. I do not know of any company that is turned down”). Thus, presumably, the refusals to deal orchestrated by the defendants would cease if the targets agreed to join the association and abide by its terms. See post, at 801 (“The refusal to deal may ... be conditional” (emphasis omitted)).

Third, contrary to petitioners’ contentions, see Brief for Petitioners in No. 91-1111, pp. 32, n. 14, 34, 38-39, a §3(b) boycott need not entail unequal treatment of the targets of the boycott and its instigators. Some refusals to deal (those, perhaps, which are alleged to violate only § 2 of the Sherman Act15) may have as their object the complete destruction of the business of competitors; these may well involve unconditional discrimination against the targets. Other refusals to deal, however, may seek simply to prevent competition as to the price or features of the product sold; and these need not depend on unequal treatment of the targets. Assuming, *789as the South-Eastern Underwriters Court appears to have done, that membership in the defendant association was open to all insurers, the association is most readily seen as having intended to treat all insurers equally: they all had the choice either to join the association and abide by its rules, or to be subjected to the “boycotts,” and acts of coercion and intimidation, alleged in that case. See post, at 808 (describing South-Eastern Underwriters as involving a “boycott, by primary insurers, of competitors who refused to join their price-fixing conspiracy”).

Fourth, although a necessary element, “concerted activity” is not, by itself, sufficient for a finding of “boycott” under §3(b). Were this the case, we recognized in Barry, §3(b) might well “ ‘devour the broad antitrust immunity bestowed by § 2(b)/ ” 438 U. S., at 545, n. 18 (quoting id., at 559 (Stewart, J., dissenting)), since every “contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce,” 15 U. S. C. § 1, involves “concerted activity.” Thus, we suggested, simple price fixing has been treated neither as a boycott nor as coercion “in the absence of any additional enforcement activity.” 438 U. S., at 545, n. 18; see post, at 804 (contending that simple concerted agreements on contract terms are not properly characterized as boycotts).

Contrary to the majority’s view, however, our decisions have suggested that “enforcement activity” is a multifarious concept. The South-Eastern Underwriters Court, which coined the phrase “boycotts[,]... coercion and intimidation,” 322 U. S., at 535; see n. 14, supra, provides us with a list of actions that, it finds, are encompassed by these terms. “Companies not members of [the association],” it states, “were cut off from the opportunity to reinsure their risks, and their services and facilities were disparaged; independent sales agencies who defiantly represented non-[association] companies were punished by a withdrawal of the right to represent the members of [the association]; and persons needing insurance who purchased from non-*790[association] companies were threatened with boycotts and withdrawal of all patronage.” 322 U. S., at 535-536. Faced with such a list, and with all of the other instances in which we have used the term “boycott,” we rightly came to the conclusion in Barry that, as used in our cases, the term does not refer to a “‘unitary phenomenon.’” 438 U. S., at 543 (quoting P. Areeda, Antitrust Analysis 381 (2d ed. 1974)).

The question in this litigation is whether the alleged activities of the domestic defendants, acting together with the foreign defendants who are not petitioners here, include “enforcement activities” that would raise the claimed attempts to fix terms to the level of § 3(b) boycotts. I believe they do. The core of the plaintiffs’ allegations against the domestic defendants concern those activities that form the basis of the First, Second, Third, and Fourth Claims for Relief in the California Complaint, and the Second Claim for Relief in the Connecticut Complaint: the conspiracies involving both the primary insurers and domestic and foreign brokers and re-insurers to force changes in the ISO CGL forms. According to the complaints, primary insurer defendants Hartford and Allstate first tried to convince other members of the ISO that the ISO CGL forms should be changed to limit coverage in the manner we have detailed above, see supra, at 773-774; but they failed to persuade a majority of members of the relevant ISO committees, and the changes were not made. Unable to persuade other primary insurers to agree voluntarily to their terms, Hartford and Allstate, joined by Aetna and CIGNA, sought the aid of other individuals and entities who were not members of ISO, and who would not ordinarily be parties to an agreement setting the terms of primary insurance, not being in the business of selling it. The four primary insurers convinced these individuals and entities, the reinsurers, to put pressure on ISO and its members by refusing to reinsure coverages written on the ISO CGL forms until the desired changes were made. Both domestic and foreign reinsurers, acting at the behest of the four pri*791mary insurers, announced that they would not reinsure under the ISO CGL forms until changes were made. As an immediate result of this pressure, ISO decided to include a retroactive-date provision in its claims-made form, and to exclude all pollution coverage from both its claims-made and occurrence forms. In sum, the four primary insurers solicited refusals to deal from outside the primary insurance industry as a means of forcing their fellow primary insurers to agree to their terms; the outsiders, acting at the behest of the four, in fact refused to deal with primary insurers until they capitulated, which, in part at least, they did.

This pattern of activity bears a striking resemblance to the first act of boycott listed by the South-Eastern Underwriters Court; although neither the South-Eastern Underwriters opinion, nor the underlying indictment, see Transcript of Record, O. T. 1943, No. 354, p. 11 (¶ 22(e)), details exactly how the defendants managed to “cut off [nonmembers] from the opportunity to reinsure their risks,” 322 U. S., at 535, the defendants could have done so by prompting reinsurance companies to refuse to deal with nonmembers, just as is alleged here.16 Moreover, the activity falls squarely *792within even the narrow theory of the § 3(b) exception Justice Stewart advanced in dissent in Barry. Under that theory,17 the § 3(b) exception should be limited to “attempts by members of the insurance business to force other members to follow the industry’s private rules and practices.” 438 U. S., at 565. I can think of no better description of the four primary insurers’ activities in this litigation. For these reasons, I agree with the Court’s ultimate conclusion that the Court of Appeals was correct in reversing the District Court’s dismissal of the First, Second, Third, and Fourth Claims for Relief in the California Complaint, and the Second Claim for Relief in the Connecticut Complaint.18

*793The majority concludes that, so long as the reinsurers’ role in this course of action was limited to “a concerted agreement to seek particular terms in particular transactions,” post, at 801-802, the course of action could never constitute a §3(b) boycott. The majority’s emphasis on this conclusion assumes an artificial segmentation of the course of action, and a false perception of the unimportance of the elements of that course of action other than the reinsurers’ agreement. The majority concedes that the complaints allege, not just implementation of a horizontal agreement, but refusals to deal that occurred “at the behest of,” or were “solicited by,” the four primary insurers, who were “competitors of the tar*794get[s].” Post, at 808 (citations and internal quotation marks omitted). But it fails to acknowledge several crucial features of these events that bind them into a single course of action recognizable as a § 3(b) boycott.

First, the allegation that the reinsurers acted at the behest of the four primary insurers excludes the possibility that the reinsurers acted entirely in their own independent self-interest, and would have taken exactly the same course of action without the intense efforts of the four primary insurers. Although the majority never explicitly posits such autonomy on the part of the reinsurers, this would seem to be the only point of its repeated emphasis on the fact that “the scope and predictability of the risks assumed in a reinsurance contract depend entirely upon the terms of the primary policies that are reinsured.” Ibid. If the encouragement of the four primary insurers played no role in the reinsurers’ decision to act as they did, then it is difficult to see how one could describe the reinsurers as acting at the behest of the primary insurers, an element I find crucial to the § 3(b) boycott alleged here. From the vantage point of a ruling on motions to dismiss, however, I discern sufficient allegations in the complaints that this is not the case. In addition, according to the complaints, the four primary insurers were not acting out of concern for the reinsurers’ financial health when they prompted the reinsurers to refuse reinsurance for certain risks; rather, they simply wanted to ensure that no other primary insurer would be able to sell insurance policies that they did not want to sell. Finally, as the complaints portray the business of insurance, reinsurance is a separate, specialized product, “[t]he availability [of which] affects the ability and willingness of primary insurers to provide insurance to their customers.” App. 18 (Cal. Complaint ¶34). Thus, contrary to the majority’s assertion, the boundary between the primary insurance industry and the reinsurance industry is not merely “technica[l].” Post, at 808.

*795The majority insists that I “disregard] th[e] integral relationship between the terms of the primary insurance form and the contract of reinsurance,” post, at 807, a fact which it seems to believe makes it impossible to draw any distinction whatsoever between primary insurers and reinsurers. Yet it is the majority that fails to see that, in spite of such an “integral relationship,” the interests of primary insurer and reinsurer will almost certainly differ in some cases. For example, the complaints allege that reinsurance contracts often “layer” risks, “in the sense that [a] reinsurer may have to respond only to claims above a certain amount. . . .” App. 10 (Cal. Complaint ¶ 4.q); id., at 61 (Conn. Complaint ¶ 4(f)). Thus, a primary insurer might be much more concerned than its reinsurer about a risk that resulted in a high number of relatively small claims. Or the primary insurer might simply perceive a particular risk differently from the reinsurer. The reinsurer might be indifferent as to whether a particular risk was covered, so long as the reinsurance premiums were adjusted to its satisfaction, whereas the primary insurer might decide that the risk was “too hot to handle,” on a standardized basis, at any cost. The majority’s suggestion that “to insist upon certain primary-insurance terms as a condition of writing reinsurance is in no way ‘artificial,’” post, at 808; see post, at 806, simply ignores these possibilities; the conditions could quite easily be “artificial,” in the sense that they are not motivated by the interests of the reinsurers themselves. Because the parties have had no chance to flesh out the facts of this case, because I have no a priori knowledge of those facts, and because I do not believe I can locate them in the pages of insurance treatises, I would not rule out these possibilities on a motion to dismiss.

Believing that there is no other principled way to narrow the §3(b) exception, the majority decides that “boycott” encompasses just those refusals to deal that are “unrelated” or “collateral” to the objective sought by those refusing to deal. Post, at 803. This designation of a single “ ‘unitary phenom*796enon' ” Barry, 438 U. S., at 543, to which the term “boycott” will henceforth be confined, is of course at odds with our own description of our Sherman Act cases in Barry19 See ibid. Moreover, the limitation to “collateral” refusals to deal threatens to shrink the § 3(b) exception far more than the majority is willing to admit. Even if the reinsurers refused all reinsurance to primary insurers “who wrote insurance on disfavored forms,” including insurance “as to risks written on other forms,” the majority states, the reinsurers would not be engaging in a § 3(b) boycott if “the primary insurers’ other business were relevant to the proposed insurance contract (for example, if the reinsurer bears greater risk where the primary insurer engages in riskier businesses).” Post, at 810 (emphasis deleted). Under this standard, and under facts comparable to those in this litigation, I assume that reinsurers who refuse to deal at all with a primary insurer unless it ceases insuring a particular risk would not be engaging in a § 3(b) boycott if they could show that (1) insuring the risk in question increases the probability that the primary insurer will become insolvent, and that (2) it costs more to administer the reinsurance contracts of a bankrupt primary insurer (including those unrelated to the risk that caused the primary insurer to declare bankruptcy). One can only imagine the variety of similar arguments that may slowly plug what remains of the § 3(b) exception. For these reasons, I cannot agree with the majority’s narrow theory of § 3(b) boycotts.

Ill

Finally, we take up the question presented by No. 91-1128, whether certain claims against the London reinsurers should have been dismissed as improper applications of the Sher*797man Act to foreign conduct. The Fifth Claim for Relief in the California Complaint alleges a violation of §1 of the Sherman Act by certain London reinsurers who conspired to coerce primary insurers in the United States to offer CGL coverage on a claims-made basis, thereby making “occurrence CGL coverage ... unavailable in the State of California for many risks.” App. 43-44 (¶¶ 131-135). The Sixth Claim for Relief in the California Complaint alleges that the London reinsurers violated § 1 by a conspiracy to limit coverage of pollution risks in North America, thereby rendering “pollution liability coverage . . . almost entirely unavailable for the vast majority of casualty insurance purchasers in the State of California.” Id., at 45-46 (¶¶ 136-140). The Eighth Claim for Relief in the California Complaint alleges a further § 1 violation by the London reinsurers who, along with domestic retrocessional reinsurers, conspired to limit coverage of seepage, pollution, and property contamination risks in North America, thereby eliminating such coverage in the State of California.20 Id., at 47-48 (¶¶ 146-150).

At the outset, we note that the District Court undoubtedly had jurisdiction of these Sherman Act claims, as the London reinsurers apparently concede. See Tr. of Oral Arg. 37 (“Our position is not that the Sherman Act does not apply in the sense that a minimal basis for the exercise of jurisdiction doesn’t exist here. Our position is that there are certain circumstances, and that this is one of them, in which the interests of another State are sufficient that the exercise of that jurisdiction should be restrained”).21 Although the *798proposition was perhaps not always free from doubt, see American Banana Co. v. United Fruit Co., 213 U. S. 347 (1909), it is well established by now that the Sherman Act applies to foreign conduct that was meant to produce and did in fact produce some substantial effect in the United States. See Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 U. S. 574, 582, n. 6 (1986); United States v. Aluminum Co. of America, 148 P. 2d 416, 444 (CA2 1945) (L. Hand, J.); Restatement (Third) of Foreign Relations Law of the United States §415, and Reporters’ Note 3 (1987) (hereinafter Restatement (Third) Foreign Relations Law); 1 P. Areeda & D. Turner, Antitrust Law ¶ 236 (1978); cf. Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U. S. 690, 704 (1962); Steele v. Bulova Watch Co., 344 U. S. 280, 288 (1952); United States v. Sisal Sales Corp., 274 U. S. 268, 275-276 (1927).22 Such is the conduct alleged here: that the London reinsurers engaged in unlawful conspiracies to affect the market for insurance in the United States and that their conduct in fact produced substantial effect.23 See 938 F. 2d, at 933.

*799According to the London reinsurers, the District Court should have declined to exercise such jurisdiction under the principle of international comity.24 The Court of Appeals agreed that courts should look to that principle in deciding whether to exercise jurisdiction under the Sherman Act. Id., at 932. This availed the London reinsurers nothing, however. To be sure, the Court of Appeals believed that “application of [American] antitrust laws to the London reinsurance market ‘would lead to significant conflict with English law and policy,’” and that “[s]uch a conflict, unless outweighed by other factors, would by itself be reason to decline *800exercise of jurisdiction.” Id., at 933 (citation omitted). But other factors, in the court’s view, including the London reinsurers’ express purpose to affect United States commerce and the substantial nature of the effect produced, outweighed the supposed conflict and required the exercise of jurisdiction in this litigation. Id., at 934.

When it enacted the FTAIA, Congress expressed no view on the question whether a court with Sherman Act jurisdiction should ever decline to exercise such jurisdiction on grounds of international comity. See H. R. Rep. No. 97-686, p. 13 (1982) (“If a court determines that the requirements for subject matter jurisdiction are met, [the FTAIA] would have no effect on the courtfs] ability to employ notions of comity ... or otherwise to take account of the international character of the transaction”) (citing Timberlane). We need not decide that question here, however, for even assuming that in a proper case a court may decline to exercise Sherman Act jurisdiction over foreign conduct (or, as Justice Scalia would put it, may conclude by the employment of comity analysis in the first instance that there is no jurisdiction), international comity would not counsel against exercising jurisdiction in the circumstances alleged here.

The only substantial question in this litigation is whether “there is in fact a true conflict between domestic and foreign law.” Société Nationale Industrielle Aérospatiale v. United States Dist. Court for Southern Dist. of Iowa, 482 U. S. 522, 555 (1987) (Blackmun, J., concurring in part and dissenting in part). The London reinsurers contend that applying the Act to their conduct would conflict significantly with British law, and the British Government, appearing before us as amicus curiae, concurs. See Brief for Petitioners Merrett Underwriting Agency Management Ltd. et al. in No. 91-1128, pp. 22-27; Brief for Government of United Kingdom of Great Britain and Northern Ireland as Amicus Curiae 10-14. They assert that Parliament has established a com*801prehensive regulatory regime over the London reinsurance market and that the conduct alleged here was perfectly consistent with British law and policy. But this is not to state a conflict. “[T]he fact that conduct is lawful in the state in which it took place will not, of itself, bar application of the United States antitrust laws,” even where the foreign state has a strong policy to permit or encourage such conduct. Restatement (Third) Foreign Relations Law §415, Comment j; see Continental Ore Co., supra, at 706-707. No conflict exists, for these purposes, “where a person subject to regulation by two states can comply with the laws of both.” Restatement (Third) Foreign Relations Law §403, Comment e.25 Since the London reinsurers do not argue that British law requires them to act in some fashion prohibited by the law of the United States, see Reply Brief for Petitioners Merrett Underwriting Agency Management Ltd. et al. in No. 91-1128, pp. 7-8, or claim that their compliance with the laws of both countries is otherwise impossible, we see no conflict with British law. See Restatement (Third) Foreign Relations Law §403, Comment e, §415, Comment j. We have no need in this litigation to address other considerations that might inform a decision to refrain from the exercise of jurisdiction on grounds of international comity.

IV

The judgment of the Court of Appeals is affirmed in part and reversed in part, and the cases are remanded for further proceedings consistent with this opinion.

It is so ordered.