I. INTRODUCTION
A contract between investor and host State contains a clause submitting disputes related to the contract to the exclusive jurisdiction of national courts or to (non-treaty) arbitration.Footnote 1 The host State breaches its substantive duties under the contract. The investor files a notice of arbitration under an investment treatyFootnote 2 between the State of its nationality and the host State, complaining that the breach of contract gives rise to a breach of the treaty.Footnote 3 Its doing so gives rise to an arbitration agreement in respect of the treaty claim since, all things being equal,Footnote 4 it has thereby accepted the host State's offer to arbitrate.Footnote 5 Which of these forum-selection agreements prevails: the earlier contractual clause or the later arbitration agreement?
Despite being of apparently foundational importance, the question has continued to arise even in the past year or so, in Anglo American v Venezuela,Footnote 6 in Nissan v India,Footnote 7 in Belenergia v Italy Footnote 8 and in CMC v Mozambique.Footnote 9 It arises most often and naturally in the context of a claim brought under an ‘umbrella clause’: a provision by which the host State undertakes that it will abide by some category of commitments entered into with the investor. The precise wording and scope of such a clause will, of course, vary from case to case; but as will be seen below it will often be wide enough to capture the State's contractual obligations to an investor. Such contractual obligations are themselves frequently encountered for the simple reason that the relationship between the investor and the State will often be governed at least in part by a contract.Footnote 10
The difficulty posed by the question has, if anything, increased with the number of awards attempting to answer it. Moreover, the answers given by some tribunals have been undermined by various forms of conceptual confusion, notably as to precisely which agreements are said to be in conflict, as to the basis on which they are thought to be in conflict and as to the relationship with other debates about the nature of claims under an umbrella clause. These problems in turn lead to persistently inconsistent approaches to the issue whether the question is one of jurisdiction or of admissibility, which though it may seem superficially unimportant can be crucial for parties seeking to enforce an award.
This article will suggest that the correct analysis is both simple and difficult to resist: that the question whether a contractual forum-selection clause ought to prevail over an agreement to arbitrate under an investment treaty is answered by assessing in each case whether the contractual clause amounts to a waiver of the investor's right of recourse to an investment tribunal.
Section II of this article sets the stage for this approach by considering the potential relevance of contractual forum-selection agreements in principle, as between the tribunal and national courts, by way of context, and to the parties’ agreement to arbitrate, in light of that. Section III then considers in detail how the contractual clause might affect the agreement to arbitrate under the treaty. It suggests that much the most satisfactory approach is to ask whether the contractual clause amounts to a waiver of the investor's right to arbitrate. That section considers alternative approaches and suggests that their merits are consistent with the waiver analysis advanced, whereas their difficulties are avoided by it. Section IV deals with whether the result of that analysis is that the issue is one of jurisdiction or admissibility, suggesting that the former is clearly the preferable analysis. It then addresses the consequences of that analysis.
II. THE RELEVANCE OF CONTRACTUAL FORUM-SELECTION AGREEMENTS IN PRINCIPLE
A. The Relationship between Investment Tribunals and National Courts
It is illuminating to begin by considering the differences between the jurisdiction of an investment tribunal and that of national courts. The investment tribunal's jurisdiction comes from the agreement to arbitrate, which is formed by the investor's acceptance of the host State's standing offer to arbitrate, expressed in the relevant treaty.Footnote 11 But frequently—perhaps even inevitably—that agreement and therefore the tribunal's jurisdiction will exist even though one or more domestic courts would consider that they also had jurisdiction over disputes arising from the same facts and legal relationship.Footnote 12
Domestic courts may have any number of bases on which they may or must exercise such jurisdiction, according to their internal rules of private international law.Footnote 13 Of particular relevance for present purposes, such bases may include that the subject matter of the proceedings is administrative action by the domestic authorities of the same State, that the proceedings concern property in that State or that there has been prorogation of that jurisdiction. It is thus evident from the outset that an apparent conflict between the contractual jurisdiction clause and the agreement to arbitrate under the treaty is only one instance of the more general phenomenon: that both the investment tribunal and the national court may often consider themselves to have jurisdiction over claims arising from the same facts and legal relationship.
This phenomenon is put in sharper relief where the claim before the investment tribunal arises as a result of breach of substantive obligationsFootnote 14 in a contract containing a forum-selection clause. But the wider point should not be obscured merely because it is this particular species of the conflict that has given rise to the most debate before tribunals and in the academic literature.
The relief may, of course, be all the sharper the more broadly the contractual jurisdiction clause is framed. So, for example, a clause submitting ‘all disputes arising under’ a particular contract may be thought less likely to give rise to a conflict with the jurisdiction of an investment tribunal than one submitting ‘all disputes arising under and related to’ the contract: the latter may be more apt to capture a claim made under an umbrella clause on the basis of an alleged breach of the contract itself. But three further preliminary points call to be made in that respect.
First, the question whether such a conflict arises is, on the hypothesis relevant to this article, to be answered from the perspective of the investment tribunal. This may affect the extent to which the contractual clause and the arbitration agreement conflict; indeed it may mean that a contractual clause which would be viewed by a domestic court or commercial arbitral tribunal as applying to treaty claims may not do so, on the investment tribunal's view, and vice versa.
For example, it is well known that English law construes (commercial) arbitration agreements widely as a matter of policy.Footnote 15 A court in England and Wales will do so provided English law applies to the arbitration agreement, according to the choice-of-law rules recently clarified by the Court of Appeal in Enka Insaat AS v OOO Insurance Company Chubb.Footnote 16 A (commercial) arbitral tribunal applying English law as its ‘curial’ law would take the same approach.Footnote 17 And a court in England and Wales is likely to adopt the same approach to construing a jurisdiction agreement in favour of a given court,Footnote 18 albeit that where the chosen court is one of an EU Member State it may be argued that treaty claims are excluded from the scope of the clause on grounds of foreseeability.Footnote 19
From the perspective of the investment tribunal, however, it cannot be assumed that any of these approaches will apply: the regimes for determining the law applicable to a contractual forum-selection clause are decidedly sparser; and the content of the rules of interpretation that a tribunal will apply is accordingly less certain. Article 42 of the ICSID Convention, for example, provides for the law(s) which the tribunal must apply in ‘decid[ing] a dispute’.Footnote 20 But Article 42 applies only to the substance of the dispute—not to questions of jurisdiction, which appear to fall in a lacuna.Footnote 21 Likewise, Article 35 of the UNCITRAL Arbitration Rules 2013 is expressly limited to the law ‘applicable to the substance of the dispute’. So, in Ruby Roz v Kazakhstan the tribunal was left to have regard to the law of Kazakhstan, the law of England and Wales (as the law of the seat) and potentially also international law as a composite—it may fairly be supposed, in the hope that no contradiction would emerge:Footnote 22
Both Parties have, correctly in the Tribunal's view, argued that the question as to whether or not there is a valid and enforceable agreement to arbitrate is a question of interpreting the relevant provisions of the FIL and the Contract, in good faith and giving full weight to the intentions of the Parties. Consideration has also to be given to Kazakh law—for instance, in so far as concerns the ‘foreign investor’ requirement; and to the law of England, as the law of the seat of the arbitration, agreed by the Parties when these arbitral proceedings began. Insofar as the Tribunal does not believe that any principles of international law would provide jurisdiction, it need not decide whether international law would have some role to play in this arbitration.
So it is understandable that, even in Cambodia Power Company v Cambodia Footnote 23—where consent to ICSID arbitration had (it was argued) been expressed in contracts expressed to be governed by English lawFootnote 24 and where the respondent therefore relied on Fiona Trust, on the (significant) assumption that the arbitration agreements in those contracts must also be governed by English lawFootnote 25—the tribunal determined the scope of its jurisdiction without apparent reference to English law or indeed specific choice-of-law rules at all.Footnote 26
Of course, it is important not to overstate the extent to which conceptual difficulties can obscure clear reality. An investment tribunal is likely to be able to identify an apparent conflict between a contractual forum-selection clause and the arbitration agreement; and often, too, it may be intuitively clear what the correct construction of the clause is. But, absent any established means of identifying which law even governs such an exercise of construction, it may be at best ambitious to expect investment tribunals to determine such conflicts systematically if the issue is framed as one of interpretation of the contractual forum-selection clause as such. Instead, as will be shown, it is possible, proper and preferable for the tribunal to analyse the potential effect of the contractual clause on the arbitration agreement under the treaty, as a matter of international law. On that approach, it can be said that there is a true conflict where, as a matter of international law, the contractual clause precludes the investor's acceptance of the State's offer to arbitrate.
Secondly, prorogation, even by an exclusive jurisdiction agreement, is not necessarily paramount even from the point of view of domestic rules of private international law. By way of a widely applied example, the ‘Brussels’ jurisdictional regime applicable throughout the European UnionFootnote 27 provides many examples of situations in which such an agreement must yield to more fundamental, conflicting rules of jurisdiction.Footnote 28
Thirdly, given both of the preceding points, there is all the less reason for an investment tribunal to assume that a contractual forum-selection clause necessarily precludes the tribunal's own jurisdiction over the dispute or the admissibility of that dispute, even if at first blush the contractual clause might have been construed as encompassing the treaty claim.
This contrasts with the early SGS v Pakistan decision on jurisdiction.Footnote 29 In that case, SGS claimed that a breach by Pakistan of its contractual obligations towards it amounted to a breach of Article 11 of the Switzerland–Pakistan BIT, which obliged Pakistan to ‘guarantee the observance of the commitments it has entered into with respect to the investments of [Swiss] investors’. The tribunal held that interpreting Article 11 along with the jurisdictional provisions of the BIT so as to allow such a claim would ‘set at naught’ the contractual clause by which the parties agreed to submit disputes arising out of or in relation to the contract to commercial arbitration.Footnote 30 By contrast, the subsequent tribunal in SGS v Philippines reached the view that the equivalent umbrella clause in the Swiss–Philippines BIT did encompass contractual obligations.Footnote 31 To that extent the tribunal's view has since become orthodoxy. But the tribunal nonetheless
… agree[d] with the concern that the general provisions of BITs should not, unless clearly expressed to do so, override specific and exclusive dispute settlement arrangements made in the investment contract itself. On the view put forward by SGS it will have become impossible for investors validly to agree to an exclusive jurisdiction clause in their contracts; they will always have the hidden capacity to bring contractual claims to BIT arbitration, even in breach of the contract, and it is hard to believe that this result was contemplated by States in concluding generic investment protection agreements. …Footnote 32
As will be suggested below, however, there is no question of the contractual agreement's being ‘overridden’ in any case nor, contrary to the approach in SGS v Philippines, is it necessary to ward off that spectre by framing the issue as one of admissibility rather than jurisdiction.Footnote 33
B. The Effect of a Contractual Forum-Selection Clause on the Agreement to Arbitrate
In short, then, there is no reason to suppose that the existence of a contractual forum-selection clause precludes absolutely the investment tribunal's jurisdiction or the admissibility of that dispute. The next question is therefore what significance an investment tribunal ought to attribute to such an agreement.
The SGS v Philippines tribunal went on to enunciate ‘[t]he basic principle in each case … that a binding exclusive jurisdiction clause in a contract should be respected, unless overridden by another valid provision’.Footnote 34 As the tribunal in Aguas del Tunari further explained, at least some threshold conditions must exist before the question of any conflict even arises:Footnote 35
Two questions are presented. First, as a threshold matter, the Tribunal observes that in order for the separate document raised by the Respondent [i.e. the putatively conflicting forum-selection agreement] to be in conflict with this Tribunal's exercise of jurisdiction, that document must both deal with the same matters and parties and contain mandatory conflicting obligations. Second, if a true conflict exists, there then arises the question of what effect such a document has on the Tribunal's jurisdiction.
Even with these qualifications in place, it is necessary to add that, at least in non-ICSID arbitration, the tribunal's second step begs yet a further question: whether the putative conflict bears on the tribunal's jurisdiction proper or on the admissibility of the dispute. That distinction is addressed in section IV, below. For the present, and for convenience rather than to beg that question, references to the investment tribunal's ‘jurisdiction’ will include both jurisdiction in the true senseFootnote 36 and the admissibility of the claim.Footnote 37
It is possible that the contractual forum-selection clause has no effect on the agreement to arbitrate. But this article will argue that the analysis advanced in Aguas del Tunari is fundamentally sound and that, for the reasons elaborated below, ‘the question of whether a conflicting mandatory [forum-selection] obligation in a separate document can affect the jurisdiction of an ICSID tribunal is a question of the intent of the Parties in concluding the separate document’.Footnote 38
The Aguas del Tunari tribunal suggested that it was
particularly helpful in such an inquiry to distinguish between: (1) a separate document that waives the right to invoke, or modifies the extent of, ICSID jurisdiction (where the intent of the parties to alter the possibility of ICSID jurisdiction is direct); and, (2) a separate document that contains an exclusive forum selection clause designating a forum other than ICSID (where the intent of the parties to alter the possibility of ICSID jurisdiction must be implied).Footnote 39
This analysis is apt to apply to agreements to arbitrate before an investment tribunal more generally—not only before an ICSID tribunal—and continues to find traction even outside that specific context.Footnote 40 As will be seen below, however, once the necessary conceptual foundations are in place the importance of the twofold distinction drawn by the tribunal falls away.Footnote 41
Taking yet a further step back shows why. The agreement to arbitrate, which is formed by the investor's acceptance of the host State's standing offer to arbitrate, is governed by international law.Footnote 42 It may therefore be formed only in a manner permitted by international law. In principle, the forum-selection agreement—whether in the form of a waiver or of a forum-selection clause—could bear on the parties’ intent in two ways: first, it could modify or rescind the host State's offer to arbitrate or, secondly, it could modify or preclude the investor's eventual acceptance of that offer. Logically, therefore, the prior question is which of the parties’ intent it affects, in any event as a matter of international law.
This is an important point of principle with ramifications in public international law more generally. The host State's offer to arbitrateFootnote 43 may perhaps most naturally be characterized as the grant of a right to the investor as a third party to the treaty.Footnote 44 This requires recognition of the investor's procedural right to make a claim to arise directly under the treaty. But if proof is needed that that is correct it can be taken in two steps: first, it is settled that treaties can grant rights directly to individuals (and not only in the context of human rights in the strict sense);Footnote 45 secondly, whereas in some cases there may be doubt whether the individual right is susceptible of an individual claim (rather than one made by diplomatic protection),Footnote 46 it would be logically incoherent—ie would defeat the object and purpose of the treaty's arbitration provisions—if the right to bring a claim before a tribunal without the assistance of the State of the investor's nationality could be exercised only through diplomatic protection.Footnote 47
The next question is therefore whether the contractual forum-selection clause operates on the State's offer of treaty arbitration, the investor's right to accept that offer, or both. The next section of this article explains why a focus on the investor's right to accept is natural and effective. The remainder of this section explains why beginning the analysis with the State's offer is apt to lead to the same place in any event.
International law is not yet well developed on the question whether, and if so when, a State may revoke a right granted to an individual third party. As to State third parties, the Vienna Convention on the Law of Treaties, Article 37(2), provides, ‘When a right has arisen for a third State in conformity with article 36, the right may not be revoked or modified by the parties if it is established that the right was intended not to be revocable or subject to modification without the consent of the third State.’Footnote 48 That appears to provide a reasonable guide to the circumstances in which a right granted to an individual might also be revoked. There is no obvious reason of principle why a different analysis should apply to third-party States and individuals in this context: one might expect there to be policy reasons why individual rights should in some cases be less readily considered revocable, such as with human-rights treaties; but such cases can be accommodated by assuming that, for example, human rights are not intended to be revocable without the individual's consent.Footnote 49
At any rate, for the sake of argument it can be assumed to be possible in principle that a contractual forum-selection clause could be viewed as a revocation of a particular investor's right to arbitrate. But that view is likely to present both conceptual and textual difficulties. A conceptual difficulty of analysing the contractual-forum selection clause as one of revocation is that, though the direct beneficiary of the offer to arbitrate is the investor, it is nonetheless part of the bargain on which the State of the investor's nationality relies in concluding the treaty.Footnote 50 So it may be difficult to establish an intention that the right to accept the offer of treaty arbitration should be revocable by the act of the host State. This shades into the textual problems, which will depend on the particular BIT. But it may be a rare treaty that expressly contemplates revocation of the right to arbitrate.Footnote 51 Indeed Oppenheim suggests that ‘an intention [that the right may not be revoked … without the consent of the (beneficiary)] is probably to be implied … in the case of those treaties which establish a status or regime valid erga omnes’.Footnote 52 It may be reasonably supposed that much the same analysis applies whether the offer to arbitrate is erga omnes or to the class of investors.
Absent express provision for the host State to revoke the right to arbitrate, the question therefore in effect resolves into whether the investor has consented to give it up.Footnote 53 The next section addresses how the investor may do so.
III. HOW CONTRACTUAL FORUM-SELECTION AGREEMENTS MAY BE RELEVANT
For the reasons given in the previous section, the focus lands on the investor's right to accept the offer of treaty arbitration. This right belongs to the investor alone and not to the State of its nationality. It is important to specify this because the point has caused previous analyses to veer from the logical path. The tribunal in SGS v Philippines, for instance said:Footnote 54
It is, to say the least, doubtful that a private party can by contract waive rights or dispense with the performance of obligations imposed on the States parties to those treaties under international law. Although under modern international law, treaties may confer rights, substantive and procedural, on individuals, they will normally do so in order to achieve some public interest. Thus the question is not whether the Tribunal has jurisdiction: unless otherwise expressly provided, treaty jurisdiction is not abrogated by contract. …
But it is difficult to see what right of the State is engaged. Even if it is conceived of as some right to allow its investors to go to arbitration, such a right is still parasitic on the first-order right of the investor itself that is truly in issue.
Though it belongs to the individual investor, the right arises in the framework of international law,Footnote 55 as shown above. So it would be incongruous (not to mention difficult, inconsistent and unpredictable) to attempt to subject it to the equivalent doctrines of any national system of law.
Waiver is the most natural lens through which international law may view this issue. That has the additional virtue of subsuming the answer to the alternative question posed above, whether the host State has validly revoked the investor's right to arbitrate and, crucially, whether the investor has consented to such revocation.
Indeed, apart from the present issue, there is a long history of ‘fork-in-the-road’ clauses in investment treaties, by which the investor must elect between arbitration under the treaty and litigation before national courts (or perhaps non-treaty arbitration), the choice of one then precluding the investor's future acceptance of the other.Footnote 56 Perhaps the most notable instance, or at least potential instance, of such a clause is now found in Article 26 of the Energy Charter Treaty.Footnote 57 That Article permits the investor to submit disputes for resolution to, among others, ICSID arbitrationFootnote 58 or otherwise ‘in accordance with any applicable, previously agreed dispute settlement procedure’.Footnote 59 But the treaty allows States parties to limit their consent to the former where an investor has already availed itself of the latter.Footnote 60
It is difficult to see that such provisions, well-established to be effective,Footnote 61 necessarily imply anything other than both (i) the right of the individual investor to accept the host State's offer to arbitrate and (ii) the possibility that such right might be waived. Indeed the more recent USMCA and Comprehensive and Progressive Agreement for Trans-Pacific Partnership describe their equivalent provisions as requiring ‘waivers’ of dispute-settlement procedures other than treaty arbitration.Footnote 62
And, as will be seen presently, a moderate body of arbitral jurisprudence has indeed adopted the view that it is as a waiver of the right to come before investment arbitration that a contractual forum-selection clause may likewise preclude the investor's acceptance of the host State's offer to arbitrate.Footnote 63 Given that, this section deals first with the analysis of contract-as-waiver (which does prove convincing) before dealing with certain other analyses that have been put forward (but which prove less convincing).
A. Waiver
If this approach is adopted, then all that is necessary is to assess in each case whether the contractual forum-selection clause expresses a waiver of the investor's right to recourse to an investment tribunal. As a matter of general international law, that comes down to the essential question whether the investor's intention to do so ‘clearly results from’ the investor's entry into the forum-selection clause.Footnote 64 This is an attractive, accurate and flexible approach: it focuses directly on the contractual clause at issue in a fact-specific way; and the answer to the question bears directly on the investor's ability to accept the host State's offer, which as identified above is precisely the relevant issue.
But the approach of many investment tribunals belies both the importance and the simplicity of this framework for analysis, taking a more circuitous and less convincing route than would have been possible merely by posing the question identified above.
The Lanco arbitrationFootnote 65 was one of the earliest to address this subject. Argentina relied on a clause in a concession agreement providing:
For all purposes derived from the agreement and the BID CONDITIONS, the parties agree to the jurisdiction of the Federal Contentious-Administrative Tribunals of the Federal Capital of the ARGENTINE REPUBLIC.Footnote 66
On the facts of that case, it was hardly necessary for the tribunal to identify any general statement of principle: as only the shareholder of the party to the concession contract, the claimant was not bound by the clause and therefore difficult to identify as the party expressing a waiver and the forum it identified was one already available to investors by virtue of Article VII(1) of the BIT. So, applying the approach above, it would have been difficult to argue that a waiver of recourse to the investment tribunal clearly resulted from such a clause. But the tribunal went further anyway and observed that it was ‘highly questionable’ whether it was even possible that such a clause could deprive an investment tribunal of jurisdiction, apparently on the basis that, ‘The alleged selection of jurisdiction, as a possibility that may be envisaged by Argentine law, ha[d] not been shown by the Respondent.’Footnote 67 Whether or not that was a correct assessment of the evidence, it seems to sidestep unnecessarily the real question identified above.
Interestingly, the forum-selection clause at issue in the subsequent Azurix Footnote 68 award had, according to Argentina,Footnote 69 been inserted into the concession contract precisely to avoid the result in Lanco. Clause 1.5.5 of the relevant terms provided for the exclusive jurisdiction of the courts for contentious-administrative matters of the city of La Plata ‘for all disputes that [might] arise out of the Bidding, waiving any other forum, jurisdiction or immunity that may correspond’.Footnote 70 The tribunal did aim its analysis at the question of waiver (though it may be questioned how far it grappled with the potentially difficult question whether a claim under the BIT's umbrella clause could be said to ‘arise out of the Bidding’, focusing instead on whether the claims before it were brought directly on the contract or the treaty);Footnote 71 but there was in the end the short answer that once again the clause had been entered into by a subsidiary of the claimant and so was not a waiver by the claimant itself.Footnote 72
The baton was soon picked up, however, by the tribunal in Aguas del Tunari. That tribunal framed the issue as being whether the ‘separate document’ (usually the contractual forum-selection agreement) could be construed as disclosing an intention to waive recourse to investment arbitration:Footnote 73
As to the … case of a separate document containing an exclusive forum selection clause that designates a forum other than ICSID, the Tribunal notes that the specific intent of the parties to preclude ICSID jurisdiction will be more difficult to ascertain than in the case of explicit waiver. The Tribunal is of the view that it is not the existence of the exclusive forum selection clause that would be given effect by an ICSID tribunal, but rather that the tribunal could, at most, give effect to a waiver implied from the existence of an exclusive forum selection clause. … A separate conflicting document should be held to affect the jurisdiction of an ICSID tribunal only if it clearly is intended to modify the jurisdiction otherwise granted to ICSID. As stated above, an explicit waiver by an investor of its rights to invoke the jurisdiction of ICSID pursuant to a BIT could affect the jurisdiction of an ICSID tribunal. However, the Tribunal will not imply a waiver or modification of ICSID jurisdiction without specific indications of the common intention of the Parties.
That tribunal's focus on express versus implied waivers is perhaps misplaced for three reasons: as Azurix demonstrates, whether a putative waiver is express or implied does not necessarily depend on whether it is in a stand-alone ‘waiver’ document or a contractual forum-selection clause; there is no obvious need for the distinction to be drawn within the framework set out above as a matter of principle.
Moreover, for the reasons given above, the reference to the ‘parties’ may in practice generally be narrowed to ‘the claimant’. That is, first, the question of loss of the right to invoke the tribunal's jurisdiction is best analysed as one of waiver. Secondly, and relatedly, since the issue is not as such what the parties’ rights are under the law governing the contractual clause but rather the characterization of the investor's act in international law, the question of common intention is anyway not of direct relevance. Thirdly, there is a more nuanced point: though at one level the proposition that a bilateral document must be construed according to the parties’ joint intention may seem obviously attractive, for the reasons given in Section II.A such a construction may ultimately prove to be a chimaera. Happily, though, that poses no problem of principle if the issue is framed as being whether the contractual clause discloses a waiver by the investor as a matter of international law, rather than what its strict effect is according to its governing law. And indeed it may be that the distinction of principle matters in only extreme cases.
At any rate, the tribunal's statement of the correct approach to contractual forum-selection clauses perhaps remains the clearest statement of principle on this point (albeit in the context of ICSID arbitration but readily applied to investment arbitration generally). The tribunal also spelled outFootnote 74 what was a corollary of the ICJ's approach in the Temple case: that it would ‘not read an ambiguous clause as an implicit waiver of ICSID jurisdiction; silence as to the question is not sufficient’. That was right enoughFootnote 75 and has been adopted and expanded upon by some subsequent tribunals, notably in TSA v Argentina Footnote 76 and in Crystallex v Venezuela.Footnote 77
But the subsequent tribunal in SGS v Paraguay, while agreeing with the approach in Aguas del Tunari,Footnote 78 took that truism too far, claiming that, ‘At least in the absence of an express waiver, a contractual forum selection clause should not be permitted to override the jurisdiction to hear Treaty claims of a tribunal constituted under that Treaty.’Footnote 79 That would be to convert a pragmatic presumption against waivers implied by silence (as in Aguas del Tunari) into a denial of the possibility of any implied waiver whatsoever, which is plainly wrong. So, too, as recently as the award in MNSS v Montenegro, the tribunal added a similar unwarranted gloss, holding that ‘investors may waive the rights conferred to them by treaty provided waivers are explicit and freely entered into by investors’.Footnote 80
It may be suggested that the addition of such glosses is the result of a deliberate policy, hardening the requirement for waiver in the way that the European Court of Human Rights has done.Footnote 81 If that is the case, it is at least not clear from these awards. And there may anyway be less justification for such a policy in a context where, though important rights in international law are engaged, they are not necessarily of the same order of importance as those under, for example, the European Convention.
What is called for is a close analysis of the individual forum-selection agreement in each case to determine whether a waiver of treaty claims ‘clearly results from it’ in the sense that it was ‘clearly intended to modify the jurisdiction otherwise granted to’ the investment tribunal. There are awards in which just such an analysis has been conducted, of which Crystallex is again a notable exampleFootnote 82 as is Occidental v Ecuador.Footnote 83 It may also then be necessary to determine precisely how far the putative waiver extends: does it, for instance, express an intention to renounce the option of invoking an investment tribunal's jurisdiction at all or does it merely require initial recourse to the contractually-chosen forum in the first instance? The answer must depend on a close analysis of the facts and above all the language used in that context.
But tribunals are far from uniform in their approach and there are at least as many—even recent—awards in which the analysis is less searching. In Gavrilović v Croatia, for example, Croatia invoked a contractual jurisdiction clause in favour of the courts of Zagreb in the purchase agreementFootnote 84 by which the investment had been obtained, claiming that it precluded the tribunal's jurisdiction (and, implicitly, was a waiver of the claimants’ right to arbitrate). In a rather terse passage, the tribunal referred to the now-orthodoxy that a claim under an umbrella clause is a separate cause of action from a claim brought on the basis of the contract itself, before concluding:Footnote 85
While a contractual forum selection clause may refer contract disputes to another forum that will decide whether a breach of contract occurred, with the consequences that may follow under the applicable law, this Tribunal must decide whether or not contractual obligations have been observed and, as a consequence, whether or not there has been a violation of the umbrella clause. The Tribunal would not fulfil its mandate if it refused to do so.
This analysis was expressly followed in the recent award in Belenergia v Italy Footnote 86 but it begs the question: it is, of course, correct that if the tribunal has jurisdiction over the claim then (subject to issues of admissibility) it should ‘fulfil its mandate’; but if it has no jurisdiction then it has no mandate to fulfil. It may be that the tribunal meant to take it as read that a contractual forum-selection clause was incapable of affecting its jurisdictionFootnote 87 and to issue a strong statement of principle that it did not consider that the contractual clause had any bearing on the admissibility of the dispute. But either way one might have expected the tribunal to undertake an analysis of the actual clause at issue in the case.
B. Other Approaches
1. ‘The “fundamental basis” of the claim’ approach
As noted above, the approach in Gavrilović was ostensibly informedFootnote 88 by the strand of jurisprudence addressing the question whether a claim brought under a treaty's umbrella clause alleging breach of a contract was a ‘contractual’ or a ‘treaty’ claim. But that debate is now of some vintage,Footnote 89 having arguably been settled by the ad hoc committee in Vivendi Footnote 90 in terms perhaps most neatly recapitulated by the tribunal in SGS v Paraguay:Footnote 91
… It has been argued that, if the umbrella clause violation is premised on a failure to observe a contractual commitment, one cannot say (in the Vivendi I annulment committee's words) that the ‘“fundamental basis of the claim” is a treaty laying down an independent standard by which the conduct of the parties is to be judged’—because, for that type of umbrella clause claim, the treaty applies no legal standard that is independent of the contract. But that argument ignores the source in the treaty of the State's claimed obligation to abide by its commitments, contractual or otherwise. Even if the alleged breach of the treaty obligation depends upon a showing that a contract or other qualifying commitment has been breached, the source of the obligation cited by the claimant, and hence the source of the claim, remains the treaty itself.
This approach, of determining the effect of the contractual forum-selection clause by asking whether the ‘fundamental basis of the claim’ is the contract or the treaty, has been widely applied.Footnote 92 As noted above there is no longer real room for doubt that it is the correct analysis of the nature of a substantive breach of an umbrella clause.
But it provides an incomplete answer to the question of the relationship between a contractual forum-selection clause and an investment arbitration agreement. It is true that the basis (fundamental or otherwise) of the claim at issue must be relevant to the detailed analysis whether it answers to the categories of claim which the investor has putatively waived its right to pursue before an investment tribunal;Footnote 93 in other words, if the claim is properly characterized as a treaty claim then there is a conflict only if a waiver expressed by the contractual forum-selection clause also extends to treaty claims. But it is not enough to answer the question by reference to the nature of the claim alone without considering the scope of the clause (and putative waiver), which is the other half of the equation.Footnote 94
The point may be illustrated quite simply. Suppose that the investor and State have agreed expressly and explicitly by contract (and, if required, for valuable consideration) that the investor will submit all disputes arising under or related to the performance of the contract to the courts of the host State; that it waives any claims it may have arising under any treaty for the protection of investors between the State of its nationality and the host State; that it will waive any equivalent claims arising under customary international law and, for good measure, that it will pursue no claims before any tribunal constituted under any treaty of the type mentioned.Footnote 95 Then suppose the investor purports to accept the host State's offer to arbitrate and claims that the State has breached its duty of fair and equitable treatment in a way somewhat related to the performance of the contract. If the analysis stops with the essential basis of the claim then the answer is clear: the duty of fair and equitable treatment is purely a creature of international law with no basis in contract so the investment tribunal has jurisdiction. But that defies common sense and fails to deal with the forceful point made by the cases on waiver, that one must still ask whether the investor's purported exercise of its right remains good in view of what it has agreed with the host State.
One possible, indirect answer to this point was suggested by the tribunal in BIVAC v Paraguay, which said that if, conversely, the fundamental basis of the claim is contractual then the investor is bound by the whole contract, including its forum-selection provisions:Footnote 96
The broader point, however, having regard to the fundamental principle that the autonomy and will of the parties is to be respected, is that the parties to a contract are not free to pick and choose those parts of the Contract that they may wish to incorporate into an ‘umbrella clause’ provision such as Article 3(4) and to ignore others. The obligation that Paraguay entered into with BIVAC was to pay its invoices in a timely way and if it failed to do so to allow any contractual dispute to go to the Tribunals of the City of Asunción. To allow BIVAC to choose those obligations it wished to incorporate into the BIT and to ignore others would seriously and negatively undermine contractual autonomy. If the parties to a contract have freely entered into commitments, they must respect those commitments, and they are entitled to expect that others, including international courts and tribunals, also respect them, unless there are powerful reasons for not doing so. BIVAC has not identified any reasons at all.
Apart from anything, this proves too much: if the investor's claim is indeed brought under a contractFootnote 97 and not under the treaty then absent another basis of jurisdiction the investment tribunal has no jurisdiction to begin withFootnote 98 and the question of the contractual forum-selection clause does not even arise. But it also involves a serious misunderstanding of the question posed by the umbrella clause in the first place, which is simply whether the State has breached its contractual duty to the investor. Unless the investor's duty to submit disputes to the contractually chosen forum is a condition precedent to the host State's substantive contractual duty, the contractual choice of forum has no bearing on the question posed by the umbrella clause. At any rate, though such a condition precedent analysis is possible in theory it would be most outlandish to encounter it in practice. And it is all the harder to imagine its working where it entails an obligation on the investor to pursue a dispute that ex hypothesi has not yet arisen. BIVAC's suggested counterargument therefore seems unlikely to be correct.Footnote 99
2. ‘The Earlier (or Later) Agreement Prevails’ Approach etc
An alternative, much simpler approach has also found some support: that where a contractual forum-selection agreement and the agreement to arbitrate conflict, the earlier in time should prevail. An equally simple approach also has its supporters: that in case of such a conflict the later agreement should prevail.
In support of the former view, Professor Douglas suggests a useful framework for the analysis:Footnote 100
If a relationship of coordination between the investment treaty regime, the municipal legal system of the host state and multilateral conventions on choice of forum agreements is to be achieved, then the general principles of generalia specialibus non derogant, prior tempore potior jure and pacta sunt servanda provide the doctrinal basis for sorting out conflicts between overlapping jurisdictions.
It is immediately obvious that the problem cannot be resolved by the third principle alone: two agreements are to be kept, which is precisely the problem. If the stalemate is to be broken then it must be, on this view, by one of the other two principles. The second works better as a matter of LatinFootnote 101 than of law: it may be thought unusual in most legal systems for parties to an agreement to find that they are incapable of varying it by reaching a subsequent agreement and that is no less the case in international law, where the VCLT, Article 30(3), provides that, ‘When all the parties to the earlier treaty are parties also to the later treaty but the earlier treaty is not terminated or suspended in operation under article 59, the earlier treaty applies only to the extent that its provisions are compatible with those of the later treaty.’Footnote 102
That is not to say, however, that timing is always irrelevant. The tribunal in BIVAC said:Footnote 103
In considering the effect of Article 9 it is appropriate to consider its relationship in time to the BIT. The BIT came first. It was concluded in The Hague on 29 October 1992, and came into effect shortly afterwards, on 1 August 1994. The Contract between BIVAC and the Ministry of Finance was executed on 6 May 1996, that is nearly two years after the BIT came into force. No evidence has been tendered by the parties as to the relevance of the BIT to BIVAC's decision to enter into the Contract. Paraguay, and its Ministry of Finance, must be taken to have knowledge of the treaties that it has entered into, and it would be surprising if those responsible for advising BIVAC on the legal aspects of the Contract were not aware of the BIT and the protections it provided, including the potential availability of ICSID arbitration for certain treaty disputes. Knowledge of the BIT was not, however, a subject to which arguments or evidence have been addressed by the parties. Nevertheless, the parties agreed to include in the Contract a provision which provides for the exclusive jurisdiction of the Tribunals of the City of Asunción to settle ‘Any conflict, controversy or claim which arises from or is produced in relation to this Contract, non compliance, resolution or invalidity’. The parties could have included a provision in Article 9(1) to the effect that the obligations it imposed were without prejudice to any rights under the BIT, including the possible exercise of jurisdiction by an ICSID Tribunal to any matters arising under the Contract which could fall to be determined under Article 3(4) of the BIT. The fact that they did not do so is not without relevance: it indicates, at the very least, that the parties to the Contract, including BIVAC, intended the exclusive contractual jurisdiction of the Tribunals of the City of Asunción to be absolute and without exception, and for it to mean what it says.
But the force of these observations does not depend on an analysis of which of the conflicting agreements came first. First, as observed above, the conflict is not between the contractual forum-selection clause and the investment treaty itself but between the putative waiver expressed by the contractual clause and the agreement to arbitrate formed under the treaty. Secondly, though it is not the significance the tribunal intended to give them, the substance of its remarks tends to show the usefulness of the waiver analysis suggested here: the real relevance of the timing was that, assuming BIVAC to have known about the BIT's existence (and therefore to have had it in mind), it nonetheless agreed to submit any disputes ‘produced in relation to’ its contract to the courts of the City of Asunción. The implicit question is therefore just whether an intention to waive recourse to investment arbitration clearly resulted from BIVAC's agreement to that clause. One may dispute what the answer should be; but it is harder to dispute the good sense of the question.
The tribunal in SGS v Paraguay suggested that BIVAC had the logic in reverse:Footnote 104
The BIVAC tribunal reasoned that because the claimant's contract post-dated the BIT, it should take precedence: ‘[t]he parties could have included a provision in [the forum selection clause] to the effect that the obligations it imposed were without prejudice to any rights under the BIT, including the possible exercise of jurisdiction by’ a treaty tribunal under the umbrella clause. While the same sequence is in play here—the Switzerland–Paraguay BIT entered into force in 1992, while the Contract was concluded in 1996—we would reverse the presumption. Given the significance of investors’ rights under the Treaty, and of the international law ‘safety net’ of protections that they are meant to provide separate from and supplementary to domestic law regimes, they should not lightly be assumed to have been waived. Assuming arguendo that the parties to the later-in-time Contract could have expressly excluded the right to resort to arbitration under the extant BIT, at least as to Contract-based claims under Article 11, they did not do so—and we would not take their silence as effecting that same waiver of Treaty rights.
It may be doubted whether either tribunal was correct to suggest, if that was the intention in SGS v Paraguay, that there ought to be any evidential presumption one way or the other, however. Arguments of common sense or policy may be made for a presumption in either direction, as both awards show. But it is difficult to see what such a presumption adds: unless the presumption is to be irrebuttable (which neither tribunal seemed to suggest) a fact-sensitive analysis must still be undertaken and it is therefore hard to see what is added even from the perspective of convenience of proof.
Those awards were focused on the evidential significance of the timing. Parra, on the other hand, argues that the subsequent agreement must prevail as a matter of law:Footnote 105
In most cases, the consent in the BIT of the host State to the submission of the investment disputes to arbitration can also be invoked in preference to any applicable previous agreement on the settlement of such disputes, such as might be embodied in the arbitration clause of an investment contract between the investor and the host State. The consent or ‘offer’ of the host State to submit to arbitration in the BIT, when accepted by the covered investor, simply supersedes their previous agreement to the extent of the overlap between that agreement and the new one formed by the offer in the BIT and its acceptance by the investor.
This, however, begs the question whether the investor retained the right to accept the host State's offer at the time it putatively did so. It is suggested here that that is the key question and that is most sensibly addressed within the framework of waiver.
That leaves the first principle identified by Professor Douglas, the application of which also finds support in the jurisprudence.Footnote 106 But the problem at hand arises only if both the contractual forum-selection clause and the investment arbitration agreement encompass the claim in question; and experience suggests that they will do so, all but invariably, not because either is specifically drafted but because of their breadth. Moreover, it again assumes that there is a subsequent, arbitration agreement, whereas the point made by the awards dealing with waiver is that that cannot be assumed. Instead, the question of ‘specificity’ to be asked is simply whether the dispute falls within the scope of the contractual forum-selection clause and, if so, whether an intention to waive recourse to investment arbitration clearly results from the investor's entry into that clause.
IV. JURISDICTION OR ADMISSIBILITY
Assuming the foregoing analysis to be accepted and that there has in a given case been a waiver of the investor's right to accept the host State's offer to arbitrate, the final question is whether that is a matter of jurisdiction or admissibility.
A. Whether Waiver a Matter of Jurisdiction or Admissibility
It is useful to begin this next stage of the analysis by recalling the investment tribunal's own fundamental requirement for jurisdiction: that each party consent to the arbitration.Footnote 107 Seen in this light, the conclusion is inescapable that, if the investor has lost the right to consent to arbitrate then that must vitiate the tribunal's jurisdiction, as recognized by the tribunal in Aguas del Tunari Footnote 108 and as is implicit in the awards in Occidental v Ecuador,Footnote 109TSA v Argentina Footnote 110 and Crystallex v Venezuela.Footnote 111
The issue has nonetheless been addressed under the rubric of admissibility. One may be cautious about criticising tribunals for doing so given that that is often the way in which the matter is presented by the parties, as it was in Gavrilović Footnote 112 and in SGS v Paraguay. Footnote 113 But it is fair to observe that the erroneous treatment of this issue under the rubric of admissibility appears to be allied to the approach, deprecated above, of focusing on the ‘fundamental basis of the claim’ without also having due regard to the scope of the contractual forum-selection clause. Gavrilović is arguably such a case for the reasons given above. So, more obviously, is SGS v Philippines, where the tribunal expressly framed the matter as one of admissibility on its own analysis (following an analysis of the claims brought under the umbrella clause and the contract itself):Footnote 114
In the Tribunal's view, this principle is one concerning the admissibility of the claim, not jurisdiction in the strict sense. The jurisdiction of the Tribunal is determined by the combination of the BIT and the ICSID Convention. It is, to say the least, doubtful that a private party can by contract waive rights or dispense with the performance of obligations imposed on the States parties to those treaties under international law. Although under modern international law, treaties may confer rights, substantive and procedural, on individuals, they will normally do so in order to achieve some public interest. Thus the question is not whether the Tribunal has jurisdiction: unless otherwise expressly provided, treaty jurisdiction is not abrogated by contract. The question is whether a party should be allowed to rely on a contract as the basis of its claim when the contract itself refers that claim exclusively to another forum. In the Tribunal's view the answer is that it should not be allowed to do so, unless there are good reasons, such as force majeure, preventing the claimant from complying with its contract. This impediment, based as it is on the principle that a party to a contract cannot claim on that contract without itself complying with it, is more naturally considered as a matter of admissibility than jurisdiction.
Another reason for the issue's treatment as one of admissibility rather than jurisdiction may be the imperfect analogy with mixed-claims commissions,Footnote 115 which can be traced to the ad hoc committee's influential decision in Vivendi I (where the claim was not under an umbrella clause). The committee said:Footnote 116
In a case where the essential basis of a claim brought before an international tribunal is a breach of contract, the tribunal will give effect to any valid choice of forum clause in the contract. For example in the Woodruff case, a decision of an American–Venezuelan Mixed Commission in 1903, a claim was brought for breach of a contract which contained the following clause:
Doubts and controversies which at any time might occur in virtue of the present agreement shall be decided by the common laws and ordinary tribunals of Venezuela, and they shall never be, as well as neither the decision which shall be pronounced upon them, nor anything relating to the agreement, the subject of international reclamation.
But the analogy, drawn in eg SGS v Philippines, is apt to mislead in the context of this issue. In those cases, the commissions’ jurisdiction was founded solely on an interstate agreement, so that the consensual basis of the commission's jurisdiction was beyond doubt. Consequently, the only relevance of the contractual choice of forum could have been to admissibility.Footnote 117
That is not to say that the same facts relevant to waiver may not also give rise to an issue of admissibility before the tribunal, such as that the investor's claim is brought in bad faith or with unclean hands. But, though it is unwise to attempt any general statement, it may be thought an exceptional case where a claim would be held inadmissible on such grounds when ex hypothesi there had been no waiver of the claim.
Possible, though more controversial, is the proposition that the concurrence of contractual and treaty jurisdictions by itself justifies a stay of the latter on pragmatic grounds even where no true waiver is present. As ever, one should be slow to suggest a general rule for fact-sensitive situations; but the admonition of the tribunal in Impregilo v Pakistan may be expected to resonate:Footnote 118
Further, if a stay was ordered, as Pakistan has sought, it is unclear for how long this should be maintained; what precise events might trigger its cessation; and what attitude this Tribunal ought then to take on a resumed hearing to any proceedings or findings that may have occurred in the interim in Lahore.
B. Significance of the Distinction
The questions of whether and, if so, how an investment tribunal ought to give effect to a contractual forum-selection clause remain of often decisive importance in the arbitration itself. This section asks what significance attaches to their classification as questions of jurisdiction or of admissibility.
Analysis on this point appears to have been sparse but the distinction is likely to be significant above all in the context of the New York Convention,Footnote 119 which is the mechanism by which successful parties in the vast majority of arbitrations in general (and some investment arbitrations) will seek recognition or enforcement of the arbitral award. That Convention plainly enables the other party to apply for refusal of recognition or enforcement where the tribunal had no jurisdiction under international law.Footnote 120 But wrong decisions as to the admissibility of the claim appear apt to permit refusal only where they are such that ‘[t]he subject matter of the difference is not capable of settlement by arbitration under the law of’ the country where recognition or enforcement is sought or that recognition or enforcement would be contrary to its public policy.Footnote 121 In other words, if the issue goes to jurisdiction then the award will be open to challenge in national courts wherever recognition or enforcement is sought, whereas if the issue goes to admissibility the matter will likely be determined by the tribunal once and for all.
There are two reasons for this. First, it is difficult to see at a purely textual level that the existence of the contractual forum-selection clause is an aspect of ‘the subject matter of the difference’. Secondly, as a matter of policy the trend has been to construe these exceptions narrowly, as exemplified by the judgment of the Supreme Court of Canada in Desputeaux v Éditions Choutte (1987) Inc, addressing the issue in the context of an award on copyright issues:Footnote 122
… a matter may be excluded from the field covered by arbitration because it is by nature a ‘matter of public order’. The concept also applies in order to define and, on occasion, restrict the scope of legal action that may be undertaken by individuals, or of contractual liberty. … The development and application of the concept of public order allows for a considerable amount of judicial discretion in defining the fundamental values and principles of a legal system. In interpreting and applying this concept in the realm of consensual arbitration, we must therefore have regard to the legislative policy that accepts this form of dispute resolution and even seeks to promote its expansion. For that reason, in order to preserve decision–making autonomy within the arbitration system, it is important that we avoid extensive application of the concept by the courts. Such wide reliance on public order in the realm of arbitration would jeopardize that autonomy, contrary to the clear legislative approach and the judicial policy based on it.
The significance between jurisdictional and admissibility matters is different—and rather less clear—where awards are made under the ICSID Convention, as many investment arbitration awards are.Footnote 123 Most obviously, Article 54 of the ICSID Convention provides that contracting States must recognize an award made under the Convention and enforce pecuniary obligations arising under such an award ‘as if it were a final judgment of a court in that State’. So no review is permitted of the award at the stage of recognition and enforcement that would allow the present issue to be revisited, whether it is classified as one of jurisdiction or of admissibility.
But it is also so because the distinction between admissibility and jurisdiction is much less clearly demarcated in the context of ICSID arbitration. This is exemplified by the remarks of the tribunal in Pan American Energy v Argentina, which was content to decide ‘the six objections raised by the Respondent’ without, it considered, ‘need[ing] to go into the possible—and somewhat controversial—distinction between jurisdiction and admissibility’.Footnote 124 Among other reasons for the obscurity of the basis and extent of this distinction before ICSID tribunals may be that Article 41(1) of the ICSID Arbitration Rules refers in English to the ‘jurisdiction of the Center’Footnote 125 and the ‘competence of the tribunal’ while the French text refers in both cases to the ‘compétence’ of each, neither leaving any obvious—at least on the text—basis for objections to admissibility to be taken. Nonetheless, as we have seen, some tribunals have explicitly addressed contractual forum-selection clauses under the rubric of admissibilityFootnote 126 and it has been argued that tribunals are entitled to take that and any other point as to admissibility as an exercise of their inherent powers.Footnote 127
The ICSID Convention does, however, allow for review by way of an application for annulment.Footnote 128 The grounds for annulment are ‘that the Tribunal was not properly constituted’, ‘that the Tribunal has manifestly exceeded its powers’, ‘that there was corruption on the part of a member of the Tribunal’, ‘that there has been a serious departure from a fundamental rule of procedure’ and that ‘the award has failed to state the reasons on which it is based’. Manifest excess of powers is naturally apt to capture a (manifestly) wrong decision on the tribunal's jurisdiction. But it is perhaps less sure to capture a wrong decision on the admissibility of the dispute. Ad hoc committees have generally followedFootnote 129 the view of the ad hoc committee in Vivendi I that, ‘It is settled that an ICSID tribunal commits an excess of powers not only if it exercises a jurisdiction which it does not have … but also if it fails to exercise a jurisdiction which it possesses.’Footnote 130 It remains unclear whether a decision that the claim itself is inadmissible is a ‘failure to exercise jurisdiction’ in the necessary sense.Footnote 131
If, however, the analysis suggested here is correct then a happy consequence is that the difficulty of determining whether ICSID arbitration allows for objections to admissibility may await resolution for another day: the effect of a competing contractual forum-selection clause is squarely on the Center's jurisdiction, the tribunal's competence, and indeed the compétence of both.
V. CONCLUSION
This article has been aimed at showing that due attention to first principles provides a clear and useful framework within which to approach what has proved a long-standing and at times intractable problem of some significance. The question whether a contractual forum-selection clause ought to prevail over an agreement to arbitrate under an investment treaty is answered by assessing in each case whether the contractual forum-selection clause amounts to a waiver of the investor's right to recourse to an investment tribunal. As a matter of general international law, that comes down to the essential question whether the investor's intention to do so ‘clearly results from’ the investor's entry into the forum-selection clause. Moreover, it follows from such an analysis that the effect of a conflicting contractual forum-selection clause must be on the tribunal's jurisdiction rather than, in general, on the admissibility of the claim.
While this framework may require a close examination of the facts of the case at hand, it is suggested that the analytical basis of that examination need no longer be obscure and that, consequently, the scope of that examination and its relationship to other difficult issues arising in the context of investment treaty arbitration may likewise be clarified.
