Introduction

Surviving means adapting. As investment treaties face increased resistance from all quarters, especially in the shadow of climate change, the existence of treaty-based ISDS as we know it has also been brought under suspicion. In recent years, the EU and the UK notified their exit from the ECT, notwithstanding the treaty’s “modernisation.” The Ecuadorian Constitutional Court found a treaty provision providing foreign investors access to ICSID arbitration unconstitutional for violating Article 422 of the Ecuadorian constitution. Ecuadorian citizens also overwhelmingly voted to retain Article 422 in their constitution, thus effectively ruling out a return to ISDS for now. And while the UNCITRAL Working Group III continues to discuss potential reforms, the process has invited criticism, and its efficacy remains uncertain. These developments have renewed calls to prioritize contract-based arbitrations pursuant to arbitration clauses in investment agreements negotiated between foreign investors and host states. In the words of Alexis Mourre, a prominent arbitrator, “the arbitration community is ‘losing’ the fight to ensure the survival of ISDS … and should instead contemplate a return to the contractual protection of investments.” This suggestion comes at the heels of the joint project on international investment contracts (IICs) commenced by UNIDROIT and ICC’s Institute of World Business (IIC Project), which aims to develop an international standard to guide the contractual relationship between states and private investors, including by providing contractual guidance on aspects of international investment law treaty standards.

Mourre’s suggestion is unsurprising. International law and arbitration have historically played an important role in limiting the reach of (host) states’ national courts and municipal laws over disputes relating to foreign investments and thus, regulating the terms of investor–state relations. ^[1] Accordingly, even if not pursuant to an investment treaty, subjecting investment disputes to international arbitration procedures—including under the ICSID Convention—could continue to shield foreign investors from the jurisdiction of host states. From this perspective, the urge to replace treaty-based ISDS with contractual arbitration is not entirely unexpected.

Given the efficiencies of arbitration, this by itself is not problematic as long as the consent of all parties remains free—both formally and substantively. Nevertheless, such arbitral proceedings invite certain risks. One such risk entails international tribunals’ temptation to marginalize the applicable municipal law governing the parties’ contractual relationship and elevate the transaction to the international plane. This way, international tribunals can test the appropriateness of the host states’ conduct by reference to not only the parties’ choice of municipal law but also international law principles. With the anticipated resurgence of contract-based ISDS, this aspect becomes more significant. Indeed, as the Report of the Fourth Session (25-27 November 2024) of the IIC Project noted in relation to choice of law clauses in IICs: “public international law could and should not be excluded from a contractual setting as that of IICs since it was particularly relevant for articulating a protection of policy goals.”

This risk is not unprecedented. Antony Anghie explains how “transnational law” was developed in the 1950s on the unfounded insistence that newly independent states’ municipal laws were not sufficiently developed to deal with cross-border disputes, and the parties could not have intended for their contracts to be governed by such laws. ^[2] In recent years, customary international law (CIL) principles relating to the protection of aliens can help reach similar conclusions. A notable example is the jurisdictional decision in Cambodia Power Company v. Kingdom of Cambodia, wherein an ICSID tribunal deriving jurisdiction from contractual agreements governed by English law found itself competent to also decide claims alleging a breach of CIL norms. It reasoned that “customary international law exists and may be applied independently of any choice of law” and that the parties’ choice of English law actually had the effect of including CIL through the doctrine of incorporation. ^[3] To the tribunal, the real question was whether the parties had excluded the application of CIL norms, which it believed they had not in this case. ^[4] As Kate Parlett demonstrates, such decisions are not exceptional. ^[5]

Given the colonial legacies of CIL, this tendency of international arbitral tribunals can be criticized from a history-driven Third World approach to international law (TWAIL) lens (as the author has argued elsewhere). However, even from a purely doctrinal perspective, several errors of this approach merit attention. These concern (i) the distinction between jurisdiction and applicable law and (ii) the unreliability of the doctrine of incorporation.

Distinguishing Jurisdiction From Applicable Law

The assertion of default CIL jurisdiction in contract-based ISDS proceedings, as indicated by the Cambodia Power tribunal, is unsound as it undermines the principle of consent.

First, sovereign equality requires that an international judicial forum can only exercise jurisdiction over a state with its consent. ISDS tribunals, whether constituted pursuant to an investment treaty or a contract, are not an exception to this rule. This imposes a burden on each claimant-investor to prove, and a responsibility on the tribunal to examine, whether the host state has authorized the tribunal to assess the compatibility of its conduct against the legal norms invoked, including CIL.

The notion of consent is, however, separate from the law governing the parties’ conduct or their contractual relationship. The latter refers to legal rules that the parties agree to abide by (such as by subjecting their contract to English law) or those that ordinarily bind them by reason of their existence (such as corporate law rules of the state where a company is incorporated or jus cogens norms). However, the mere existence of such legal rules does not imply that the parties have also authorized an arbitral tribunal to apply them for deciding an investment dispute. Such authorization cannot be deduced from the customary nature of the invoked norms but must be independently traced to the source of the tribunal’s jurisdiction, such as an arbitration clause in a contract.

As the International Court of Justice had explained in the Fisheries Jurisdiction case: “[t]here is a fundamental distinction between the acceptance by a State of the Court’s jurisdiction and the compatibility of particular acts with international law. The former requires consent. The latter question can only be reached when the Court deals with the merits, after having established its jurisdiction.” ^[6]

Accordingly, confronted with arguments similar to those in Cambodia Power, international tribunals have often declined jurisdiction. In the MOX Plant Case, a tribunal constituted under the United Nations Convention on the Law of the Sea (UNCLOS) had to determine if it could decide Ireland’s claims arising under instruments other than the Convention. In support, Ireland cited Article 293 of the UNCLOS, titled Applicable Law, which provided that “[a] court or tribunal having jurisdiction under this section shall apply this Convention and other rules of international law not incompatible with this Convention.” However, the tribunal differentiated applicable law from jurisdiction—the latter addressed in Article 288 of the UNCLOS—and found Ireland’s additional claims inadmissible for want of consent. ^[7]

Second, even beyond international law, consent remains the foundation of arbitration. No investor has an inherent right to access arbitral jurisdiction and seek remedy for an injury caused to it by a state’s alleged breach of an otherwise binding legal norm. This is a privilege to be enjoyed, subject not only to the existence but also the scope of the parties’ consent to arbitration. The parties to a contract can, after all, grant a limited consent to arbitrate certain—but not all—disputes arising from their transaction, which would limit the scope of the tribunal’s consequent jurisdiction. Simply put, there can be no arbitral jurisdiction unless it is created and covered by the parties’ mutual consent. This contradicts the Cambodia Power decision, which presupposes a default jurisdiction to decide CIL claims unless it is explicitly excluded by the parties.

Two ICSID decisions, albeit rendered by in proceedings pursuant to an investment legislation and treaty respectively, corroborate this position.

In Inceysa v. El Salvador, the parties’ dispute arose from the alleged non-performance of a service contract between the claimant and the El Salvadoran Ministry of Environment and Natural Resources; Clause 21 of which required the submission of any disputes to arbitration in accordance with Salvadoran law. The claimant interpreted this reference to include Article 15 of El Salvador’s Investment Law, which mentioned the ICSID Convention. El Salvador objected to the ICSID’s jurisdiction to decide the claimant’s contractual claims, pointing to a disparity between the invoked source of the tribunal’s jurisdiction (i.e., the municipal investment law) and the legal basis of the claims submitted (i.e., breach of contract). The tribunal concurred with El Salvador. It held that “to invoke the arbitration jurisdiction provided in the Investment Law, there must be a claim with substantive grounds in said law, a situation which does not exist in the case at hand.” ^[8] This did not mean that the parties’ contract was no longer binding but that the legislation containing the state’s consent to ICSID arbitration did not authorize the tribunal to decide claims alleging a breach of contract.

The decision in Emmis v. Hungary is to the same effect. The claimant requested an ICSID tribunal constituted pursuant to the Hungary–Netherlands BIT and Hungary–Switzerland BIT to decide claims alleging a breach of CIL norms. Hungary requested a preliminary dismissal under Rule 41(5) of the ICSID Arbitration Rules for want of jurisdiction, which it believed could not be remedied by reference to the expression “such rules of international law as may be applicable” in Article 42(1) of the ICSID Convention dealing with applicable law. The tribunal agreed in principle. It clarified that the issue was not about the determination of the content of CIL norms but whether it had been authorized to decide the claimant’s “additional separate cause of action.” ^[9] This was a question of jurisdiction, not applicable law, to be answered by reference to the scope of consent granted in the treaties invoked. The tribunal ultimately found that while Article 10 of the Hungary–Switzerland BIT did not grant such authorization, Article 10 of the Hungary–Netherlands BIT was more permissive. ^[10]

There is no cogent reason for contract-based ISDS tribunals to adopt a different approach. After all, irrespective of the precise source of the host states’ consent to arbitrate, the customary or binding nature of an international law obligation remains distinct from an ISDS tribunal’s jurisdiction to examine that state’s compliance with such obligation.

Revisiting the Doctrine of Incorporation

A further obstruction to the potential internationalization of investment contracts relates to the doctrine of incorporation, which the Cambodia Power tribunal had referred to.

In the United Kingdom, the doctrine of incorporation was traditionally understood as generally implying that CIL rules automatically became part of English law. ^[11] However, English scholars and courts have since doubted its acceptance and contested the unqualified terms in which the doctrine is stated. For instance, in In re McKerr, the House of Lords clarified that incorporation of international law had little relevance where the law enacted by the Parliament occupied the field. Accordingly, it refused to create an overriding common law obligation analogous to Article 2 of the European Convention on Human Rights in an area of law for which the Parliament had long legislated since that would contradict how common law is ordinarily developed. ^[12] The UK Supreme Court reinforced this position in Keyu v. Secretary of State, clarifying that “it would be inappropriate for English courts to import the suggested CIL principle … into domestic law, because Parliament has effectively pre-empted the whole area.” ^[13]

Given this uncertainty about its status and scope, the doctrine of incorporation hardly provides a reliable basis to casually elevate municipal law-governed contracts to the international plane.

Even otherwise, the doctrine of incorporation—or as it was understood in Cambodia Power—lacks widespread acceptance in other regions. Different states, including those of common law tradition, address the relationship between municipal and international law differently. Many give effect to international law principles only to the extent they are consistent with domestic constitutional principles, ^[14] thus undermining any prospect of automatic incorporation. Consequently, in contracts not governed by English law, there is no reasonable basis to insist that by subjecting their contract to a particular municipal law, the parties had simultaneously intended to create a default CIL jurisdiction through the implied incorporation of CIL norms.

For these reasons, international law scholars now downplay the utility of the doctrine of incorporation in ISDS cases. Christoph Schreuer opines that “it would not be wise to rely on the incorporation of international law into domestic law as a general proposition. The status of international law under domestic constitutions varies greatly.” ^[15] Hege Kjos concurs, observing that “[s]tates differ with regard to the extent to which they incorporate international law; and investors should therefore be warned against relying on the automatic application of international law via national law.” ^[16] Malcolm Shaw likewise concludes that “a considerable degree of caution may therefore now be necessary with regard to the traditionally and baldly expressed proposition that customary international law is part of English law.” ^[17]

One hopes that future ISDS tribunals are guided by these observations, as opposed to the reasoning offered by the Cambodia Power tribunal.

Conclusion

The anticipated resurgence of contract-based ISDS brings to attention the risk of arbitral tribunals internationalizing investor–state relations that are ordinarily governed by municipal laws. As the Cambodia Power decision showed, this could create a potentially infinite jurisdiction wherein, irrespective of the scope of the parties’ consent, tribunals could decide claims alleging a breach of any international norm perceived to be customary. Given the imperial legacies of CIL and international investment law, this outcome is hardly satisfactory. It is thus necessary to guard against these risks. While it is difficult to predict the approach of future contract-based ISDS tribunals, one must ensure that the discontents of treaty-based ISDS do not spill over to the contractual domain. Otherwise, to paraphrase Mourre, the international arbitration community could risk losing the latter fight, too.

Author

Harshad Pathak is an Indian lawyer, presently a doctoral candidate at the University of Geneva and a consultant with the International Arbitration team at Mayer Brown Paris; [email protected].

The author thanks the participants at the Lillehammer workshop on international investment contracts (2024)—particularly Dr. Yuliya Chernykh and Dr. Natalia Torres Zuniga—for their reflections on an initial draft. The contents of this article reflect the views of the author alone and not of any organization he may be associated with.

^[1] Miles, K. (2013). The origins of international investment law: Empire, environment and the safeguarding of capital. Cambridge University Press, at 56–58, 67–68.

^[2] Anghie, A. (2005). Imperialism, sovereignty and the making of international law. Cambridge University Press, at 228.

^[3] Cambodia Power Company v Kingdom of Cambodia ICSID Case No ARB/09/18, Decision on Jurisdiction (22 March 2011) [332–333].

^[4] ibid 335.

^[5] Parlett, K. (2016). Claims under customary international law in ICSID arbitration. ICSID Review 31(2), 434–456.

^[6] Fisheries Jurisdiction (Spain v Canada) Jurisdiction of the Court, Judgment (4 December 1998), ICJ Reports 1998, p 432, 456.

^[7] The MOX Plant Case (Ireland v United Kingdom) ITLOS Case No 10, ICGJ 343 (ITLOS 2001), Order No 3 (24 June 2003) [19].

^[8] Inceysa Vallisoletana SL v Republic of El Salvador ICSID Case No ARB/03/26, Award (2 August 2006) [333].

^[9] Emmis International Holding, BV and others v Hungary ICSID Case No ARB/12/2, Decision on Respondent’s Objection Under ICSID Arbitration Rule 41(5) (11 March 2013) [79].

^[10] ibid., 80–83.

^[11] Trendtex Trading Corporation v Central Bank of Nigeria [1977] QB 529, 554.

^[12] In re McKerr [2004] UKHL 12 [32].

^[13] Keyu and others v Secretary of State for Foreign and Commonwealth Affairs and another [2015] UKSC 69 [151].

^[14] For instance, see Adigun, M. (2019). The status of customary international law under the Nigerian legal system. Commonwealth Law Bulletin, 45(1), at 115; Commissioner of Customs, Bangalore v GM Exports Supreme Court of India, (2016) 1 SCC 91 (23 September 2015).

^[15] Schreuer, C. H., Malintoppi, L., Reinisch., A., & Sinclair, A. (2009). The ICSID Convention – A commentary (2nd ed.). Cambridge University Press, at 582. https://icsid.worldbank.org/sites/default/files/parties_publications/C8394/Claimants%27%20documents/CL%20-%20Exhibits/CL-0077.pdf

^[16] Kjos, H. E. (2013). Applicable law in investor-state arbitration: The interplay between national and international law (1st ed). Oxford University Press. 182–183.

^[17] Shaw, M. N. (2017). International law (8th ed.). Cambridge University Press, at 112.