Other Publications

Executive Summary

An investor may bring an arbitral claim against a state under the rules of the International Centre for Settlement of Investment Disputes when three conditions are met: (1) the parties to the dispute consent in writing to jurisdiction, (2) the dispute arises directly out of an investment, and (3) the dispute arises between a Contracting State and a national of another Contracting State. Similar conditions apply to investor state claims brought under other rules. This chapter explains the rules defining the kinds of cases that may be brought to investor-state arbitration. The next chapter discusses other jurisdictional issues, in particular temporal questions, and exceptions and carve-outs.

1.0 Introduction

A Greek citizen alleged that the British Mandate in Palestine refused, in violation of international law, to recognise rights he had acquired by agreement with the previous government, the Ottoman Empire. Who had the right to bring an international claim against the British Government?

In 1924, the Permanent Court of International Justice gave the traditional answer: Greece could bring the claim.¹ In so doing, Greece was “asserting its own rights”, namely its right “to protect its subjects” and its right “to ensure …respect for the rules of international law”. By bringing an international claim, Greece transformed the dispute from one “between a private person and a state” to “a dispute between two states”.

This was the answer of inter-national law, law between nations. Only a state could initiate a claim against other states through a process known as diplomatic protection or espousal. When a state chose to espouse a claim, it was deemed to be “asserting its own rights” as the “sole claimant”. Indeed, only states were deemed “subjects” of international law, while human beings and companies were treated as mere “objects”.²

As late as 1970, the International Court of Justice stated, “The State must be viewed as the sole judge to decide whether its protection will be granted, to what extent it is granted, and when it will cease… [T]he State enjoys complete freedom of action.”³

By then, however, the wall dividing states from private persons was cracking. Four years earlier, the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (the ICSID Convention) had entered into effect. As its name suggests, the Convention focused on disputes between states and foreign nationals — that is, private persons. This focus gave rise to the jurisdictional requirements set forth in the first sentence of Article 25(1):

The jurisdiction of the Centre shall extend to any legal dispute arising directly out of an investment, between a Contracting State (or any constituent subdivision or agency of a Contracting State designated to the Centre by that State) and a national of another Contracting State, which the parties to the dispute consent in writing to submit to the Centre.

Accordingly, tribunals acting under the auspices of the International Centre for Settlement of Investment Disputes (ICSID or the Centre) have jurisdiction over any legal dispute when three conditions are met: (1) the parties to the dispute consent in

[Page89:]

writing to jurisdiction, (2) the dispute arises directly out of an investment, and (3) the dispute arises between a state that has ratified the ICSID Convention (Contracting State) and a national of another Contracting State.

Investors considering whether to start an investment claim should focus early on whether these three conditions are met. Tribunals have addressed the meaning of these conditions in many cases, but the conditions often remain ambiguous and disputed, and important differences could result from seemingly minor differences in the text of a treaty or contract. Accordingly, investors with potential claims should consult with qualified counsel about whether the legal requirements for jurisdiction are met in a particular case.

This chapter addresses each of these conditions, starting with consent because it is the “essential characteristic of the entire system of the Convention”.⁴ Other issues concerning jurisdiction are addressed in the next chapter.

This chapter focuses on ICSID jurisdiction, but investors should be aware that they may have the right to bring to an investor-state claim outside ICSID. Non-ICSID investment cases are often conducted under the UNCITRAL arbitration rules, but they can also sometimes be brought before other arbitral institutions, such as the International Chamber of Commerce or the Stockholm Chamber of Commerce. There are important differences between ICSID and non-ICSID cases,⁵ to be sure, including some differences in jurisdiction, but in broad strokes investors will still want to consider the issues addressed in this chapter:

• Did the parties c on sent to the tribunal’ s jurisdiction ?
• Does the tribunal have jurisdiction over the subject of the dispute?
• Does the tribunal have jurisdiction over the parties to the dispute?

2.0 Consent

The founding of ICSID opened the door of international adjudication to private persons. Private persons are now capable of enjoying and asserting their own rights under international law.⁶

This change engendered controversy, as it required the states negotiating the ICSID Convention to afford substantial power to private persons. ICSID’s architects squared the circle by building ICSID on the foundation of consent. The consent requirement allows each ICSID Contracting State to decide for itself how far to go into the brave new world of investor claims — to decide how many and what kinds of cases it is willing to allow investors to bring against it. ICSID Contracting State status does not automatically open the door to any private claims; no case can be brought against a Contracting State unless that state also consented to jurisdiction over a dispute. This has been called “the doubly voluntary character” of ICSID: states may choose whether or not to join ICSID and, if they join, Contracting States remain “free to decide whether or not to utilise the Centre’s facilities by consenting to its jurisdiction in respect of particular arrangements or disputes.”⁷

State consent is the “essential prerequisite” for ICSID jurisdiction and “the cornerstone of the jurisdiction of the Centre”.⁸ So woven into ICSID’s fabric is the consent requirement that it appears three times in the preamble alone.⁹ This focus on consent anchored ICSID in traditional international legal norms,¹⁰ even as ICSID otherwise broke radically from those norms by empowering private persons to act in their own interests.

Various other provisions of the ICSID Convention also show the importance of consent. Once mutual consent is given, it cannot be unilaterally withdrawn.¹¹ Consent by an agency or subdivision of a state does not suffice alone; it requires action by the state itself, either state approval or state notice that no such approval is needed.¹² States may notify ICSID of the types of cases that they will (or will not) submit to ICSID jurisdiction, but such a notice does not constitute consent.¹³ Consent excludes any other remedy (unless otherwise stated) and, in particular, excludes diplomatic protection by the investor’s home state.¹⁴ Finally, neither denunciation of the ICSID

[Page90:]

Convention nor amendments to the Convention affect “the rights or obligations under this Convention …arising out of consent” given before the denunciation or amendment.¹⁵

Article 25 of the ICSID Convention specifies that consent must be expressed in writing, but does not otherwise address the form of consent. Parties to a contract typically express consent to arbitration at the same time (“synchronously”) when they sign the contract and agree to arbitrate a particular dispute or all disputes arising out of the contract. Parties to a dispute may also express their consent at different times (“nonsynchronously”). ¹⁷ For example, a state may enact an investment statute giving rights to investors and that statute may express the state’s standing consent to arbitration brought later by the investors, who typically consent when or shortly before they file the case. Most commonly, as in the AAPL v Sri Lanka case mentioned above, states include investor rights and consent to arbitration in treaties — in particular bilateral investment treaties (BITs),¹⁸ specialised treaties like the Energy Charter Treaty, and regional trade agreements like NAFTA.

Investors should look for expressions of the state’s consent to arbitration in applicable contracts, statutes and treaties.

Treaty-based consent has driven the growth of ICSID’s docket, with treaty-based cases accounting for 91% of all ICSID cases and statute-based cases 9.0%.¹⁹ The vast majority of ICSID cases have been filed since the Argentine financial crisis in 2001, which triggered a surge in treaty-based cases. ICSID registered a record 52 new cases in 2015, 48 in 2016, and 53 in 2017 — more new cases in 2015-2017 (153) than in the first 35 years combined (1966-2000).²⁰

3.0 What is an Investment?

Article 25 of the ICSID Convention specifies, as mentioned, that disputes must “aris[e] directly out of an investment”. The Convention does not define the word “investment” — because the negotiators could not agree on a definition and decided to omit one.²¹ Before Salini v Morocco in 2000, ICSID tribunals did not dwell on this word.

Salini took the view that the word “investment” in Article 25 must have a meaning that restricts the kinds of cases that parties can submit to ICSID. Salini then described four elements of “investment” for purposes of Article 25:

1. A “contribution” of capital or know-how, etc.;

[Page91:]

2. A “certain duration of performance”;

3. A “participation in the risks of a transaction”; and

4. A “contribution to the economic development of the host state”.²³

Salini started a lively debate. Its supporters insisted that Article 25 established a jurisdictional floor, even if a state used a broader definition of “investment” in the investment treaty where it consented to ICSID jurisdiction. Salini’s critics argued that tribunals should defer to the definition of “investment” used by the state.²⁴ Among Salini’s supporters, there was further debate about the proper definition of “investment” — tribunals have variously applied three-, four-, five-, and six-part definitions.²⁵

Newer treaties tend to have more detailed definitions of “investment”, plainly influenced by Salini — an example of states taking arbitral interpretations into account when drafting later treaties. This trend suggests that states generally support the concept of Salini, while rejecting its particular definition as too strict. The newer treaties often list several “characteristics” of an investment typically mentioned in the cases, but the treaty lists are generally far more liberal than the cases. In place of a fixed multipart test, the treaties typically preface their lists of characteristics with the phrase “such as” and tie them together with the word “or” (rather than “and”), making clear that an investment need not have all of the listed characteristics.²⁶

Although Salini originated in the ICSID Convention, which does not apply to investment arbitrations heard by non-ICSID tribunals, the Salini approach seems to be spreading outside ICSID.²⁷ This trend may be confirmed as more states adopt more detailed definitions of “investment” in their new investment treaties.

Salini observed that promoting development is one of ICSID’s key goals, but may have erred by translating that goal into a rule that a claimant had to prove, as a condition of jurisdiction, that it had made a contribution to the development of the host state.²⁸ However, Salini’s development test is faring poorly in recent treaties: it is absent from the listed characteristics of “investment” in treaties signed by a broad swath of countries, including the ASEAN Comprehensive Investment Agreement of 2009, the China-Japan-Korea trilateral agreement of 2012, and the Canada-EU Comprehensive Economic and Trade Agreement which entered into force provisionally on 21 September 2017.²⁹

The trend towards more detailed definitions should make it easier for tribunals to decide whether the claimant has made a qualifying “investment”, which should help reduce the cost of this aspect of investment arbitration.

4.0 State v Foreign National

4.1 A Contracting State

Under Article 25 of the ICSID Convention, one party to each case must be a “Contracting State” to the Convention. This may include “any constituent subdivision or agency of a Contracting State designated to the Centre by that State”. It also includes “all territories for whose international relations a Contracting State is responsible”, unless a territory is excluded in writing.³⁰

Although in principle the Contracting State may be either the claimant or respondent, in practice states hardly ever initiate ICSID claims. In part, this is because treaty-based cases, which have become dominant, can only be started by investors. Even contract-based cases, however, are usually started by investors — this may simply reflect that private persons have a stronger need to resort to international arbitration as a means of resolving disputes with governments.

[Page92:]

Determining whether a party is a Contracting State is normally a straightforward task: ICSID has 154 Contracting States (as of July 2018), listed on its website.³¹ Some questions about timing can arise, however, when a state denounces its membership, as Bolivia, Ecuador and Venezuela did between 2007 and 2012. These issues are discussed in the next chapter.

In considering whom to name as a respondent in an ICSID case, claimants should ensure that the respondent is both subject to ICSID jurisdiction and responsible for the injuries claimed.

4.2 A National of Another Contracting State

Each ICSID case must also have as a party “a national of another Contracting State” that opposes the Contracting State party. This is one of the hallmarks of ICSID arbitration — distinguishing it from (for example) disputes at the International Court of Justice and the World Trade Organization, where private parties cannot bring actions.

The foreign national may be either a natural or legal person. Nationality for legal persons (companies) is determined on the date of consent, while natural persons (human beings) must be nationals of “another Contracting State” on both the date of consent and the date the claim is registered at ICSID.³²

The definition of “nationality” is more restrictive for natural persons than for legal persons. Natural persons must be nationals of another state, of course, but there is a further ICSID requirement that they cannot also be nationals of the adverse state party — that is, they cannot be dual nationals.³³ By contrast, some companies organised under the laws of the adverse state party may be deemed to be foreign nationals for the purpose of this jurisdictional requirement. That is the case where, “because of foreign control, the parties have agreed [that the company] should be treated as a national of another Contracting State for the purposes of this Convention.”³⁴ For example, the Argentine subsidiary of a US corporation may bring an ICSID case against Argentina, because the BIT treats the subsidiary as a US person for this purpose.³⁵

The home state of the private party must also be a Contracting State of ICSID to obtain ICSID jurisdiction. However, the ICSID “Additional Facility” is available in cases where only one of the two states is an ICSID Contracting State. Additional Facility awards do not benefit from the ICSID Convention’s special enforcement provisions.

[Page93:]

Article 25(1) requires that the investor be from “another” state than the adverse state party and BIT-based cases demand that the investor be from a particular state, namely the other party to the BIT. The scenarios above, Tokios Tokeles v Ukraine and Mobil Corp. v Venezuela, illustrate two issues that can arise when one maps the changeable nature of corporate groups against the nationality requirements of the legal regime. Both scenarios question not the claimant’s formal nationality, but its good faith. Accordingly, these issues are sometimes called, derisively, “round-tripping” and “treaty shopping”.

The Tribunals in Tokios Tokeles and Mobil each upheld the claimant’s formal nationality and hence its jurisdictional eligibility as a foreign national. Neither saw abuse on the facts before it. Yet, both considered that there is a limit to corporate flexibility, a point at which corporate structuring is deemed to circumvent abusively the other-nationality requirement. In this regard, the Mobil Tribunal distinguished Phoenix Action v Czech Republic, a case in which the ICSID Tribunal found that the claimant had not made “an economic investment … but indeed, simply a rearrangement of assets within a family, to gain access to ICSID jurisdiction [for a pre-existing dispute] to which the initial investor was not entitled.”³⁸ The Phoenix Action Tribunal refused jurisdiction, finding that the claimant had engaged in “abusive manipulation of the system of international investment protection” by trying to disguise domestic investments as international “for the sole purpose of access” to ICSID.³⁹

Similarly, in a recent high-profile investment case outside ICSID, jurisdiction turned on the question whether the claimant had restructured abusively.

[Page94:]

To minimise questions about restructuring, it is prudent for investors to consider BIT protection (among other considerations) when initially structuring an international investment.

5.0 Conclusion

Tokios Tokeles and Mobil both focused on the particular BITs at issue in those cases. Neither BIT required the tribunal to look past the claimant’s formal nationality. States may make other choices, if they wish, in their BITs. For example, the US Model BIT includes a “denial of benefits” clause that allows (inter alia) one party to deny benefits of the BIT to an investor from the other party if the investor is an enterprise that “has no substantial business activities in the territory of the other Party and persons of a non-Party, or of the denying Party, own or control the enterprise.”⁴³ The US Model BIT thus imposes a substantial business activities test that would have been relevant in the circumstances presented by both Tokios Tokeles (where “persons …of the denying Party” own the investor) and Mobil (where “persons of a non-Party” own the investor). Denial of benefits clauses are discussed in more detail in Chapters 6 and 11.

The corporate structuring cases raise important policy questions that may affect investors. Should international investment arbitration be available where an “essentially domestic” investment is structured through a foreign company? Should the protections of a BIT be available to third-country investors that incorporate an entity in a state party? Should it matter whether the claimant company is a holding company or has substantial activities at the place of incorporation? Should tribunals delve into the reasons for corporate structuring decisions?

Indeed, should tribunals insist on any international element at all? To be sure, as discussed, the other-nationality requirement of Article 25(1) demands an international element — but this is not the only possible policy choice. States may choose to afford the benefits of neutral adjudication to their own citizens. The other-nationality element is a textual rule, not a general requirement of international law. Indeed, its absence is fundamental to modern human rights law, where international tribunals routinely hear disputes between citizens and their own states. This idea could be extended to the investment sphere, which has more in common with human rights law than is often appreciated,⁴⁴ especially if one follows the European example by treating property as a human right.⁴⁵

In the end, jurisdiction is more than a collection of technical requirements. It is a reflection of values. In the case of ICSID, those values include the centrality of consent, the promotion of development, and making neutral and effective international adjudication open to private persons. The other-nationality requirement is written into ICSID’s constitution — it may have seemed 50 years ago to be a necessary, even inherent aspect of ICSID’s profoundly progressive opening to private actors, but it is fair today to question the on-going value of this limitation on party autonomy.

While traditional international tribunals like the International Court of Justice and the World Trade Organisation may be seen as horizontal institutions governing relations between governments, and domestic administrative litigation is typically vertical between citizen and state, the other-nationality requirement makes ICSID a diagonal institution.⁴⁶ It is devoted to disputes between a state and a national of another state, excluding both state-state and state-citizen cases. It may be time, however, to think about adding a vertical element to ICSID’s diagonal structure by allowing states, if they choose, to submit disputes with their own nationals. This could be done through the Additional Facility,⁴⁷ which already minimises the Convention’s textual requirement that both the home and host states must be ICSID Contracting Parties. Of course, as with the Additional Facility and ICSID itself, there would only be jurisdiction for vertical arbitration cases between a state and its own nationals where the parties consented.

[Page95:]

Notes

---

---