1. Introduction
It is now standard in contemporary international law commentary to note that the latter part of the twentieth century has seen a move away from the traditional understanding of international law as fundamentally based on the consent of states.Footnote 1 As just two examples, customary international law has in some contexts become more influential than treaties,Footnote 2 and human rights obligations are now recognized as often binding states, even when they have signed no treaty acknowledging their existence.Footnote 3
Treaties, too, have changed. They remain an agreement between statesFootnote 4 and, as with any agreement, the consent of the parties is essential.Footnote 5 However, states have demonstrated an increased willingness to place their treaty-making powers in the hands of other actors.Footnote 6 Treaties have, of course, long been used to promote non-state interests, particularly to protect participants in international business.Footnote 7 More recently, however, actors who may traditionally have benefited from the exercise of a state's treaty-making powers have been allowed a more active role in the creation and even execution of treaties.
This change has, however, had little impact on the interpretation of treaties. Tribunals may adopt a teleological approach to the interpretation of treaty language where a treaty is seen as intended by the contracting states to achieve certain goals not easily captured by precise language.Footnote 8 However, even a teleological approach to treaty interpretation remains fundamentally focused on the contracting states, as the tribunal simply understands those states as having granted it the authority to fashion its own rules to achieve the treaty's goals, rather than being strictly bound by the specific language of the treaty.Footnote 9
The purpose of this article is to highlight the existence of a subset of treaties for which even a teleological approach to interpretation fails to capture the central importance for the treaty of entities other than the contracting states. These ‘power-conferring treaties’ do not merely entrust tribunals with the power to effectively fashion the means by which a treaty's goals should be achieved. To varying degrees, they grant control of the treaty itself, including, at times, both its enforcement and the very meaning of its terms, to entities other than the contracting states. As a result, the traditional emphasis in treaty interpretation on the ‘object and purpose’ of the treaty, and the precise language in which the treaty is written, will fail to generate an interpretation that faithfully captures the manner in which the treaty genuinely functions.
This is not to say that the traditional rules of treaty interpretation, as codified in the Vienna Convention on the Law of Treaties (VCLT), have no role in the interpretation of power-conferring treaties. They are, after all, now recognized as reflecting customary law regarding treaty interpretation.Footnote 10 This article will attempt to demonstrate, however, that recognition of the power-conferring nature of a treaty can have an important impact on the way that these traditional tools must be applied.
It will do this through an analysis of a highly contentious topic in contemporary international investment law – the meaning of the term ‘investment’ in Article 25(1) of the International Convention for Settlement of Investment Disputes (the ‘Convention’). Article 25(1) identifies the jurisdictional limits of arbitrations that can be pursued at the International Centre for Settlement of Investment Disputes (ICSID). However, although it specifies that any dispute arbitrated under the Convention must arise directly out of an ‘investment’, the Convention does not clarify what the term ‘investment’ means.Footnote 11 The meaning of ‘investment’ in Article 25(1) has, as a result, become a primary area of dispute within ICSID arbitration, with arbitral tribunals reaching conflicting interpretations of the term, based on differing conceptions of the rationale for the term's inclusion.Footnote 12
This article will demonstrate that recognizing the ways in which the ICSID Convention confers powers upon entities other than the ICSID contracting states to determine the meaning of ‘investment’ in Article 25(1) provides important insights into how the term should be interpreted.
Section 2 of the article will clarify what it is for a treaty to be power-conferring, and what the impact on the interpretation of a treaty will be of the recognition that it is power-conferring. Section 3 will then turn specifically to international investment law, explaining the particular importance power-conferring treaties have within that field.
Section 4 of the article will commence the discussion of the ICSID Convention itself, presenting the background to the contemporary dispute regarding the meaning of the term ‘investment’ in Article 25(1) of the Convention. Section 5 will then demonstrate that the Convention confers upon the parties to any given arbitration at ICSID the power to determine, to some degree, the meaning of the term ‘investment’ in Article 25(1), in so far as it concerns their own arbitration. Section 6 will then address Article 25(4) of the Convention, which confers upon individual contracting states the power unilaterally to define the meaning of ‘investment’ for any arbitration in which they are a party. Section 7 will then examine Article 64 of the Convention, which, it will be argued, confers upon any two or more contracting states the power to agree upon a definition of ‘investment’ under Article 25(1), which is then binding in any arbitration brought by an investor from one state party to that agreement against any other state party to that agreement. Finally, section 8 will address the ‘minimum content’ of the term ‘investment’ in Article 25(1), which provides an outer limit beyond which no entity has been granted the power to extend the definition of ‘investment’.
2. Treaty interpretation and power-conferring treaties
The observation that some legal rules confer powers on individuals or entities, rather than regulating them, is far from original, being most famously made by the legal theorist H. L. A. Hart in his criticisms of the jurisprudence of John Austin.Footnote 13 Austin's ‘command’ theory of law had attempted to explain laws entirely in terms of commands issued by a sovereign to those over whom the sovereign has power.Footnote 14 Hart's objection to this command-centred view of law emphasized that there are certain types of law that such a description simply does not seem to match.Footnote 15 Contract law, for example, does not mandate that contracts be made, or even that they be made in certain ways. Rather, it presents an opportunity. Individuals are told the conditions under which the state will enforce agreements that they have made. As a result, they are given what Hart described as a limited legislative power to determine the legal framework that will apply to their own actions.Footnote 16
Hart's recognition of the power-conferring nature of certain legal rules is not important because power-conferring laws cannot plausibly be described as a form of command. Contract law, for example, can be certainly understood as constituting a set of commands from a sovereign that agreements be made in certain ways.Footnote 17 Hart's point, rather, was that doing so simply mischaracterizes the true nature of the laws in question.Footnote 18 Contract law can indeed be described as a form of command, but such an understanding does not faithfully capture the true manner in which those laws operate.
It is this notion of fidelity that is important in understanding the impact of the recognition that a treaty is power-conferring on the manner in which that treaty should be interpreted. The rules of treaty interpretation listed in the VCLT are now recognized as reflecting customary international law, and there is no reason why they should not be used to interpret power-conferring treaties. However, recognition that a treaty is power-conferring means that while traditional methods of treaty interpretation are still applicable, applying them in the same way as would be done for a non-power-conferring treaty will lead to an interpretation unfaithful to the treaty itself.
Power-conferring treaties cannot be properly understood solely through an emphasis on such things as the ‘object and purpose’ of the treaty or the specific language used in the treaty.Footnote 19 Such considerations are important in determining which elements of a treaty are power-conferring and also in interpreting those elements that are not. However, a power-conferring treaty is, by definition, not an attempt to capture an agreement between the parties to the treaty. Rather, it creates a framework in which the wishes and views of entities and individuals other than the parties to the treaty can have a central role in the treaty's operation.
At the most basic level, then, recognition that a treaty is power-conferring highlights the need to identify precisely which entities have been granted powers under the treaty and how they are entitled to use them. As a result, aspects of the treaty that would ordinarily have been interpreted in a manner intended to achieve the goals of the parties to the treaty must instead be interpreted to achieve the goals of a non-party.
In addition, recognition that a treaty is power-conferring can even affect how the VCLT's rules of treaty interpretation are applied. For example, the purpose of the ICSID Convention has often been described in terms such as ‘to promote the flow of private investment to contracting countries by provision of a mechanism which, by enabling international settlement of disputes, conduces to the security of such investment’.Footnote 20 Applying the rules of the VCLT in accordance with this understanding of the Convention's ‘object and purpose’ would justify taking an expansive view of the jurisdiction of ICSID tribunals. After all, the Convention cannot ‘promote the flow of private investment’ if tribunals decline jurisdiction over cases in which an investor alleges mistreatment by a host state.Footnote 21
However, when the Convention is recognized as power-conferring, this description of the object and purpose of the Convention becomes less plausible. Promoting the private flow of investment certainly remains part of the Convention's object and purpose, but if, for example, powers to constrain ICSID jurisdiction have been conferred on entities other than the parties to a dispute, then it is clearly not a full account of what that object and purpose is. After all, such parties will have the ability to exercise that power in ways that may limit the promotion of the flow of private investment, even where that flow is desired by the host state receiving the investment.
The remainder of this article will attempt to illustrate the impact of recognizing a treaty's power-conferring nature by focusing on a single example of power-conferring in a treaty: the definition of ‘investment’ in Article 25(1) of the ICSID Convention. As will be shown, this aspect of the Convention is a particularly complex example of power-conferring. Consequently, it serves as a useful example of the effect the power-conferring nature of a treaty can have upon a treaty's interpretation.
3. The rise of power-conferring treaties in international investment law
The ability of legal rules to confer powers, rather than just regulate behaviour, has come to have particular importance for international investment law since the middle of the twentieth century. Traditionally, investors in a foreign state who believed they had been mistreated by that state were generally required to pursue their claim in the state's own domestic courts.Footnote 22 If this failed to generate adequate compensation, the investor was then forced to turn to its home stateFootnote 23 and ask its government to undertake ‘diplomatic protection’ efforts on its behalf.Footnote 24 If the investor's plea was successful, its government would enter into negotiations with the government of the host state as a means of settling the investor's claim. Potentially, the two governments might both consent to submit the dispute to arbitration.Footnote 25
Diplomatic protection, however, has significant weaknesses as a means of resolving international investment disputes.Footnote 26 First, the home state has no obligation to pursue the investor's claim at all.Footnote 27 It may, for example, conclude that pursuing the investor's claim would be incompatible with its own foreign-policy objectives with respect to the host state. Second, even if the home state does pursue the investor's claim, the investor has no control over how this is done.Footnote 28 Consequently, the investor may find that its claim has been settled for far less than it believes it is actually worth. Alternatively, if arbitration is commenced, it is the home state and not the investor that controls the arbitration, and the investor only participates to the extent permitted by the home state.Footnote 29
While this system arguably functioned adequately for hundreds of years, the globalization of business in the twentieth century brought out clearly the significant limitations of diplomatic protection.Footnote 30 Increasing amounts of foreign investment were being made into an increasing diversity of countries, which, in turn, led to an increasing number of investment-related disputes.Footnote 31 As a result, states found themselves petitioned more and more often to intervene on behalf of a citizen or domestic corporation that had invested abroad. Similarly, states that were recipients of foreign investments found their domestic activities more frequently challenged by other states, whose investors had been affected by the measures in question. Finally, investors themselves complained of having to avoid potentially profitable investments due to concerns that they would not receive adequate compensation if mistreated by the host state.
It was in this context that, in the mid-twentieth century, states began collectively to examine methods for encouraging international investment that would both provide an alternative to diplomatic protection and minimize the involvement of the home state in the resolution of investment-related disputes. ICSID, created by the ICSID Convention, was a primary embodiment of this goal.Footnote 32 Expressly designed as a forum in which investors and host states could resolve disputes directly with one another, it was seen as a means of de-politicizing investment disputes, allowing them to be resolved on a purely legal basis.Footnote 33 More importantly for present purposes, however, since the structure of ICSID arbitration precludes the home state from any direct participation in the arbitral process, it also represented a grant of power to investors, who now had primary control over their own claims.
ICSID arbitration is not mandatory, however, and will only occur where both the host state and the investor have consented to their dispute being heard at ICSID.Footnote 34 This consent might be found in an agreement directly between the host state and the investor; however, it has increasingly been found in bilateral investment treaties (BITs) between the host state and the home state.Footnote 35 This is significant for present purposes because such treaties will include a range of substantive promises between the two states regarding how they will treat one another's investors.
This means, however, that although it is the investor that brings and controls an investor–state arbitration under a BIT, the substantive rights being protected by the arbitration are actually those of the home state, not of the investor.Footnote 36 Nonetheless, even though the investor is ultimately enforcing its home state's rights, it has been granted the power to pursue a claim that its home state might prefer be left alone or refuse to pursue a claim if it finds this to be advantageous to its own business purposes. Moreover, when pursuing a claim, it is the investor, not the investor's home state, that determines what arguments will be advanced regarding how the treaty in question should be interpreted.Footnote 37 Consequently, contemporary BITs, through their standard incorporation of investor–state arbitration, grant investors the power not only to protect their own investments, but to determine when the home state's own substantive rights should be protected, and even the interpretation that those rights should be given.Footnote 38
Contemporary international investment law, then, has come to be dominated by power-conferring international agreements, negotiated directly between states, but giving significant powers to investors to control both when those agreements are enforced and even how they should be interpreted.
4. The controversy regarding the meaning of ‘investment’ in the ICSID Convention
If the ICSID Convention is to be understood as a power-conferring treaty, it is necessary to clarify precisely who has been granted the right to exercise powers under it and what those powers are. This article is not an exhaustive analysis of the Convention and so will not attempt to identify every way in which the Convention might be power-conferring. It will instead focus upon the power-conferring aspects of one element of Article 25(1) of the Convention, which addresses the jurisdiction of an ICSID tribunal.
One of the foundational concepts of any form of arbitration is that parties are only obligated to arbitrate in accordance with the consent to arbitration that they have given.Footnote 39 That consent need take no particular form and may be merely formal, rather than considered and fully voluntary,Footnote 40 but, ultimately, arbitration is consensual and unless consent to the arbitration by both parties can be found, an arbitral tribunal has no jurisdiction.Footnote 41
However, while consent is fundamental to all forms of arbitration, institutional arbitrations include the additional possibility of a constraint based upon the nature of the parties to the dispute or the dispute itself, rather than just party consent. While parties may, after all, arbitrate any dispute they wish to arbitrate, arbitral institutions have no obligation to administer any arbitration that does not meet whatever standards they wish to impose.
In accordance with this feature of institutional arbitration, one of the distinctive elements of the jurisdiction of a tribunal constituted under the ICSID Convention is that the mere consent to arbitrate by the parties to the dispute does not suffice to ground the jurisdiction of the tribunal. Instead, the ICSID Convention explicitly restricts the jurisdiction of ICSID tribunals to disputes ‘arising directly out of an investment’.Footnote 42 Consequently, even if both parties to a dispute agree to arbitrate at ICSID, an ICSID tribunal must decline jurisdiction unless it finds that this requirement has been met.Footnote 43
While potentially a fairly straightforward constraint, this aspect of the jurisdiction of ICSID tribunals ultimately has come to be one of the most controversial topics in ICSID arbitration, due to the way in which the constraint is written into the Convention.Footnote 44 Whilst Article 25(1) of the Convention explicitly restricts ICSID arbitration to disputes ‘arising directly out of an investment’, no indication is given anywhere in the Convention as to exactly what constitutes an ‘investment’ for the purposes of the jurisdiction of an ICSID tribunal.Footnote 45 As a result, ICSID tribunals have been left to determine for themselves how to interpret this term and thus how extensive their own jurisdiction should be.
While early ICSID arbitrations concerned paradigmatic cases of international investment, such that the existence of an investment under Article 25(1) was not challenged, the enormous growth since the mid 1990s in the number of disputes brought to ICSID has resulted in a number of arbitrations in which the existence of an investment was less clear.Footnote 46 As a result, in a series of recent cases, states have argued that the requirement for the existence of an investment was not met and the tribunal was therefore obligated to decline jurisdiction.Footnote 47
Initially, when faced with these arguments, tribunals undertook case-by-case evaluations, with each tribunal adopting its own preferred approach to the question. However, by the late 1990s, enough decisions had been made by ICSID tribunals that it was possible to discern certain characteristics common among transactions that had been held to constitute an investment under the Convention.Footnote 48 While the lack of precedent in arbitration meant that tribunals were not actually required to conform to these prior decisions, their existence provided a foundation for the reasoning of subsequent tribunals, who were conscious of the need to develop some form of consistency in the interpretation of this requirement.Footnote 49
What resulted was what has come to be known as the ‘Salini’ criteriaFootnote 50 – five elements that can serve as a guide for tribunals as to whether the transaction underlying a given dispute constitutes an investment for the purposes of Article 25(1) of the Convention. As usually listed, the criteria are (i) a certain duration, (ii) a regularity of profit and return, (iii) an assumption of risk by both parties, (iv) a significant financial commitment by the investor, and (v) a contribution to the development of the host state.Footnote 51
The Salini criteria are argued by proponents merely to serve as an indicative list that can assist a tribunal in determining that an investment exists, rather than a set of criteria that must be met by any transaction underlying a dispute able to be arbitrated at ICSID.Footnote 52 However, not all tribunals and commentators agreed with the formalization embodied in the Salini criteria, with some arguing instead that so long as both parties to the dispute agreed that the transaction constituted an investment, then Article 25(1)'s requirement for an investment was met.Footnote 53 As a result, in recent years, both commentators and tribunals have divided into two broad camps, following what are traditionally referred to as the ‘objective’ and ‘subjective’ approaches to the existence of an investment under the Convention.
According to the ‘objective’ approach, the term ‘investment’ in Article 25(1) has a solid and discernible meaning, often described in accordance with the Salini criteria, that serves as an absolute constraint on the jurisdiction of an ICSID tribunal. As a result, even if both parties to a dispute have agreed that the transaction underlying the dispute constitutes an investment and both wish their arbitration to be held under the auspices of ICSID, a tribunal must decline jurisdiction where the objective requirements of an Article 25(1) ‘investment’ have not been met.
By contrast, proponents of the ‘subjective’ approach argue that while the use of the term ‘investment’ in Article 25(1) means that any dispute held at ICSID must indeed arise directly from an investment, the failure of the Convention to provide any definition of ‘investment’ results from a decision to allow parties to a dispute to determine the meaning of the term for themselves. Consequently, when both parties to the dispute have agreed that the transaction underlying the dispute is an investment, an ICSID tribunal should not conduct any further analysis of the existence of the investment and merely hold Article 25(1)'s requirement of an investment to have been met.
The remainder of this article will demonstrate that understanding the ICSID Convention as power-conferring with respect to the definition of ‘investment’ in Article 25(1) helps to resolve this dispute. While proponents of the subjective approach are demonstrably correct that the Convention confers upon the parties to a dispute a central role in providing content to the meaning of ‘investment’ under Article 25(1), they are nonetheless mistaken in viewing the parties to a dispute as the only entities granted such a power. Consequently, while the parties to any individual arbitration do indeed have the central role in determining whether the transaction underlying the dispute constitutes an investment for the purposes of Article 25(1), they can only do so within the constraints of the similar powers granted to other entities.
5. The power of the parties to a dispute to define ‘investment’ under Article 25(1) of the Convention
The first step in establishing which entities have been conferred powers under the ICSID Convention to determine the meaning of ‘investment’ in Article 25(1) must be to demonstrate that Article 25(1) is indeed power-conferring. As noted already, the term ‘investment’ in Article 25(1) is entirely undefined and hence potentially subject to a wide variety of interpretations. However, of itself, this cannot suffice to demonstrate that Article 25(1) is power-conferring, as this vagueness may simply have resulted from poor drafting.
There is, though, clear evidence from the travaux préparatoires to the Convention that the vagueness of the term ‘investment’ in Article 25(1) not only was deliberate, but was adopted with the explicit intent that the content of the term would be filled by entities other than the contracting states to the Convention. Travaux préparatoires must certainly be used carefully, as the VCLT is uniformly recognized as embodying a primarily textual approach to treaty interpretation, in which the actual language incorporated into a treaty has priority over any unexpressed intentions that the contracting states may have had when negotiating the treaty.Footnote 54 Consequently, the use of travaux préparatoires to interpret a treaty can be controversial.Footnote 55
Use of travaux préparatoires is, however, accepted practice in interpreting the meaning of ‘investment’ in Article 25(1) of the ICSID Convention, as the VCLT's rules on treaty interpretation clearly allow that use of travaux préparatoires is acceptable where the meaning of a term is ambiguous or otherwise unclear.Footnote 56 Yet, if any treaty term has ever been ambiguous, it is ‘investment’ in Article 25(1) of the Convention.Footnote 57 Not only does the term have no uniformly accepted meaning outside the Convention, but no indication is given in the Convention itself as to what the term should be understood as meaning. There has been little controversy, then, surrounding the repeated references that have been made to the travaux préparatoires of the ICSID Convention in order to understand the meaning of ‘investment’ in Article 25(1).Footnote 58
Clear evidence that the vagueness of Article 25(1) constitutes a grant of power to entities other than the contracting states to the Convention can be found in a famous statement in the Report of the Executive Directors on the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (‘Report of the Executive Directors’), published alongside the Convention, that ‘[n]o attempt was made to define the term “investment” given the essential requirement of consent by the parties’.Footnote 59 That is, neither states nor investors are required to bring investment disputes to ICSID, even if both the home and host states involved in the dispute are signatories to the Convention. Instead, the Convention requires an independent consent, from both the host state and the Investor, for any specific dispute to be brought before an ICSID tribunal.Footnote 60 As indicated by the above statement in the Report of the Executive Directors, however, the negotiating history of the Convention clearly illustrates that the requirement of this second consent played a primary role in convincing many contracting states to accept the vagueness of the definition of ‘investment’ in Article 25(1). As each state retained the ability to decide for itself which disputes it was willing to consent to bring to ICSID, a precise definition of the term ‘investment’ was felt to be unnecessary, as each state could effectively define the term for itself. In this respect, then, the decision to incorporate into Article 25(1) only the term ‘investment’, with no clarifying language, constituted a conferral of powers upon the parties to any given dispute to define that term for themselves.
Nonetheless, this should not be understood as meaning that the parties to a dispute have unconstrained freedom with respect to the definition of ‘investment’ that they have adopted. After all, the contracting parties could have omitted the term ‘investment’ completely from Article 25(1), thereby giving parties to a dispute total freedom to determine for themselves which of their disputes should be brought before an ICSID tribunal. The deliberate retention of this constraint, then, indicates clearly that while parties to a dispute were to be largely free to define the term ‘investment’ however they wished, it must nonetheless accord with some broader, party-independent meaning of the term.
Indeed, the travaux préparatoires make it clear that the statement quoted above from the Report of the Executive Directors, if taken at face value, is highly misleading, as it is simply not accurate to say that ‘no attempt’ was made to define ‘investment’.Footnote 61 Rather, while a decision was indeed ultimately taken not to adopt a specific definition of ‘investment’, this decision followed a protracted series of disputes regarding the precise definition that the term should be given.
While the Working Paper prepared to serve as an initial foundation for negotiation of the Convention had completely omitted the term ‘investment’ from the jurisdictional constraints of the Centre, this approach was rejected by negotiators and a reference to ‘investment’ was included in all subsequent drafts.Footnote 62 In addition, once the term was incorporated within the jurisdictional constraints of ICSID tribunals, the states negotiating the Convention did not simply proceed to ignore it, accepting that it had no content in and of itself. Rather, the definition of ‘investment’ in Article 25(1) became a point of high contention, with many proposals being made as to the precise way that it should be defined.Footnote 63 It was only when it became clear that no agreement was going to be reached on the precise meaning of an Article 25(1) investment that a proposal was accepted that no definition be adopted at all, leaving the term ‘investment’ included within Article 25(1) but with no indication given as to its meaning.Footnote 64
The travaux préparatoires for the Convention, then, certainly make it clear that the vagueness of ‘investment’ in Article 25(1) was deliberate and was intended to confer power upon the parties to a dispute to determine for themselves whether their transaction constituted an ‘investment’. Nonetheless, they are also clear that this conferral of power was not an unrestricted one, as the insistence of the contracting states that the term ‘investment’ be added to the Convention, combined with the extensive efforts made during negotiations to reach an acceptable definition of the term, is simply inconsistent with such a claim.Footnote 65 The remainder of this article will therefore examine other entities that can be understood to have conferred the power to determine the definition of ‘investment’ in Article 25(1) and thereby place limitations on the power conferred on the parties to a dispute.
6. The power of host states to constrain the definition of ‘investment’ through Article 25(4) of the Convention
The travaux préparatoires of the Convention make it clear that the requirement for a second consent to ICSID arbitration convinced the contracting states that it was safe to confer upon the parties to a dispute the central role in determining the meaning of ‘investment’ under Article 25(1). However, the requirement for a second consent was not the only motivating factor behind the willingness of the contracting states to accept the vagueness of ‘investment’ under Article 25(1). Developing states in particular had expressed concerns over the possibility that they might be pressured by developed states or by powerful investors into consenting to arbitrate a dispute at ICSID. In such a situation, the need for a second consent would provide them with absolutely no protection.
An essential part of the proposal to include ‘investment’ in Article 25(1) as an undefined term, then, was the addition to the Convention of Article 25(4), under which contracting states possess the power to notify the Centre, and thereby the other contracting states, of ‘the class or classes of disputes which it would or would not consider submitting to the jurisdiction of the Centre’.Footnote 66 Through such a notification, states could protect themselves against unacceptable pressures or overly expansive interpretations by tribunals. A state may, for example, submit a notification that it is unwilling to arbitrate at ICSID any disputes arising out of its domestic oil industry, thereby removing the necessity to make such a statement in every document including a consent to ICSID arbitration.Footnote 67
On its face, this provision appears to provide states with a unilateral means of defining ‘investment’ under Article 25(1). On such an interpretation, a state that has declared under Article 25(4) that it will not arbitrate oil-related disputes at ICSID cannot be forced to do so, even were it to include a consent to ICSID arbitration in an oil-related investment contract.
Both the Report of the Executive Directors and subsequent ICSID jurisprudence make clear, however, that Article 25(4) does not have this type of preclusive effect on the jurisdiction of an ICSID tribunal.Footnote 68 Consequently, even were a state to have lodged an Article 25(4) notification that it would not arbitrate oil-related disputes at ICSID, if it then consented to ICSID arbitration of such a dispute, an ICSID tribunal would uphold its jurisdiction despite the Article 25(4) notification.
Nonetheless, the role that Article 25(4) notifications had in convincing the contracting states to accept the vague usage of ‘investment’ in Article 25(1) means that Article 25(4) notifications cannot be taken to be purely symbolic. They could, after all, hardly have served as reassurance to states concerned about the potential breadth of ICSID jurisdiction if they ultimately had no effect on that jurisdiction.
There are, then, three primary interpretations that an Article 25(4) notification can be given: (i) a communication to investors, (ii) a reservation to the Convention, and (iii) an exercise of a power to define ‘investment’ under Article 25(1).
6.1. Article 25(4) notifications as communications to investors
There is no question that an Article 25(4) notification can serve as a communication to investors regarding the types of dispute a given state is willing to arbitrate at ICSID. It merely has to be communicated to a potential investor. In such a case, an Article 25(4) notification would function no differently from any other non-contractual communication made between contractual partners, only having effect to the extent allowed by the applicable contract law.
The difficulty with this interpretation is that an Article 25(4) notification is not something that a state issues as party to a dispute. That is, although it might at some point be communicated to investors, the mechanism for making Article 25(4) notifications does not require this to be done. Rather, an Article 25(4) notification is communicated to ICSID and is subsequently communicated to the other contracting states.Footnote 69 It is, therefore, an action taken by a state in so far as it is a contracting state to the ICSID Convention, not in so far as it is a party to an arbitration agreement with an investor. Indeed, Article 25(4) does not require that investors be notified of the existence of an Article 25(4) notification for it to have its effect.
It is, therefore, simply inconsistent with the structure of Article 25(4) notifications for them to be understood as merely communications to investors, serving only to modify a state's consent to arbitration.
6.2. Article 25(4) notifications as reservations to the Convention
The second possibility is to consider Article 25(4) notifications as reservations to the Convention. Under such an interpretation, Article 25(4) gives each contracting state to the Convention the right to exclude certain kinds of transaction from the Article 25(1) definition of investment, so long as the exclusion is not inconsistent with the ‘object and purpose’ of the Convention.Footnote 70
Understanding an Article 25(4) notification as a form of reservation to the Convention, however, is inconsistent with the clear recognition in the Report of the Executive Directors and subsequent case law that an Article 25(4) notification cannot override a clear subsequent consent to ICSID arbitration. A reservation to a treaty, after all, affects both a state's rights and its responsibilities under a treaty. That is, it is not something that a state can turn on and off at will. Once a reservation is made, it is applicable until the state in question withdraws it.Footnote 71 Consequently, if an Article 25(4) notification constituted a reservation to the Convention, states could not arbitrate any dispute at ICSID that was inconsistent with that reservation, even if they consented to do so.
As already noted, however, this is simply not the way that Article 25(4) notifications function. Rather, even if a state has an Article 25(4) notification in place, it can still arbitrate at ICSID a dispute inconsistent with that notification. Moreover, that a state has consented to arbitrate at ICSID in a manner inconsistent with its Article 25(4) notification has no effect upon that notification's existence. The arbitration merely constitutes an exception to the notification, not a means of terminating it.Footnote 72
Furthermore, a fundamental element of the legal rules applicable to treaty reservations is the right of other states party to the treaty to object to the reservation. There is, however, no available mechanism for any contracting state to the ICSID Convention to object to an Article 25(4) notification. Indeed, a state could potentially become party to the Convention and then submit a notification effectively precluding every form of investment dispute from being arbitrated at ICSID. A reservation of this breadth would be challenged and removed as clearly inconsistent with the ‘object and purpose’ of the Convention. As a notification under Article 25(4), however, it could remain.
Article 25(4) notifications, then, simply do not function in a manner consistent with the recognized legal rules applicable to reservations to treaties. They cannot, therefore, properly be understood as constituting reservations.
6.3. Article 25(4) notifications as an exercise of a power to define ‘investment’ under Article 25(1)
An Article 25(4) notification must be understood, then, as something more than merely informational, but less than a reservation. That is, they are best understood as a conferral of power on each individual contracting state, to place limits on the definition of ‘investment’ in Article 25(1) that will be applicable in any dispute in which it is a party. They can, therefore, include any substantive restrictions that the state wishes to include, without thereby being subject to a challenge from another contracting state. Moreover, an Article 25(4) notification does not prevent a state's arbitrating a dispute at ICSID that is inconsistent with its notification. Such an arbitration simply reflects a decision by the state not to insist upon its power to preclude arbitration of such disputes.
Recognizing Article 25(4) notifications as reflecting a conferral of power, however, nonetheless leaves the question of how they should be understood as operating. Adapting a proposal initially advanced by Julian Mortenson, the most persuasive interpretation is that while an Article 25(4) notification can give specificity to the limits of a vague consent to arbitration, it cannot do so where the consent to arbitration clearly covers the dispute in question.Footnote 73
This interpretation of Article 25(4) notifications can be best illustrated through an example. Although the example will be based upon consents to arbitration contained in BITs between states, the same analysis can be applied to consents to arbitration contained in agreements directly between investors and host states.
Presume that after ratifying the ICSID Convention, State A provides a notification under Article 25(4) that it does not agree to arbitrate any oil-related disputes at ICSID. However, it has already signed a bilateral investment treaty with State B that explicitly includes oil-related disputes, as well as including a consent to arbitrate any disputes at ICSID. In addition, State A subsequently signs a bilateral investment treaty with State C that also includes consent to ICSID arbitration and, again, specifically includes oil-related disputes in its definition of ‘investment’. Finally, State A then signs a bilateral investment treaty with State D that, again, includes a consent to ICSID arbitration, but makes no reference to oil in its broad and generalized definition of investment.
Under this example, State A's Article 25(4) notification would have no effect on arbitrations brought under the pre-existing bilateral investment treaty with State B. State A lacks the power unilaterally to alter the terms of its agreement with State B and hence is bound, despite its Article 25(4) notification. Moreover, as the Article 25(4) notification reflects a power possessed by State A to define ‘investment’ under Article 25(1), it is exercisable at State A's discretion. Consequently, as State A did not denounce its BIT with State B, the best understanding of the conflict between State A's Article 25(4) notification and its BIT with State B is that State A chose not to exercise that power with respect to any arbitrations brought under the BIT with State B.
Similarly, State A's Article 25(4) notification will have no effect on arbitrations brought under the subsequent bilateral investment treaty with State C. As an Article 25(4) notification reflects a power possessed by State A, rather than an agreement between State A and the other contracting states to the Convention, State A retains the right to act inconsistently with its notification whenever it wishes to do so. Moreover, State A's agreement to arbitrate oil-related disputes with investors from State C does not cancel or modify State A's Article 25(4) notification, which continues to be applicable in the future.
However, State A's Article 25(4) notification has an entirely different impact on arbitrations brought under the bilateral investment treaty with State D, due to the vagueness of the definition of ‘investment’ contained in the treaty. This vagueness means that there is inadequate evidence that State A did not intend to have its power to define ‘investment’ under Article 25(1) be effective with respect to disputes arising under the A–D BIT. Consequently, the Article 25(4) notification applies to all claims brought under the A–D BIT. It does not modify the terms of that BIT and so does not preclude oil-related disputes being arbitrated under any forum other than ICSID that is specified in the BIT. However, it does preclude any claims under it being arbitrated at ICSID.
Article 25(4), then, is best understood as constituting a conferral of power to individual contracting states to define ‘investment’ under Article 25(1). As such, it can be excluded from any individual transaction, when desired by the state making the notification, without affecting the notification's ongoing effectiveness.
7. The power of home and host states to define ‘investment’ in Article 25(1) through mutual agreement
As has just been shown, home states can unilaterally alter the definition of ‘investment’ under Article 25(1) via an Article 25(4) notification. This section of the article will demonstrate that the Convention also provides a mechanism for the definition of ‘investment’ to be defined bilaterally between a home and a host state. Moreover, unlike the unilateral definition available via Article 25(4), a bilateral agreement between the home and host states is binding on both and hence is not overridden even when the host state includes a clear but inconsistent consent in an agreement directly with an investor.
Unlike the other powers examined in this article, the power of the home and host states to agree on a binding definition of ‘investment’ does not arise directly out of the incorporation of vagueness into Article 25(1). Instead, it comes from a completely separate article of the Convention – Article 64. As will be demonstrated below, Article 64 gives any grouping of contracting states to the Convention the power to agree on an interpretation of any element of the Convention, which will then be binding between them. Consequently, although this power is not specifically one to define ‘investment’ under Article 25(1), it does enable states to agree on such a definition.
As discussed earlier, prior to the creation of ICSID, investors with a legitimate claim against a foreign government were usually forced to rely upon often unsatisfactory diplomatic protection efforts by their home state. Unhappiness with this system was the most important motivation behind the creation of ICSID.
Given this context, although the potential risks for a host state of consenting to arbitrate with an investor at ICSID are clear, a home state is itself not unaffected by its ratification of the ICSID Convention. A home state that ratifies the ICSID Convention does not just risk having claims brought against it as a host state. It also surrenders the traditional power it had to use its investors’ claims against host states for its larger political purposes. After all, there is no provision within the ICSID Convention that allows a home state to prevent one of its investor's bringing a claim against a host state or even just allows it to intervene in an ICSID arbitration in order to express its own views on the dispute.Footnote 74 Ratification of the ICSID Convention, then, represents a significant loss of power even for states that expect predominantly to be home states in disputes, rather than host states.
However, although a home state does not possess any power to intervene directly in an ICSID arbitration between one of its investors and a foreign state, Article 64 means that a state need not be entirely without influence in such a dispute. Under this Article, contracting states are permitted to resolve any dispute regarding the interpretation of the Convention through negotiation, prior to instituting a case at the International Court of Justice (ICJ).Footnote 75 While commentary on this Article has focused on its grant of adjudicatory power to the ICJ, the recognition of the power of ICSID contracting states to resolve disputes through negotiation is more important in the present context.Footnote 76
Although Article 64 has never been used, either as the foundation for a case at the ICJ or as the ground for a negotiated agreement, its language is nonetheless clear as to the power it grants to groups of ICSID contracting states. While a provision acknowledging that the contracting states to a treaty retain the power to agree on the interpretation of any element of that treaty might be unremarkable,Footnote 77 Article 64 expressly states that the power to negotiate agreed interpretations of Convention provisions is exercisable by limited groups of contracting states and does not require the participation of all ICSID contracting states.Footnote 78
As a result, Article 64 confers on groups of two or more contracting states the power to adopt an interpretation of the Convention that will be applicable between them, but that is potentially inconsistent with an interpretation adopted by one or more other groups of contracting states. It is, therefore, a mechanism by which a single treaty can ultimately be transformed into what is effectively a variegated set of treaties, consistent in most provisions, but diverging in important ways on certain points.
It should be emphasized, however, that Article 64 only gives contracting states the power to negotiate agreements regarding the ‘interpretation or application’ of the Convention. Consequently, it cannot be a means of altering or terminating the Convention. The Convention, therefore, remains a single treaty, subject to a variety of interpretations, rather than actually constituting a linked set of bilateral treaties.
With respect to the definition of ‘investment’ in Article 25(1), then, Article 64 confers on groups of contracting states the power to negotiate an agreed definition that will be applicable in any arbitration between any contracting state party to that agreement and an investor from another contracting state also party to the agreement.
Article 64 does not itself specify any formal requirements for such an agreement. Consequently, while an agreement to a particular interpretation of ‘investment’ in Article 25(1) may take the form of an independent document, explicitly referring to Article 64 of the Convention, nothing in Article 64 requires that this be so. Rather, so long as an agreement between the home and host states was reached through ‘negotiation’ or ‘another method of settlement’ agreed between the states concerned, Article 64 makes this agreement a binding interpretation of the Convention.
The lack of formal requirements for an Article 64 agreement is important because, as discussed earlier, a significant majority of recent ICSID arbitrations have been based on BITs negotiated between states, rather than on agreements directly between the investor and the host state. In contrast to the ICSID Convention, however, these BITs often include very detailed definitions of what constitutes an ‘investment’ under the BIT.
The definition of ‘investment’ in a BIT is, of course, not directly an agreement on the definition of ‘investment’ in Article 25(1) of the Convention. Consequently, it might initially seem that BIT definitions of ‘investment’ are simply irrelevant to Article 64, which only applies to agreements relating to the ‘interpretation or application’ of the Convention.Footnote 79
However, this ignores the role that Article 25(1)'s definition of ‘investment’ plays in the Convention. As noted above, negotiations regarding how to define ‘investment’ in Article 25(1) did not focus upon what an ‘investment’ should or should not be understood as being. Rather, incorporation of the requirement that a dispute ‘arise directly out of an investment’ was based on the more practical question of which types of disputes states were willing to arbitrate directly with investors.
That is, the term ‘investment’ in Article 25(1) ultimately has a functional definition. It does not refer to a substantive type of entity or activity, but rather to the willingness of states to arbitrate directly with investors. Recognition of this point is important because it means that an agreement between two or more contracting states regarding the meaning of ‘investment’ in Article 25(1) does not constitute an agreement on the particular types of things that are or are not investments under the Convention. It is, rather, an agreement regarding the category or categories of disputes that the states concerned are willing to have arbitrated at ICSID between one of the states party to the agreement and an investor from another.
Consequently, when two states enter into an agreement that includes both consent to ICSID arbitration and a definition of investment that acts as a jurisdictional constraint to the states’ consent to ICSID arbitration, they have thereby reached agreement upon precisely what constitutes the meaning of ‘investment’ under Article 25(1). They were, of course, likely not attempting to do so, as they were merely placing a jurisdictional constraint on their own consent to arbitration under the BIT. However, once it is recognized that ‘investment’ under Article 25(1) is a functional term relating to the willingness of states to arbitrate directly with investors, rather than a substantive type of transaction, then evidence of this willingness can provide content to the term no matter its source.
In accordance with the power granted to contracting states by Article 64, then, a BIT that includes both a definition of ‘investment’ and a consent to ICSID arbitration constitutes an agreement between the two states party to the BIT that, for the purposes of disputes arising between one of those states and an investor from the other, ‘investment’ under Article 25(1) of the Convention has the content of the definition provided in the BIT.
Of course, this argument will have no direct effect where jurisdiction of an ICSID tribunal is based on the BIT in question, as, if the transaction does not fall within the definition of ‘investment’ in the BIT, then an ICSID tribunal has no jurisdiction, no matter what ‘investment’ means in Article 25(1). It can, however, have important consequences where the consent to arbitration is contained in an agreement directly between the investor and the host state, but the transaction underlying the dispute does not qualify as an ‘investment’ under an existing BIT between the host and home states.
In such a situation, the definition of ‘investment’ in the BIT becomes incorporated into the ICSID Convention as an agreed refinement of the definition of ‘investment’ in Article 25(1), for any disputes involving one of the two states party to the BIT and an investor from the other. Consequently, if the transaction underlying the agreement between the investor and the state does not fall within the definition included in the BIT, it thereby also does not fall within Article 25(1) of the Convention.Footnote 80
That is, whenever a BIT includes both a consent to ICSID arbitration and a definition of ‘investment’ that serves as a jurisdictional limit on that consent, neither state party to the BIT may arbitrate a dispute at ICSID that arises out of a transaction that would not qualify as an ‘investment’ under the BIT, even though the arbitration is brought based on a consent in a separate agreement directly between the investor and the host state.
8. The minimum content of the definition of ‘investment’ in Article 25(1)
This article has been concerned with elucidating the entities that have been granted the power to define ‘investment’ under Article 25(1) of the Convention and the means by which they can exercise that power. However, the history of the negotiations that resulted in Article 25(1) makes clear that these powers were not unconstrained. That is, while entities other than the contracting states to the ICSID Convention have been granted the power to define ‘investment’ under Article 25(1), they have not been given complete freedom to do so. Such a claim would, after all, make nonsense of the insistence that the term ‘investment’ be included within Article 25(1), as it would eliminate the ability of the term to act as a constraint. There must, then, be some element of control over the definition of ‘investment’ that has been retained by the contracting states to the Convention – some ‘minimum content’ that even those entities granted the power to define ‘investment’ do not have the power to eliminate. The concluding section of this article will be dedicated to explaining the nature of this ‘minimum content’ and how it can be identified.
8.1. The problem of interpretative subjectivity
Before addressing directly the manner in which the minimum content of Article 25(1)'s definition of ‘investment’ should be analysed, it is important to note the unavoidable need for a tribunal-independent understanding of this minimum content.
That is, even proponents of the ‘subjective’ approach to the interpretation of Article 25(1) concede that there must be some outer limit to the meaning of ‘investment’, such that even party agreement to bring certain types of dispute to ICSID cannot suffice to ground a tribunal's jurisdiction.Footnote 81 However, it is clearly untenable to maintain that the decision not to incorporate a precise definition of ‘investment’ into Article 25(1) was taken because it was decided that the meaning of ‘investment’ was clear and no further specificity was required. Although the principle architect of the Convention, Aron Broches, did indeed suggest that one justification for omitting a detailed definition of ‘investment’ was precisely that the existence of an investment was readily recognizable, the inability of the negotiating states to reach a consensus on the definition despite extensive efforts indicates clearly that if any consensus existed on the definition of ‘investment’, it was at a very basic level.Footnote 82
There is, then, simply no single clear definition of ‘investment’ to which tribunals attempting to interpret Article 25(1) can appeal. As a result, each tribunal has, of necessity, been forced to rely upon its own intuitive understanding of what is or is not ‘obviously’ not an investment.Footnote 83
The difficulty this creates is that while the Convention clearly conferred upon the parties to a dispute the power to define ‘investment’, it did not do the same for arbitral tribunals. Yet, any tribunal that bases an interpretation of Article 25(1) on the intuitions of its members regarding what does or does not constitute an investment has indeed exercised this power. That is, the tribunal has determined the meaning of ‘investment’ under Article 25(1), but not through interpretation of the Convention or identification of the views of the parties to the dispute. It has, rather, simply imposed the view that its own members prefer.
The difficulty in such a view is apparent. The arbitrators have been selected by the parties in a procedure that allows them to be selected because of certain substantive views they hold, including on the definition of ‘investment’ in Article 25(1).Footnote 84 However, when ruling on the outer boundaries of Article 25(1)'s definition of ‘investment’, tribunals are determining precisely the minimum content of the term that was explicitly reserved from the definitional power conferred on the parties. Consequently, by appealing to their own intuitions in determining the ‘minimum content’, the tribunal allows the parties to take control of precisely that element of the definition of ‘investment’ that the contracting states restricted from party control.
This situation causes a serious practical problem because, under Article 54 of the Convention, contracting states, even though not themselves either the host state or the home state in a given dispute, undertake to enforce the pecuniary obligations of any ICSID award.Footnote 85 Consequently, unless there is a minimum content to the definition of ‘investment’ that is beyond the power of either the parties or the home and host states to define, other contracting states will be obligated to enforce awards resulting from disputes concerning transactions that are plausibly an ICSID investment only because the parties or the home and host states agreed that they wanted it to be. This approach entirely eviscerates the constraint that ICSID be a centre for the resolution of investment disputes, and only investment disputes.Footnote 86
Despite the variety of ways already discussed in which the host state, home state, and investors can give content to Article 25's definition of investment, then, it is clear that there must be a constraint on their freedom. Moreover, this constraint must be one objectively ascertainable and not merely confined to the discretion of the arbitrators in each dispute.
Nonetheless, this does not entail that the minimum content of ‘investment’ in Article 25(1) must be invariant. As the next subsection of this article will argue, the minimum content of an Article 25(1) ‘investment’ is best understood as being evolutionary in nature, changing over time to reflect changing understandings of the meaning of the term. As a result, the meaning of the term is to be determined through an examination of the contemporary practice of all contracting states to the Convention, rather than by attempting to discern the meaning of ‘investment’ at the time the Convention was signed or by imposing any other form of invariant definition.
8.2. Determination of the minimum content of an Article 25(1) investment
As has been argued above, the term ‘investment’ in the Convention has content beyond that provided by either party agreement or agreement between the host and home states. However, it is also indisputable that no single definition of the term was acceptable to all contracting states when the Convention was negotiated. The best insight into the nature of the minimum content of an Article 25 investment comes, therefore, from considering the terms under which the contracting states to the Convention would plausibly see the term ‘investment’ as providing a substantive jurisdictional constraint, even without the incorporation of a clear definition.
The inability to reach agreement on even the most basic elements of a definition preclude the notion that the term was simply intended to refer to some extra-Convention definition of ‘investment’, such as might be found in economics or some other branch of international economic law.Footnote 87 In addition, the willingness of the contracting states to allow parties to disputes and home and host states to define ‘investment’ differently at different times and contexts is inconsistent with the idea that they simultaneously agreed to limit their own impact upon the definition of ‘investment’ to a single usage currently being made of the term at the time the Convention was concluded.
Rather, as has already been noted, the travaux préparatoires show that the concern motivating the disputes over the definition of ‘investment’ in Article 25(1) concerned the obligation of states to enter into investor–state arbitration, not what does or does not actually constitute an investment. Recognizing this fact, it is clear that a tribunal attempting to determine whether a transaction falls within the minimum content of the term ‘investment’ in Article 25(1) of the Convention should similarly focus on identifying whether that transaction is a type of transaction that contracting states to the Convention are currently willing to submit directly to investor–state arbitration. This can be done through examining whether the transaction would constitute an ‘investment’ under other agreements concluded by the contracting states to the Convention that also include consent to investor–state arbitration.
Importantly, however, as the concern is not with the definition of ‘investment’, but with the willingness to submit to investor–state arbitration, the only agreements that are relevant to this determination are those that include consent to investor–state arbitration.Footnote 88 Where the term ‘investment’ is used in agreements that do not include consent to investor–state arbitration, even if they are also in some way related to international investment, such usages of ‘investment’ are not directly relevant for determining the meaning of the term under Article 25(1).
Of course, the practice of states in this respect changes constantly and has evolved considerably in the period since the ICSID Convention came into force.Footnote 89 The minimum content of ‘investment’ under Article 25(1), then, must be understood as evolutionary, reflecting at any given time the types of ‘investment’ that states are willing to arbitrate directly with investors.Footnote 90
The consequence of such a conclusion is that it is simply impossible to make any form of absolute statement regarding what is or is not an investment under Article 25(1) of the Convention. Rather, while a certain form of transaction may not currently constitute an Article 25(1) investment, changing practice may result in its becoming one in the future. Similarly, a type of transaction currently regarded as a paradigm form of investment may well lose its status in the future if practice begins to exclude it.
The question remains, however, how widespread the agreement to arbitrate a particular type of transaction must be before it is taken to have been incorporated into Article 25(1)'s definition of ‘investment’. It is tentatively suggested here that the appropriate standard is that used to determine whether a rule of customary international law has developed. That is, if the inclusion of a particular type of transaction within the definition of ‘investment’ in relevant international agreements has not reached a level of uniformity and diversity that would support, in another context, a finding of the existence of a rule of customary international law, then the type of transaction in question should not be found to have been incorporated into Article 25(1)'s definition of ‘investment’.Footnote 91 Only with such a standard can the consent of the large number of states signatory to the ICSID Convention be respected.
9. Conclusion
States have long used treaties as a means of protecting or advancing the interests of entities who themselves have no direct role in the treaty's formation or execution. The goal of this article has been to emphasize that there is, however, a subset of treaties that are best understood not merely as attempting to benefit entities other than the states creating them, but as conferring on those entities significant powers to affect the treaty's operation. Moreover, recognition that a treaty is power-conferring can significantly affect the interpretation of a treaty, with the traditional focus on the goals and intentions of the states party to the treaty no longer always being appropriate. This has been illustrated through an analysis of one aspect of power-conferring that occurs in the ICSID Convention. The ICSID Convention is, however, by no means the only power-conferring treaty in existence. Moreover, adoption of such treaties can reasonably be expected to increase in the future, in reflection of the partial retreat of the state from many areas of traditional governmental control.
The recognition that a treaty is ‘power-conferring’, however, does not ultimately determine anything specific about the way the treaty should be interpreted, but merely indicates an important element of the treaty that must be taken into consideration. Each power-conferring treaty must, therefore, be analysed on its own terms, with the traditional rules of treaty interpretation providing the means of determining what powers have been conferred, and on whom.
Nonetheless, as the analysis of the power-conferring aspects of the definition of ‘investment’ under Article 25(1) of the ICSID Convention has attempted to demonstrate, recognition that a treaty is power-conferring can substantially alter the way that a treaty is interpreted, and therefore cannot merely be subsumed under a traditional contracting state-centred approach to treaty interpretation.
