Definition of “Investment”—A Voice from the Eye of the Storm

Abstract

This article traces the development of the definition of “investment” under Article 25(1) of the ICSID Convention. It proposes that the definition should act as an outer limit to the usually broad definition of investment (encompassing every kind of asset) in international investment agreements (IIAs). The article discusses the various characteristics (hallmarks) of an investment which should constitute the definition. It argues that the hallmark of “significant contribution to economic development” can be refined to reduce uncertainty while giving effect to the intent of the ICSID contracting states by drawing a distinction between an ordinary commercial transaction and an investment. Recent IIA definitions of investment adopting Salini hallmarks show that states adopt the “every kind of asset” definition of investment in IIAs out of a concern that the form which the investment may take should not be restricted and that states do not necessarily view the Salini hallmarks as unwelcome.

I. Definition of “investment”

The definition of “investment” is important to International Centre for Settlement of Investment Disputes (ICSID) arbitrations because unless an asset or economic activity constitutes an investment under Article 25 of the Convention on the Settlement of Investment Disputes Between States and Nationals of Other States [ICSID Convention],1 it is not subject to ICSID jurisdiction. Unfortunately, the drafters of the ICSID Convention chose not to define the meaning of investment within the Convention, sparking off a stormy definitional debate which rages today.

Article 25(1) of the ICSID Convention merely states:

(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. When the parties have given their consent, no party may withdraw its consent unilaterally. [Emphasis added.]

The ordinary meaning of investment is defined as “the action or process of investing. A thing worth buying because it may be profitable in future”.2 “Invest” is defined as:

v.1 put money into financial schemes, shares or property with the expectation of achieving a profit. Devote (one’s time or energy) to an undertaking with the expectation of a worthwhile result. (Invest in) informal: buy (a product) whose usefulness would repay the cost.3

Some tribunals, such as the majority of the Annulment Committee of the Malaysian Historical Salvors, SDN, BHD v. Malaysia [MHS Annulment] in 20094 and Biwater Gauff (Tanzania) Limited v. United Republic of Tanzania [Biwater],5 have rejected the outer limit approach,6 holding that there is no definition of investment in Article 25 of the ICSID Convention that is separate from the definition of investment in the relevant treaty or agreement from which the dispute arises [the bilateral investment treaty (BIT) definition].7

Other tribunals have accepted that “investment”, as used in Article 25 of the ICSID Convention, has its own definition and criteria separate from the BIT definition. However, many of these tribunals differ as to which criteria should constitute the Article 25 definition and whether these criteria should be considered as jurisdictional or simply treated as the “typical characteristics” of an investment. This causes complications when the tribunals try to measure whether a particular transaction meets the definition of investment under Article 25 of the ICSID Convention.

The MHS Annulment 8 is another development in the definitional debate. However, one would be wrong to think that this storm has ended with the MHS Annulment Committee’s decision—we are merely in the eye of the storm. Indeed, the case of Saba Fakes v. Republic of Turkey [Saba Fakes]9 released just as this article was being published, now supersedes the MHS Annulment as the last word (for the time being) on the definition of investment under Article 25 of the ICSID Convention. As the awareness and popularity of ICSID arbitration grows, claimants who have entered into transactions which are not the traditional infrastructural or mining type investments will continue to test the boundaries of the definition of investment.

This multifaceted definition of investment reminds us that the word “investment” can be understood in many ways, even by applying its plain (dictionary) meaning. It is therefore necessary to consider the context in which the word “investment” is used in Article 25 of the ICSID Convention. It is necessary to consider previous ICSID decisions on investment because, despite such decisions having no strict binding effect, there is a growing trend in ICSID jurisprudence for tribunals to recognize the role of precedent in ICSID cases and even holding that

unless there are compelling reasons to the contrary, [an ICSID tribunal] ought to follow solutions established in a series of consistent cases that are comparable to the case at hand, subject to the specificity of the treaty under consideration and the circumstances of the case.10

Hence, the storm surrounding the definition of investment in Article 25 of the ICSID Convention must be clarified and discussed. There are three aspects to this debate, namely:

1. 1. Whether the definition of investment in Article 25 of the ICSID Convention acts as an outer limit to any bilateral investment treaty [BIT] definition of investment [the outer limit approach];

2. 2. Assuming that Article 25 does act as an outer limit to any BIT definition of investment, what the proper definition of the term “investment” under Article 25 of the ICSID Convention is; and

3. 3. Whether the test in Salini Costruttori S.p.A. and Italstrade S.p.A. v. Kingdom of Morocco 11 [the Salini test] adequately represents such a definition with the result that failure to satisfy the Salini test will mean that there is no investment under Article 25 of the ICSID Convention [the “jurisdiction” approach]. An alternative approach is to regard the Salini test only as a yardstick indicating the typical characteristics of an investment [the “typical characteristics” approach].

This article will focus its discussion only on (1) and (2) above as the principle author has, in writing the MHS Award in 2007, already taken the view that (3) is academic in scope.12 In doing so, the authors will consider the relevant case-law as how investment in Article 25 of the ICSID Convention should be interpreted. It is hence necessary to begin with an explanation of the outer limit approach as well as the hallmarks of an investment identified in the seminal case of Salini.13

A. The Outer Limit Approach

The Salini Tribunal held, following the decision of the tribunal in Československa Obchodní Banka, a.s. v. Slovak Republic [CSOB],14 that it had to apply a twofold test in order to determine its jurisdiction. The twofold test involved determining:

1. 1. Whether the dispute arises out of an investment within the meaning of the ICSID Convention; and, if so,

2. 2. Whether the dispute relates to an investment as defined in the parties’ consent to ICSID arbitration, in their reference to the BIT, and the pertinent definitions contained in Article 1 of the BIT.

The twofold test is a manifestation of the outer limit approach. In CSOB, where the test was first derived, the CSOB Tribunal explained:

The concept of an investment as spelled out [in Article 25(1) of the ICSID Convention] is objective in nature in that the parties may agree on a more precise or restrictive definition of their acceptance of the Centre’s jurisdiction but they may not choose to submit disputes to the Centre which are not related to an investment.15

In other words, an agreement of the parties describing their transaction as an investment was not conclusive in resolving the question whether the dispute involves an investment under Article 25(1) of the ICSID Convention.

Some cases, such as Phoenix Action Ltd v. Czech Republic [Phoenix Action],16 adopt the outer limit approach on the basis that bilateral agreement between an investor and a state cannot contradict the definition of investment in the ICSID Convention, a multilateral treaty. As long as it fits within the ICSID notion, the BIT definition is acceptable; it is not if it falls outside such a definition. For example, if a BIT provides that ICSID arbitration is available for sales contracts which do not imply any investment, such a provision cannot be enforced by an ICSID Tribunal.

Other tribunals, such as those in CSOB and RSM Production Corporation and Others v. Grenada [RSM]17 take the approach that an express acceptance and specific consent to ICSID jurisdiction (such as an ICSID arbitration clause in a direct contract between investor and the state) creates a strong presumption that parties considered their transaction to be an investment within the meaning of the ICSID Convention. RSM goes further to qualify that:

only where the economics of the disputed transaction are clearly lacking one or more of the recognized characteristics of an investment should an ICSID tribunal decline to enforce the parties’ will and find that it has no jurisdiction; other than that, the true abuse of power would be to defeat their expectations.18

In other words, under the RSM approach, a broad BIT definition would not be taken to trump the definition of investment under Article 25 of the ICSID Convention. However, a specific consent or agreement between parties to refer disputes arising out of a particular agreement to ICSID arbitration can create a presumption that the definition of investment under Article 25 of the ICSID Convention has been fulfilled, even though not all of the recognized characteristics of an investment have been satisfied on the facts. The authors find that this approach strikes the right balance between flexibility and the legitimate expectations of the ICSID contracting states.

B. The Salini Test

The Salini test stands for two propositions: first, investment in Article 25 of the ICSID Convention has an intrinsic definition of its own; and second, this definition holds four characteristics of an investment within the meaning of the ICSID Convention. The Salini Tribunal famously identified these four characteristics [hallmarks] as:

1. 1. contribution;

2. 2. a certain duration of performance of the contract;

3. 3. a participation in the risks of the transaction; and

4. 4. contribution to the economic development of the host state of the investment (derived from the ICSID Convention’s preamble).19

The Salini Tribunal considered that these four hallmarks are interdependent, such that one may depend on the other and that the various hallmarks should be assessed globally.20

The requirement of “regularity of profit and return” was not mentioned by the Salini Tribunal although it had been mentioned in Fedax N.V. v. Republic of Venezuela [Fedax].21 However, by the time the award in Joy Mining Machinery Limited v. Arab Republic of Egypt [Joy Mining]22 was issued, regularity of profit and return was included as a fifth hallmark in the definition of investment, although this requirement was not popularly adopted by later tribunals. In addition, the case of Phoenix Action 23 recently added two further requirements. It remains to be seen whether the two further Phoenix Action requirements would be followed in future ICSID awards. Currently, the hallmarks of an investment identified by the ICSID jurisprudence appear to be:

1. 1. a certain duration of performance;

2. 2. assumption of risks by both sides;

3. 3. a substantial commitment or contribution24 which should be looked at not only in financial terms but also in terms of know-how, equipment, personnel, and services;25

4. 4. contribution or significance to the (economic) development of the host state;26

5. 5. regularity of profit and return (added to the original Salini test by Joy Mining);

6. 6. investment made in good faith (added by Phoenix Action); and

7. 7. investment made in accordance with the law (added by Phoenix Action).

Of the hallmarks listed above, the first three are commonly accepted by ICSID Tribunals which agree in principle with the Salini twofold test. There is less agreement with respect to the remaining hallmarks, especially with the hallmark of contribution or significance to economic development.

Most tribunals27 which adopt the outer limit approach and the Salini test recognize the need to distinguish investments from ordinary sales contracts28 and the prevalent problem of BITs and international investment agreements (IIAs) defining investment so broadly that even ordinary sales and service contracts could qualify as an investment.

II. The fate of the outer limit approach

The majority decision of the MHS Annulment stated that the definition of investment for the purposes of ICSID arbitration must yield to the definition of investment in any particular BIT. This is similar to the approach adopted by the tribunal in SGS Société Générale de Surveillance S.A. v. Islamic Republic of Pakistan [SGS],29 as well as a line of other cases above. This creates the immediate expansion of the claims available for ICSID arbitration.

According to the MHS Annulment, the BIT is the:

[58] medium through which the Contracting States have given their consent to the exercise of jurisdiction of ICSID … [71] By terms of their consent, they could define jurisdiction under the convention … [73] some 2,800 bilateral, and three important multilateral, treaties have been concluded, which characteristically define investment in broad, inclusive terms … It is those bilateral and multilateral treaties which today are the engine of ICSID’s effective jurisdiction. To ignore and depreciate the importance of the jurisdiction they bestow upon ICSID, and rather to embroider upon questionable interpretations of the term “investment” as found as Article 25(1) of the Convention, risks crippling the institution.30 [Emphasis added.]31

In contrast, a dissenting member of the MHS Annulment Committee, Judge Mohamed Shahabuddeen, was of the view that the word “investment” in Article 25(1) of the ICSID Convention had to be construed in order to place an outer limit to an ICSID investment beyond which party agreement to what constitutes an investment would be ineffectual to create an ICSID investment. He drew a distinction between the contents of jurisdiction and the limits within which those contents exist.

The dissenting opinion in the MHS Annulment represented a significant and irreconcilable difference in principle between the tribunal members. Indeed, the debate over the outer limit approach in the MHS Annulment was, as Judge Shahabuddeen described in his dissent, the result of “a titanic struggle between ideas, and correspondingly between capital exporting countries and capital importing ones”.32 He observed:

A reasonable inference is that Contracting States [to the ICSID Convention] did not agree that these burdens on them would apply to benefit transactions which did not promote the economic development of the host State. It is difficult to see why a purely commercial entity, intended only for the enrichment of its owners and not connected with the economic development of the host State, is entitled to bring before ICSID a dispute concerning an investment in the host State. Schreuer notes that “it was always clear that ordinary commercial transactions” would not be covered by the Centre’s jurisdiction … It is pedantic to spend time on the meaning of “ordinary commercial transactions”.33

The authors agree with Judge Shahabuddeen and disagree with the MHS Annulment majority decision that the word “investment” in Article 25 of the ICSID Convention has no meaning independent of the BIT definition of investment because many BITs contain an overbroad definition of investment that would capture transactions which the drafters of the ICSID Convention have arguably never contemplated would constitute an investment under the ICSID Convention.

A salvage award like the one in MHS, a shell company like the one in Phoenix Action buying shareholdings for the purpose of bringing an ICSID arbitration, and commercial bank guarantees like the one in Joy Mining are clearly not the kind of investments which states had intended to be at the heart of any ICSID arbitration. Even if ultimately dismissed, defending such claims would be a drain on state resources. Without Article 25(1) as a device to control access to ICSID arbitration, using the BIT definition alone opens a possible floodgate of arbitration claims against the state because most BIT definitions of investment are broad to the point of being unhelpful. For example, Article 1(a) of the UK-Malaysia BIT defines investment as “every kind of asset and in particular, though not exclusively, includes … (iii) claims to money or to any performance under contract, having a financial value”.34 Although there are differing views as the definition of investment under Article 25(1) of the ICSID Convention, the hallmarks are capable of providing a common platform for discussion and coherent development by all ICSID Tribunals, unlike the BIT definitions of investment, which may differ from case to case.

As an update, the most recent decision of Saba Fakes 35 also adopts the outer limit approach.

A. Why Future Tribunals May Hold that the Outer Limit Approach Should Apply

1. It is difficult to imagine that states intended very broad BIT definitions such as “every kind of asset” to constitute a definition of investment for the purposes of ICSID jurisdiction

As Schreuer observes:

[T]he BIT clause providing for ICSID jurisdiction is drafted in general terms referring to future disputes …

Almost all BITs contain definitions of the term investment. In modern BITs, these have very similar features, which may be described in generalized way. They are introduced by a broad general description followed by a non-exhaustive list of typical rights. The general description frequently refers to “every kind of asset”. The list of typical rights usually includes:

• Traditional property rights;

• Participation in companies;

• Intellectual and industrial property rights;

• Concession or similar rights.36 [Emphasis added.]

Model BITs from the UK, Germany, France, and the Netherlands contain definitions similar to the one described by Schreuer in the above paragraph.37

The UK-Malaysia BIT in the MHS Annulment case also defined investment as “every kind of asset” with a similar list of typical rights as Schreuer noted.38 The MHS Annulment found that a salvage contract was “a claim to money and to performance under a contract having financial value; the contract involves intellectual property rights; and the right granted to salvage may be treated as a business concession conferred under contract”.39 On a literal wording of a typical BIT which allows for investment disputes to be referred to ICSID arbitration, it would seem that (using the MHS Annulment approach) nearly any dispute involving any kind of asset would qualify for ICSID arbitration as long as the definition of investor is fulfilled.

For example, the US-Congo BIT in Patrick Mitchell v. Democratic Republic of the Congo [Patrick Mitchell]40 defined investment as “every kind of investment”.41 This inherent circularity renders the definition virtually useless. Similarly, since the common BIT definition (“every kind of asset”) usually includes claims to money having a financial value, they could arguably include every conceivable contractual claim. It is noteworthy that the tribunal in Joy Mining could only say that a bank guarantee was “different” from a claim in money (comparing it to promissory notes) and ultimately referred to Article 25 of the ICSID Convention because the BIT definition of investment was too broad to be helpful. In situations where the BIT definition is all-encompassing, it is difficult to imagine that states had, by signing the ICSID Convention, intended to open themselves up to ICSID arbitration on disputes relating to every kind of asset or every kind of investment.

2. The “party autonomy” reasoning should be applied with circumspection with respect to IIA definitions of investment, owing to the nature of an IIA

The approach taken in MHS Annulment is similar to the party autonomy principle in civil law jurisdictions with respect to contract law.

However, an IIA differs from a regular contract. An IIA is primarily an agreement between two or more contracting states. When an IIA is negotiated and signed, states have every interest in ensuring that the definition of investment is as wide as possible in order to ensure that the treaty obligation owed to the other state is to protect all investments emanating from the other states, a fortiori, when the term “investment” is defined for the purposes of a multilateral treaty. The consent to investor-state arbitration then makes present and future investors “third-party beneficiaries” to the obligation undertaken by the state and is usually contained in a single separate clause, while the term “investment” permeates all parts of the treaty.

In addition, countries increasingly are negotiating broad IIAs, such as free-trade agreements (FTAs) and economic co-operation agreements that cover a far wider scope of issues than a traditional BIT. While states intend the definition of investment to be all-encompassing with respect to defining their treaty obligations to other states, it is reasonable to presume that their primary concern in adopting the broadest possible definition of investment would be to promote freedom of trade, and it is even clearer that they may not necessarily have intended such a definition of investment to grant wholesale access to ICSID arbitration.

3. There is unlikely to be an international consensus on BIT definitions of investment

In Biwater,42 the tribunal had raised the possibility that there might be an “international consensus” to be found in the fact that substantial numbers of BITs express the definition of investment more broadly than the Salini test. The majority in MHS Annulment seemed to agree, as they had quoted the relevant passage in Biwater to that effect.43 However, as pointed out by one commentator, “the mere reproduction of similar definitions in various international treaties, be they bilateral or multilateral, is insufficient to establish a kind of ‘customary definition’ of the term”.44

One of the requirements for the establishment of customary international law is the evidence of state practice. As will be seen below, states themselves have argued in various ICSID arbitrations against the applicability of such broad BIT definitions and in support of an autonomous definition of investment under the ICSID Convention (e.g., Morocco in Salini, Grenada in RSM, Malaysia in MHS, Turkey in PSEG Global Inc., The North American Coal Corporation and Konya Ilgin Elektrik Uretin ve Ticaret Limited Sirketi v. Republic of Turkey [PSEG],45 Bangladesh in Saipem S.p.A. v. People’s Republic of Bangladesh [Saipem],46 Pakistan in Bayindir Insaat Turizm Ticaret Ve Sanayi A.S. v. Islamic Republic of Pakistan [Bayindir],47 and Egypt in Jan de Nul N.V. and Dredging International N.V. v. Arab Republic of Egypt [Jan de Nul]48). Moreover, as illustrated in section V below, BIT definitions of investment have been evolving in various ways in response to the Salini test. Accordingly, there is no consistent state practice which would support a finding of such an international consensus.

4. The definition of investment in Article 25(1) of the ICSID Convention serves a protective function

Although it is possible to see Article 25 of the ICSID Convention as a “specification of the concept of investment under the BIT” (as suggested by the Bayindir Tribunal49), it would be conceptually better to think of Article 25 as an independent requirement which is part of the state’s consent to the ICSID Convention. As stated above, the definition of investment in an IIA may be pertinent to aspects of the IIA other than the consent to ICSID arbitration. Moreover, Article 25 is more than a specification of the concept of investment of the BIT as it arguably serves a protective function.

The Report of the International Bank for Reconstruction and Development (IBRD) Executive Directors50 stated:

23. Consent of the parties is the cornerstone of the jurisdiction of the Centre …

25. While consent of the parties is an essential prerequisite for the jurisdiction of the Centre, consent alone will not suffice to bring a dispute within its jurisdiction. In keeping with the purpose of the Convention, the jurisdiction of the Centre is further limited by reference to the nature of the dispute and the parties thereto.

27. No attempt 51was made to define the term “investment” given the essential requirement of consent by the parties, and the mechanism through which Contracting States can make known in advance, if they so desire, the classes of disputes which they would or would not consider submitting to the Centre (Article 25(4)).52 [Emphasis added.]

The second paragraph of the quotation above was relied upon by both the Biwater Tribunal and the MHS Annulment majority. It links the definition of investment with the concept of consent to ICSID jurisdiction and reminds us that, for the purposes of the ICSID Convention, the definition of investment in Article 25 specifically concerns access to ICSID arbitration. The statement that the jurisdiction of the Centre is “further limited” by reference to the nature of the dispute, the express reference to the word “investment” in Article 25, and the suggestion that the mechanism in Article 25(4) could be used by states to shape the definition of investment in Article 25 all point towards the elements of Article 25 (including the word “investment”) being used as a safeguard.

Compared to the BIT definition of investment, which was usually formulated in the general context of encouraging trade relations, Article 25 is more specific to the context of consent to ICSID arbitration. It is unlikely that the ICSID contracting states intended a general definition of investment in a BIT to equate to consent to ICSID arbitration without the further consideration of the definition of investment in Article 25, which (together with the other elements of Article 25) more specifically represented the outer limit of the ICSID contracting states’ consent to arbitration under the ICSID regime.

One should note the comments of Dr Aron Broches, General Counsel of the World Bank, who chaired the consultative meetings at which the 15 October 1963 preliminary draft of the ICISD Convention was also discussed, set out below:

I believe that [giving up the effort to devise a definition of investment] was a wise decision, fully consonant with the consensual nature of the Convention, which leaves a large measure of discretion to the parties. It goes without saying, however … that this discretion is not unlimited and cannot be exercised to the point of being clearly inconsistent with the purposes of the Convention.53 [Emphasis added.]

In short, the consent of the parties, subject to consistency with the purposes of the ICSID Convention, is the predominant basis of ICSID jurisdiction.

If one accepts that the definition of investment in Article 25 is linked to consent to ICSID jurisdiction, it would become easier to explain why states can contractually agree that a specific transaction is an investment even if it is outside the scope of a BIT or does not satisfy the hallmarks (cumulatively or otherwise) which constitute the Article 25 definition, e.g., the RSM case.54 Some tribunals have taken the view that Article 25 is absolute because the ICSID Convention is a multilateral treaty, whose terms cannot be changed by bilateral agreement. This might be too rigid an approach.

The outer limit represented by Article 25(1) exists solely for the benefit of the contracting states in defining their consent to jurisdiction, and it is submitted that a state can waive such an outer limit by specifically consenting to submit a particular dispute with a known investor to ICSID arbitration or by specifically stating that a particular transaction will be considered an investment. This would justify why the Joy Mining Tribunal alluded to an exception to the requirements of Article 25 if there was an “arbitration clause”, namely an ICSID arbitration clause between the parties, as opposed to a general consent to an unknown class of investors in a BIT.

However, the authors submit that words such as “every kind of asset” are too broad to be construed as the parties’ consent to ICSID jurisdiction in a situation where the investment is inconsistent with the purposes of the Convention. There should be a stronger or more expressly worded indication sufficient to amount to a waiver of the Article 25(1) definition of investment in order for the BIT definition of consent to prevail in a case where the investment does not satisfy the definition of investment under the ICSID Convention.

Moreover, apart from the seventy-two BITs which were concluded before the ICSID Convention was drafted, most were signed after the ICSID Convention had come into existence in 1966, and states might have assumed that Article 25(1) acting as an outer limit to the definition of investment would be part of their consent to ICSID arbitration.

The MHS Annulment majority was concerned with respecting the intention of contracting states in their BITs. However, states themselves have not violently objected to the concept that investment pursuant to Article 25 of the ICSID Convention has a definition independent of the BIT definition as well as the concept that the subject matter of the dispute should satisfy some basic criteria before it can be called an investment.

It has been nine years since the decision of the tribunal in Salini 55 endorsed the concept of the “hallmarks” of an investment in 2001. In this period, various state respondents (e.g., Malaysia in MHS, Tanzania in Biwater,56 the Czech Republic in Phoenix Action,57 the Slovak Republic in CSOB,58 and Venezuela in Fedax,59) have adopted the position that investment within the meaning of Article 25(1) of the ICSID Convention can be different from the BIT definition. This is the clearest illustration that Article 25(1) was intended by ICSID contracting states to serve a protective function in their favour.

5. Article 25(4) notifications are not the appropriate tool to limit broad BIT definitions of investment

If one accepts the MHS Annulment majority’s approach, the consequence would be that the BIT definition prevails over the Salini test (in whatever version).

Article 25(4) states:

(4) Any Contracting State may, at the time of ratification, acceptance or approval of this Convention or at any time thereafter, notify the Centre of the class or classes of disputes which it would or would not consider submitting to the jurisdiction of the Centre. The Secretary-General shall forthwith transmit such notification to all Contracting States. Such notification shall not constitute the consent required by paragraph (1).60

As suggested by the CSOB and Fedax Tribunals, the most convenient way for ICSID contracting states to limit the scope of ICSID jurisdiction is to make a declaration under Article 25(4) of the ICSID Convention as the Report of the IBRD’s Executive Directors expressly refers to the Article 25(4) mechanism when explaining why investment was not defined in the ICSID Convention.61 However, as explained below, this method is not feasible.

PSEG held that notifications only help the interpretation of the parties’ consent but does not have an autonomous legal operation. The PSEG Tribunal stated that notifications are not reservations, and:

to be effective, the contents of a notification will always have to be embodied in the consent that the Contracting Party will later give in its agreements or treaties. If, as in this case, consent was given in the Treaty before the notification, that treaty could have been supplemented by means of a Protocol to include the limitations of the notification into the State’s consent. Otherwise the consent given in the Treaty stands unqualified by the notification.62

The PSEG Tribunal raises the example of the various BITs entered into by the People’s Republic of China which reproduce the terms of the notification made by the People’s Republic of China. If one adopts the PSEG approach in conjunction with the approach adopted by the MHS Annulment majority, states will not be able to use Article 25(4) as a convenient way of limiting the scope of ICSID jurisdiction.

Given that some Article 25(4) notifications are strongly worded as a clear refusal of consent to ICSID jurisdiction (e.g., Ecuador’s Article 25(4) notification) and that treaties such as the Energy Charter Treaty 63 clearly give “unconditional consent” to international arbitration or conciliation, it would be interesting to see how future tribunals will resolve conflicting expressions of consent and non-consent in the BIT and Article 25(4) notifications respectively. Since Article 25(4) notifications are not a reservation to the ICSID Convention, it is likely that a clearly worded BIT would prevail even over a strongly worded refusal to consent contained in an Article 25(4) notification.

Moreover, only seven out of 143 states have made Article 25(4) notifications in practice.64

Article 25(4) is thus not likely to be a useful, or popular, device through which states can refine the definition of investment. If the approach of the MHS Annulment majority is correct, states which wish to further refine the definition of investment must do so by amending their BITs. However, it is arguable that amending BITs to qualify the broad definition of investment will create its own problems.

Even if a state were to renegotiate the definition in some of its BITs, it is at least arguable that most-favoured nation (MFN) clauses may cause any broader definition of investment in another BIT to apply to the renegotiated BIT,65 unless the renegotiating state has the foresight and bargaining power to expressly preclude the application of the MFN clause in this respect.

Any class or classes of disputes submitted to the jurisdiction of the Centre may still be overinclusive if not carefully drafted, in the sense that certain investments may superficially fall within such classes (whether by accident or design) but in substance would still not be an investment envisaged in ICSID practice and jurisprudence. It is entirely foreseeable that states would still need to rely on tribunals to filter out claimants such as Phoenix Action and “ordinary commercial claims” which states enter into as part of normal commercial life.

6. ICSID jurisprudence would benefit from a common definition of investment under Article 25(1) of the ICSID Convention

The confusion over the definition of investment has not gone unnoticed by states. In the Report of the Multi-Year Expert Meeting on Investment for Development,66 attended by representatives of eighty-one state members of the United Nations Conference on Trade and Development (UNCTAD), it is noted that:

21. A third trend related to the divergent interpretations of treaty obligations made by international tribunals. While some suggested that these divergences of interpretation were sometimes more related to differences in the assessment of facts, and less to differences in interpretation, a great number of examples of divergent interpretations were discussed by experts. These included … the scope of covered investments with some tribunals considering the “Salini” criteria as specific requirements and others considering them to be merely possible aspects for determining whether an investment was covered by the IIA or not …

22. The potential for divergent interpretations was seen as a source of great concern and lack of predictability. As a response to these developments, some IIAs included specific interpretations of key provisions, with a view to fostering a more consistent and rigorous application of international law in arbitral awards and in order to prevent divergent interpretations. Some speakers noted the absence of general principles of law in the area of investment.67 [Emphasis added.]

The IIA system is highly atomized, with different tests under different BITs. It is submitted that the formulation of a clear and uniform test would help in reducing the number of divergent interpretations and aid in articulating some general principles in relation to the concept of investment. The idea is to have a common framework while still allowing for “customization” of that framework by states through express provision in IIAs and specific investor-state agreements. Accordingly, the concept of Article 25(1) of the ICSID Convention operating as a common standard for outer limits to the definition of investment still remains attractive in terms of promoting consistency and predictability of access to ICSID arbitration, especially because BIT definitions tend to be overly broad.

B. Other Concerns of the MHS Annulment Majority with the Outer Limit Approach

The other policy-based concerns of the MHS Annulment majority with the outer limit approach will be dealt with in this section. One concern is that it would deprive investors of their only arbitral recourse contrary to the intent of the state parties to the BIT. As stated by the MHS Annulment Committee:

It cannot be accepted that the Governments of Malaysia and the United Kingdom concluded a treaty providing for arbitration of disputes arising under it in respect of investments so comprehensively described, with the intention that the only arbitral recourse 68provided between a contracting state and a national of another Contracting State, that of ICSID, could be rendered nugatory by a restrictive definition of a deliberately undefined term of the ICSID Convention, namely “investment” as it is found in the provision of Article 25(1). It follows that the Award of the Sole Arbitrator is incompatible with the intention and specifications of the States immediately concerned, Malaysia and the United Kingdom.69 [Emphasis added.]

The outer limit approach does not equate to a “restrictive definition of a deliberately undefined term of the ICSID Convention”.70

First, the fact that the drafters of the ICSID Convention left the term “investment” undefined is a familiar refrain in many awards which take a similar position to that taken by the MHS Annulment Committee. However, the reason that the definition of investment was deliberately left undefined was simply because states could not agree on an acceptable definition and not because a definition under Article 25(1) was unnecessary. As Schreuer noted: “The subsequent discussions [of the Preliminary Draft of the ICSID Convention] showed a widely held opinion that a definition of the term ‘investment’ was necessary”.71

Second, an outer limit approach need not be restrictive. Indeed, most tribunals which support the outer limit approach acknowledge that an Article 25 definition of investment must be flexible.

Third, even though investment was not defined, “it was always clear that ordinary commercial transactions would not be covered by the Centre’s jurisdiction no matter how far-reaching the parties’ consent might be”.72 In adopting the outer limit approach, tribunals are not seeking to adopt a more restrictive definition of investment, but rather to give effect to the substance of the contracting states’ consent to the ICSID Convention.

Moreover, it can be argued that the very act of choosing ICSID arbitration illustrates the intent of state parties to a BIT to overlay the requirements of the ICSID Convention (including the interpretation of Article 25 in relation to the scope of their consent to ICSID arbitration) over the broad BIT definition of investment. The state parties to a BIT could easily have provided for non-ICSID arbitration (e.g., United Nations Commission on International Trade Law (UNCITRAL), International Court of Arbitration (ICC), or London Court of International Arbitration (LCIA) arbitration), in which case the tribunal would be free to apply the IIA definition of investment.73 Alternatively, if the investor possessed sufficient bargaining power, it could obtain the state’s consent to arbitration under other rules.

Another concern of the MHS Annulment majority is that the ICSID institution would be “crippled” by “questionable interpretations” of the term “investment” as found in Article 25(1) of the ICSID Convention.74 Alternatively, investors (or even states) might seek another forum which would not require parties to jump over the ICSID hurdle of the Article 25 definition of investment. However, these concerns seem to be exaggerations. Despite the brewing debate, ICSID cases have increased rather than decreased in the recent years.

Accordingly, the above two peripheral, policy-based concerns of the MHS Annulment majority do not detract from the desirability of having an Article 25 definition of investment serve as a unifying minimum standard for ICSID jurisdiction.

III. An interesting alternative to the Salini-type test: to construe the bit definition of investment as having the same characteristics as the Salini test

Interestingly, the recent Permanent Court of Arbitration (PCA) case of Romak S.A v. Republic of Uzbekistan [Romak]75 (which was under the UNCITRAL Rules and not under the ICSID regime) has advanced a new approach. Romak was a case where the claimant tried to claim that a contract for the supply of wheat was an investment under the Switzerland-Uzbekistan BIT. The tribunal in Romak agreed that a literal interpretation of the BIT definition of investment in that case (i.e. every kind of asset) would render meaningless the distinction between investments and purely commercial transactions. It held that a mechanical application of the usual listed categories in such a definition (e.g., claims to money, etc.) would produce a result which is “manifestly absurd or unreasonable”. Such an outcome was contrary to Article 32(b) of the Vienna Convention on the Law of Treaties [Vienna Convention].76 If the intention of the contracting states was to accord to the term “investment” an extraordinary and counter-intuitive meaning, the wording must leave no room for doubt that this was the intention.

Although the Romak Tribunal opined that there was no basis to suppose that the word “investment” had a different meaning in the context of the ICSID Convention than it bore in relation to the Switzerland-Uzbekistan BIT, it examined the ICSID jurisprudence and concluded that the inherent meaning of investment under the BIT entailed a contribution that extends over a certain period of time and that involves risk. In other words, the tribunal construed the Salini test developed in ICSID jurisprudence into the BIT definition of investment.

The approach of the Romak Tribunal is similar to the one proposed by Zachary Douglas in his book The International Law of Investment Claims, in which he took the view that:

the open-textured nature of the standard formulation in investment treaties [defining “investment” to be “any asset” and then providing a non-exhaustive list of assets that might qualify as an investment] preserves the ordinary meaning of the term “investment” and therefore its consistency with the characteristics that must be attributed to the same term as employed in Article 25 of the ICSID Convention.77

As the saying goes, “there are many ways to skin a cat”. It is evident that non-ICSID Tribunals face the same problem of trying to define the difference between an investment and a commercial transaction for the supply of goods and services. ICSID Tribunals have given the word “investment” in Article 25 an inherent meaning, while a non-ICSID tribunal in Romak gave the already defined term “investment” in the BIT a further inherent meaning that coincided largely with the inherent meaning ICSID Tribunals had previously given. If non-ICSID Tribunals are also of the view that investment in the BIT is the same as that in Article 25, this further reinforces the practical need for an inherent meaning of investment.

IV. Discussion of the hallmarks of investment

The discussion in this section takes place on the premise that Article 25(1) of the ICSID Convention acts as an outer limit to the BIT definition of investment and will focus on which hallmarks should constitute the defining characteristics of “investment” under Article 25(1).

The key weakness of the Salini test and the jurisdiction approach was that such a definition might be too narrow and inflexible, resulting in the arbitrary exclusion of the disputes relating to certain transactions from the jurisdiction of the Centre. This was one of the reasons why the tribunal in the Biwater case did not support a jurisdiction approach.78

Emmanuel Gaillard was a member of the tribunal in the recent case of Saba Fakes, where the tribunal affirmed the three criteria of: “(i) a contribution, (ii) a certain duration, and (iii) an element of risk, are both necessary and sufficient to define an investment within the framework of the ICSID Convention”.79 However, the tribunal in Saba Fakes rejected a contribution to the host state’s economic development, good faith, and legality of the investments as hallmarks.

However the hallmarks may be described, the authors of this article agree with a need for a certain and yet non-restrictive test. Even tribunals in support of a jurisdiction approach have recognized that the criteria cannot operate too restrictively, as they have been careful to say that the criteria should be evaluated as a whole and not individually. As observed above, and in the MHS Award,80 the distinction between the jurisdiction or the typical characteristics approach may be academic. However, with either approach, the larger problem is what the hallmarks of investment should be.

At present, the existing hallmarks are at risk of being over- or underinclusive, and the search for a “perfect” definition of investment continues. It is suggested that the criteria of investment should be:

1. 1. a financial commitment (in money or other terms);81

2. 2. risk;

3. 3. duration; and

4. 4. significant contribution to the host state’s development.

In addition, the “good faith” or bona fides of an investment can be considered as a separate principle or subsumed under the fourth hallmark of “significant contribution”.

“Regularity of profit and return” is now understandably not favoured as a hallmark. Such a hallmark will not be applicable where, as in Biwater, the project was a loss leader. Some projects may be contingent on extraneous events (e.g., the successful discovery of natural resources) and may not be structured in such a way that there would always be an expectation of a regular profit and return. An article by Luke Eric Peterson even raises the possibility of not-for-profit organizations bringing claims in ICSID arbitration.82

Douglas proposes two new rules for the definition of investment, accompanied by a comprehensive analysis of the authorities:

Rule 22: The legal materialization of an investment is the acquisition of a bundle of rights in property that has the characteristics of one or more categories of an investment defined by the applicable investment treaty where such property is situated in the territory of the host state or is recognized by the rules of the host state’s private international law to be situated in the host state or is created by the municipal law of the host state.83

Rule 23: The economic materialization of an investment requires the commitment of resources to the economy of the host state by the claimant entailing the assumption of risk in expectation of a commercial return.84

Douglas’s Rule 23 retains only three of the Salini hallmarks, namely commitment, risk, and expectation of a commercial return. It does not include duration or economic development. In conclusion, Douglas emphasizes that whatever may be the definition of investment, it must be certain so that investors will know at the time of investment whether they are making an investment to which the ICSID Convention applies. One cannot have an inchoate list of criteria.

On the other hand, Devashish Krishnan proposes that the definition of investment under the ICSID Convention should be consistent with what is treated as an investment under a country’s capital account, as well as the IMF’s description of investment as direct, portfolio, and other investment.85

Currently, the development of ICSID case-law appears to favour the adoption of the three hallmarks of contribution, risk, and duration. Gaillard86 supports the outer limit approach, but suggests that only three criteria (contribution, risk, and duration) should be used as the hallmarks of investment (without a separate criterion of contribution to the economic development of the host state) as a solution which would keep the definition flexible and non-restrictive. This position was largely consistent with Consorzio Groupement L.E.S.I.-DIPENTA v. People’s Democratic Republic of Algeria [Consorzio]87 and Saba Fakes,88 where Gaillard was part of a three-member tribunal in both cases. The Consorzio Tribunal elaborated upon the three factors of contribution, duration, and risk as follows:

1. 1. contribution must be made at least in part in the host country and bring with it economic value;89

2. 2. with respect to duration, there must be economic commitments of significant value sufficient for one to agree that the operation is of a nature to promote the economy and development of the country concerned;90 and

3. 3. with respect to risk, any contract that implies risk for the contracting party such that there should be a particular guarantee of jurisdiction to firms seeking to invest in another country allowing for intervention by international arbitrators, in addition to ordinary mechanisms.91

On the face of it, Consorzio is in support of only the first criteria of contribution. However, it can be seen that the “economic development” requirement seems elided into the “duration” requirement in Consorzio.

If one takes the Consorzio definition of duration as authoritative, the debate over whether economic development constitutes a hallmark may well be academic. However, the duration requirement is usually understood as the length of the investment in terms of time.

A. Contribution/Significant Contribution to Economic Development

This hallmark (which was identified as a possible additional requirement in Salini) is derived from the Preamble of the ICSID Convention, which states: “the Contracting States [are] considering the need for international cooperation for economic development, and the role of private international investment therein”.92

The description of this hallmark varies from a significance for the host state’s development,93 “an international transaction which contributes to cooperation designed to promote the economic cooperation of a Contracting State may be deemed an investment [under the ICSID Convention]”,94 “contribution to the economic development of the host State as an essential … characteristic of the investment”,95 “a significant contribution to the host State’s economy”, “a contribution to the economic and social development of the host state”96 to “an operation made in order to develop an economic activity in the host state”.97

The requirement of contribution/significant contribution to the economic development of the host state (while not perfect) is a way of capturing the amorphous distinction between an investment in the ICSID sense and an ordinary commercial transaction. Moreover, it is exactly the idea behind the ICSID Convention, which is to afford investors an avenue to arbitration with the state in order to encourage them to invest in activities which would benefit the state.

Some commentators have questioned whether the significant contribution should be to the economic, as opposed to political, social, or cultural development of the state. Schreuer comments:

[I]t does not follow that an activity that does not obviously contribute to economic development must be excluded from the Convention’s protection. Any concept of economic development, if it were to serve as a yardstick for the existence of an investment and hence for protection under ICSID, should be treated with some flexibility. It should not be restricted to measurable contributions to GDP but should include development of human potential, political and social development and the protection of the local and the global environment.98

Others have questioned whether economic development should be a hallmark at all. For example, Gaillard99 takes the position that the Preamble was a mere acknowledgment that investment fosters economic development but did not mean that economic development is essential to the notion of investment. The recent case of Saba Fakes 100 takes a similar position. Douglas takes the view that the economic development criterion is unworkable owing to its subjective nature, because “whether or not a commitment of capital or resources ultimately proves to have contributed to the economic development of the host state can often be a matter of appreciation and generate a wide spectrum of reasonable opinion”.101 Krishnan is of the view that an economic transaction constituting an investment by definition contributes to development, that no economist would consider private foreign investment to be anti-development, and, in any event, it is not for ICSID Tribunals and arbitrators to pronounce on what type of investment have deleterious or anti-development effects.102

The concerns of the various commentators are not without merit. However, it might be possible to further refine the concept of “economic development” to reduce subjectivity and uncertainty instead of eliminating this hallmark from the tribunals’ consideration altogether. The reason why economic development may reasonably be regarded as a necessary criterion is as follows.

Although the Preamble of the ICSID Convention is not binding, it does reflect a major (if not primary) basis upon which contracting states entered into the ICSID Convention. The encouragement of non-economic development (such as arts and culture) was not one of the stated purposes of the ICSID Convention. In the modern era, many “non-economic” activities, e.g., cultural and historical preservation, have also been credited with an economic value (e.g., Egypt’s economy relies heavily on its tourist industry). It is also logical to use economic development as a marker, as the investor would invariably be compensated for his improvement to the state with money or something of monetary value; hence the tribunal will ultimately have to attach an economic value to any commitment made by the investor. That economic value would indirectly represent, in economic terms, the value that the investor has brought to the state.103

Moreover, non-economic development to which no economic value can be credited would be difficult to measure. One needs only to imagine the type of evidence and submissions needed in order to prove a contribution to the political, social, or cultural development of the state in order to realize that any tribunal would be hard put to make any firm assessment of such a contribution in practice.

The exercise undertaken by tribunals is not to pronounce on whether investments have deleterious or anti-development effects, but rather to distinguish between ordinary commercial transactions (e.g., buying a metro ticket may in its small way “contribute” to the economy) and the type of investments to which parties to the ICSID Convention envisioned being submitted to ICSID arbitration.

If one were to completely exclude the criteria of economic development, the danger is that the three criteria of commitment (in money or other terms), risk, and duration (if understood only in terms of number of years) can easily be superficially satisfied. For example, many commercial loans can fulfil the three criteria but probably only certain loans will qualify as an investment under Article 25 of the ICSID Convention. In the CSOB case, the tribunal, in determining that a particular loan was an investment, stated:

[T]he Tribunal considers that the broad meaning which must be given to the notion of an investment under Article 25(1) of the Convention is opposed to the conclusion that a transaction is not an investment merely because, as a matter of law, it is a loan. This is so, if only because under certain circumstances a loan may contribute substantially to a State’s economic development …104 [Emphasis added.]

The fourth hallmark of significant contribution to host state development focuses a tribunal’s attention on the object of the ICSID Convention in such cases.

At least some states have reaffirmed that economic development is a major purpose of concluding IIAs. The role of IIAs on development was discussed at length in the Report of the Multi-Year Expert Meeting on Investment for Development, which states:

29. Experts also discussed the impact of IIAs on FDI [foreign direct investment] flows. Views concurred regarding the difficulty of establishing tangible proofs that the conclusion of IIAs would increase FDI inflows to developing countries … participants also stressed the importance of IIAs for providing a stable and predictable investment framework … However, for some countries that had concluded IIAs in the hope of attracting FDI, the particular experience was that IIAs turned into a source of litigation, instead. Some countries felt that such agreements had not achieved their initial aim of increasing FDI flows.

30. In that context, there was a debate about the objectives of IIAs and the extent to which they should reflect development aspects. In discussing IIA objectives, experts distinguished between the protection of foreign investors; the enhancement of FDI flows and the furtherance of economic development. With countries reporting on their particular experience and expectations, different nuances about the importance of the different objectives emerged.

31. Although some participants noted that IIAs, at their core, were intended to support economic development, others believe that they needed to do more to reflect development objectives and incorporate investment promotion. Suggestions included a closed definition for investment that provided certainty and clarity …105 [Emphasis added.]

Sebastien Manciaux, in his article “The Notion of Investment: New Controversies”,106 described the criteria of “significant contribution” to economic development as a quantitative threshold that is unspecified. He also points out that mergers and acquisitions do not result in a significant contribution to economic development but yet account for more than half the annual flow of foreign direct investment. Another example raised was that of a failed construction project. He argues that this hallmark should be discounted or ignored because it does not constitute a discriminating criterion from a legal perspective.

Manciaux’s example of the failed construction project was considered in the case of RSM, and the tribunal was not deterred from finding that there had been a contribution to the economic development of the host state, notwithstanding the lack of actual oil exploration activities. In RSM, the tribunal stated:

243. [T]here would be no need for actual expenses to have been incurred … the relevant criterion being the commitment to bring in resources toward the performance of such exploration … Had the Exploration Licence been issued, RSM would have been irrevocably committed to bring in, directly or indirectly by turning to other sources, the necessary capital. If oil was not found, or was not found in sufficient quantities, or was found to lie in locations that did not make exploitation economically viable, that capital would have been spent in vain.

244. [A]s to the contribution to the economic and social development of the host State, in the unlikely situation where the exploration expenses themselves would not be sufficient to satisfy it, the condition must be assessed in consideration of a successful adventure. It is not the actual or the final contribution that matters, precisely because the exploration may not lead to exploitation.107 [Emphasis added.]

The hallmark of significant contribution is not meant to be a “hard” or calculable measurement. As a legal norm, it does not necessarily have to translate into an appreciable increase in gross domestic product or any other economic indicator. It is simply no more than what common law countries would deem in domestic law as purposive interpretation of legislation. The question is: What sort of investment did contracting states intend to protect by signing the ICSID Convention?

The contracting states clearly did not intend ordinary commercial transactions to be protected, and the “economic development” hallmark is a useful tool for a tribunal to distinguish between an ordinary commercial transaction and an investment contemplated by the ICSID contracting states. A failed project which never took off and never contributed in the hard mathematical sense to the growth of the economy could still satisfy this hallmark if the investor had significantly invested in the project to such a degree that it could have made a significant contribution to economic development had the project been brought to fruition. As for mergers and acquisitions made with genuine business goals, a reasonable tribunal could also consider that a particular acquisition did satisfy this hallmark. Conversely, one could argue that a large enough “ordinary commercial transaction” (e.g., a hypothetically very large purchase—or very expensive—order for a particular good) might cause gross domestic product to increase and thereby contribute to the economy of a state, but as mentioned above, the economic development hallmark is not to be mechanistically applied. Ultimately, each case is fact-specific, and the tribunal would have to assess the investment carefully to determine whether this hallmark is satisfied.

Although the hallmark of economic development might be considered a “softer” standard than the other hallmarks, the law does have a place for such standards, e.g., “good faith” or “unconscionability”. In any event, the tribunal can be guided by previous case-law and developments in investor-state practice in determining what investments should satisfy this hallmark. This should address the concerns of uncertainty and subjectivity raised by some tribunals and commentators.

Ultimately, the Article 25 definition is a definition protective of the scope of the state’s consent to jurisdiction that fills the gap, especially in circumstances when the BIT definition of investment is overly broad. A criterion of “significant contribution to economic development” would help to sieve out vexatious and de minimis claims, reducing the financial cost to states of defending such claims. If the circumstances show that the state has already contemplated what would constitute an investment in detail, for example in the newer IIAs discussed below, then it is acknowledged that the tribunal should consider the wording of the BIT as a relevant reflection of what the state considers to be a fulfilment of the Article 25 definition. Accordingly, the four hallmarks as proposed in section I.B should suffice as the key components of an Article 25 definition.

V. Developments in treaty drafting

States are also exploring new ways of defining investment. The draft Free Trade Area of the Americas [FTAA]108 itself contains nine proposed definitions of investment. This shows that the struggle to define investment exists not only in ICSID jurisprudence, but also continues in the arena of treaty drafting.

In response to the debate on definition, the newer IIAs and treaties have adopted negative definitions of investment. For example, the multiple definitions proposed in the third draft agreement of the FTAA 109 contain many suggestions as to what the word “investment” does not mean. The common theme to these proposals is that investment should exclude commercial contracts for goods and services and the extension of credit in commercial contexts. Among the many versions of negative definitions are:

[I]nvestment does not mean:

i) a debt instrument of the State;

j) claims to money that arise solely from:

i) commercial contracts for the sale of goods or services by a national or enterprise in the territory of a Party to an enterprise in the territory of another Party; or

ii) the extension of credit in connection with a commercial transaction, such as trade financing, other than a loan covered by subparagraph d) supra; or

k) any other claims to money, that do not involve the kinds of interests set out in subparagraphs a) through h) supra;

…

[Investment] does not include:

a) a payment obligation of the State or a State enterprise and the granting of such credit to the State or a State enterprise; nor

b) claims to money derived exclusively from:

i) commercial contracts for the sale of goods and services by a national or enterprise in the territory of a Party to a national or enterprise in the territory of another Party; or

ii) the granting of credit in relation to a commercial transaction, whose period of maturity is less than three (3) years, such as financing of trade;

…

This definition [of investment] does not include:

a) assets not directly linked to a productive activity; and

b) loans and other operations resulting in debt, as well as flows of capital related strictly to a commercial transaction;

…

But investment does not mean:

f) merely financial flows, such as those destined only to gain indirect access to the financial market of the other Party;

g) claims to money that arise solely from:

i) commercial contracts for the sale of goods or services by a national or enterprise in the territory of a Party to an enterprise in the territory of another Party; or

ii) the extension of credit in connection with a commercial transaction, such as trade financing, other than a loan covered by subparagraph d) supra; or

h) any other claims to money, that do not involve the kinds of interests set out in subparagraphs a) through e) supra;110

The 2004 US Model BIT111 also describes certain exclusions to the definition of investment:

¹ Some forms of debt, such as bonds, debentures, and long-term notes, are more likely to have the characteristics of an investment, while other forms of debt, such as claims to payment that are immediately due and result from the sale of goods or services, are less likely to have such characteristics.

² Whether a particular type of license, authorization, permit, or similar instrument (including a concession, to the extent that it has the nature of such an instrument) has the characteristics of an investment depends on such factors as the nature and extent of the rights that the holder has under the law of the Party. Among the licenses, authorizations, permits, and similar instruments that do not have the characteristics of an investment are those that do not create any rights protected under domestic law. For greater certainty, the foregoing is without prejudice to whether any asset associated with the license, authorization, permit, or similar instrument has the characteristics of an investment.

³ The term “investment” does not include an order or judgment entered in a judicial or administrative action.112

The Chile-Korea FTA113 combines one of the proposed definitions in the FTAA with an exclusion from the 2004 US Model BIT:

• [but] investment does not mean,

• (i) claims to money that arise solely from:

  1. (i) commercial contracts for the sale of goods or services by a national or enterprise in the territory of a Party to an enterprise in the territory of the other Party; or

  2. (ii) the extension of credit in connection with a commercial transaction, such as trade financing; and

• (j) an order entered in a judicial or administrative action.114

Some states have shifted back to traditional concepts of foreign direct investment. The EFTA-Mexico FTA115 restricts investment to “direct investment … for the purpose of establishing lasting economic relations with an undertaking such as, in particular, investments which give the possibility of exercising an effective influence in the management thereof”.116

Other states have resorted to using a “closed-list” definition of an investment. For example, Article 96 of the Japan-Mexico FTA117 adopts such an approach, stating:

(i) the term “investment” means:

(AA) an enterprise;

(BB) an equity security of an enterprise;

(CC) a debt security of an enterprise;

(aa) where the enterprise is an affiliate of the investor; or

(bb) where the original maturity of the debt security is at least 3 years,

but does not include a debt security, regardless of original maturity, of a Party or a state enterprise;

(DD) a loan to an enterprise:

(aa) where the enterprise is an affiliate of the investor; or

(bb) where the original maturity of the loan is at least 3 years,

but does not include a loan, regardless of original maturity, of a Party or a state enterprise;

(EE) an interest in an enterprise that entitles the owners to share in income or profits of the enterprise;

(FF) an interest in an enterprise that entitles the owner to share in the assets of that enterprise on dissolution …;

(GG) real estate or other property … and any related property rights such as leases, liens and pledges, acquired in the expectation or used for the purpose of economic benefit or other business purposes; and

(HH) interests arising from the commitment of capital or other resources in the Area of a Party to economic activity in such Area, such as under:

(aa) contracts involving the presence of an investor’s property in the Area of the Party, including turnkey or construction contracts, or concessions, or

(bb) contracts where remuneration depends substantially on the production, revenues or profits of an enterprise;

but investment does not mean,

(II) claims to money that arise solely from:

(aa) commercial contracts for the sale of goods or services by a national or enterprise in the Area of a Party to an enterprise in the Area of the other Party; or

(bb) the extension of credit in connection with a commercial transaction, such as trade financing, other than a loan covered by subparagraph (DD) above; or

(JJ) any other claims to money that do not involve the kinds of interest set out in subparagraphs (AA) through (HH) above;118 [Emphasis added.].

In the face of such a specific “definition” of investment that already indicates the duration (three years) of certain types of investment, it is foreseeable that the tribunal will not be able to contradict the BIT when assessing whether the duration characteristic under Article 25 has been fulfilled for a loan or debt security to an enterprise. The closed list of investments also indicates to some extent what sort of risk or commitment is acceptable as an investment and the tribunal should also take that into account in considering whether the Article 25 definition has been fulfilled. However, there is still room for the fourth hallmark (if this is accepted as a hallmark), i.e. significant contribution to economic development, to act as a sieve for frivolous, insignificant, or ordinary commercial claims, although it is arguable that an investment within the closed-list would prima facie be a contribution to economic development. States may have to prove that the investment is as egregiously fictional as the one in Phoenix Action 119 in order for the tribunal to strike out a claim for the failure to satisfy the Article 25 definition.

Accordingly, an Article 25 definition will still have a useful, though very limited, part to play where newer, more comprehensive closed-list BITs are involved. However, it is acknowledged that its greatest role is as a protective definition where the BIT definition is “every kind of asset” or “every kind of investment”.

In fact, one reason why the MHS Annulment majority decision will not end the debate over the scope of the Salini test is because

numerous recently negotiated IIAs incorporate a definition of “investment” in economic terms—that is, they cover, in principle, every asset that an investor owns and controls but add the qualification that such assets must have “the characteristics of an investment”. For this purpose, they refer to the criteria developed in ICSID practice, such as “the commitment of capital or other resources, the expectation of gain or profit or the assumption of risk”.120

For example, Article 10.1 of the Chile-Korea FTA states:

“Investment” means every kind of asset that an investor owns or controls, directly or indirectly, and that has the characteristics of an investment, such as the commitment of capital or other resources, the expectations of gains or profits and the assumption of risk …121 [Emphasis added.]

(Note that the US-Chile FTA122 also contains a similar reference to the “characteristics of investment”.)

The 2004 US Model BIT contains a similar but slightly longer definition:

“Investment” means every asset owned or controlled, directly or indirectly by an investor that has the characteristics of an investment. Where an asset lacks the characteristic of an investment, that asset is not an investment regardless of the form it may take. The characteristics of an investment include the commitment of capital, the expectation of gain or profit or the assumption of risk. Forms that an investment may take include [an enterprise, shares, futures, options, etc.]123 [Emphasis added.]

Most recently, the ASEAN Comprehensive Investment Agreement 124 states:

²Where an asset lacks the characteristics of an investment, that asset is not an investment regardless of the form it may take. The characteristics of investment include the commitment of capital, the expectation of gain or profit or the assumption of risk.

³For greater certainty, investment does not mean claims to money that arise solely from:

(a) commercial contracts for sale of goods and services; or

(b) the extension of credit in connection with such commercial contracts.125 [Emphasis added.]

The definitions in the above BITs are interesting as they adopt hallmarks identified in the Salini test but do not mention the hallmarks of duration or significant contribution to economic development. Arguably, the use of the words, “such as” and “include” indicates that the three listed hallmarks are merely examples of a characteristics of an investment and case-law can be referred to supply other characteristics. It would be interesting to see if a loss-leader project would be considered an investment under this treaty, given the express mention of “expectations of gains of profits” as a characteristic of investment.

The 2004 US Model BIT and the ASEAN Comprehensive Investment Agreement is perhaps a good reflection of what the typical modern state intends when adopting an “every kind of asset” definition of investment. In doing so, the state is concerned that the form which the investment may take should not be restricted in any way and accordingly provides that an investment means every kind of asset. However, the state, cognizant of the fact that every kind of asset may be an overinclusive definition, may still wish the asset, whichever form it takes, to meet the characteristics of investment. The state is happy to leave the assessment of whether there are characteristics of investment in particular cases to the individual tribunals, but provides some guidance in the BIT with respect to what may be considered a characteristic of an investment, without precluding the tribunal from finding that there are other characteristics that have evolved, perhaps from ICSID case-law or other investment treaty jurisprudence.

A question that may loom ahead for future tribunals would be whether a BIT which has been recently entered into and does not contain refinements of the definition of investment described in this section, but instead still defines investment as “every kind of asset”, would be construed as an intention to rely on Article 25(1) and the existing ICSID case-law on Article 25(1) to supply the outer limits of the definition of investment, or would now be construed as a conscious adoption of a broad definition intended to remove any barriers to ICSID arbitration which Article 25(1) may pose in relation to the definition of investment.

Some tribunals or commentators would adopt the view that it would not be possible for a bilateral or unilateral act of the parties to widen the scope of Article 25(1) as the ICSID Convention is a multilateral treaty.126 However, as discussed in section II(A)(4) above, one might also take the view that the protections of Article 25(1) are for the benefit of the individual host state, which the host state may expressly waive in a particular circumstance(s).

Even if the second approach is adopted, it might still be that an unequivocal waiver would be required in order for the tribunal to disapply the usual requirements of Article 25(1) and assume jurisdiction on the basis of state consent to the specific subject matter of the arbitration.

Accordingly, even if future IIAs adopt the general wording “every kind of asset”, they should not be taken as trying to expand the Article 25 definition.

The trend towards more comprehensive definitions in newer BITs show that at least some states have appreciated the failings of an overbroad BIT definition and are helping to make the definition of investment more precise. This is a laudable process as both states and ICSID Tribunals can mutually gain valuable feedback from ICSID awards and trends in treaty-drafting respectively.

VI. Conclusion

The definition of investment under Article 25(1) should consist of the hallmarks of commitment, duration, risk, and significant contribution to economic development, with “good faith” being subsumed under the fourth hallmark, or considered separately as a general principle applicable to the interpretation of treaties. These four hallmarks can operate as a useful and general outer limit to the BIT definition of investment, especially when the BIT definition is a broad one such as “every kind of asset [or investment]”. Whether the four hallmarks are treated as jurisdictional (assessed cumulatively) or merely as typical characteristics may be an academic distinction.

The important point is that that the tribunal must consider the degree to which the hallmarks have been fulfilled and, if any of the hallmarks are not satisfied or only superficially satisfied, the tribunal must balance the fulfilment of the other satisfied hallmarks against any hallmarks that are not satisfied in its determination as to whether it has jurisdiction.

If none of the hallmarks are satisfied or if all of the hallmarks are satisfied, the residual consideration of the tribunal would be whether:

1. 1. Despite the satisfaction of all the hallmarks, the investment is nevertheless contrary to the purpose of the ICSID Convention and the consent of the contracting state to ICSID jurisdiction; or

2. 2. Despite the non-satisfaction of all the hallmarks, the investment is still consistent with the purpose of the ICSID Convention and the consent of the contracting state to ICSID jurisdiction.

Postscript

Let the principal author close with a personal note. When I wrote the MHS Award,127 I did not have the benefit of reading those awards which were decided after MHS, viz M.C.I.,128Biwater, RSM, and Phoenix Action, nor did I have the benefit of the Claimant’s arguments before the MHS Annulment Committee (which were presented by a different legal team from the one that appeared before me). Were I to write the Award today, I would certainly write it in a different way, but it would be wrong for me (or anyone else) to speculate whether my ultimate decision would be different. The point of this article is not to defend my Award, but to point out that, in the light of the conflicting jurisprudence, the question of what is an investment still remains an open one in the future—even for me.