The Scope and Effect of Umbrella Clauses:

The Need for a Theory of Deference?

 

 

Mihir C. Naniwadekar*

 

 

 

This note offers support for the view that all investment-related contracts are within the scope of bilateral investment treaty (BIT) protection under a typical umbrella clause; while ordinary commercial contracts are not. In this category of investment-related contracts, the author argues that the protection is wide and is not, at the jurisdictional threshold, subject to contractual dispute-settlement provisions outside the BIT. This view is based on distinguishing between a treaty breach and a contractual breach – though the latter may result in the former, the two are conceptually distinct and give rise to separate causes of action. Consequently, BIT tribunals have jurisdictions over treaty breaches and are not affected in this regard by contractual adjudicatory mechanisms. At the same time, this note recognizes the practical and efficacy-based difficulties of allowing simultaneous proceedings in both treaty and contract based forums; and possible abuses of turning to treaty-based adjudication simply to get around failure in contract-based adjudication. These difficulties – this note suggests – will best be solved not by treating the matter as one of jurisdiction of treaty-based tribunals, but rather by arriving at a coherent theory of deference to be granted by one tribunal to the other as a matter of merits.

 

 

 

 

 

 

 

 

Table of Contents

 

I.    Introduction

II.   The Arbitral Case Law: Is a Common Thread Visible?

      A.   The Pakistan Case

      B.   The Phillipines Case

      C.   Can a Common Thread be Seen?

      D.   Subsequent Arbitral Awards

III. The Scope of the Umbrella Clause

      A.   The Basic Principles of Interpretation

      B.   A Short Look at the History of the Umbrella Clause

      C.   The Relevance of the History

      D.   Interpreting the Clause

      E.   A Conflict in Methodologies

      F.   The Investment Test to Limiting the Scope of the Umbrella Clause – Reconciling Object, Context and Text

IV.  Umbrella Clauses and Contractual Forum Selection Clauses

      A.   Violations of BITs as well as Contracts

      B.   Countering the Counters

      C.   A Theory of Deference on Merits

V.   Conclusion

 

I.    Introduction

 

      Several layers of protection can exist under law in order to ensure that an investment in a foreign state is safeguarded.[1] At the first level, as a matter of customary international law, host states are prohibited from illegally expropriating the investments of foreign nationals. Host states are under an obligation to treat a foreign investment in accordance with an international minimum standard.[2] However, protections under customary international law are not enforceable directly by the affected party, and there is no clear forum for resolution of disputes.[3] These factors perhaps influenced the development of the second level of protection – the Bilateral Investment Treaty (BIT).[4] Under the typical BIT, most of the customary protections are codified and further strengthened, and a dispute-resolution mechanism (often in the shape of an arbitration clause referring potential disputes to an international institution) is included.^[5] This allows the investor to directly make claims against the host state for violation of the BIT.^[6] A further level of protection may be provided in specific contracts entered into between the investor and the host state. These contracts typically contain a more detailed record of obligations than the average BIT; and they may well have their own dispute settlement mechanism.

 

      So far, all seems clear. Those obligations arising out of the contract would be subject to the contractual dispute resolution mechanism; and those disputes arising out of the BIT would be governed by the mechanism provided in the BIT. However, in reality, the position is rather more complex. This is because of the umbrella clause – a rather widespread feature in todays BITs.[7] An umbrella clause essentially requires each party to the BIT to observe any obligations it may have entered into with regard to investments of nationals or companies of the other party. The umbrella clause, at least arguably, blurs the line between contract and treaty.^[8] A contract is also an obligation on the host state – does the umbrella clause then mean that breach of the contract is to be construed as a breach of the BIT as well? The answer to this question is of significance, as it will determine several issues such as applicable law, standards for determining breach, correct forum for resolution of the dispute etc.

 

      The umbrella clause began making its appearance in treaties in the 1950s,[9] yet,

 

  the question of whether an international arbitration tribunal had jurisdiction over contractual counter-claims was never fully examined, nor was the question of whether contractual jurisdiction clauses should oust – or precede – the jurisdiction of treaty-based tribunals.[10]

 

The present decade has, perhaps, more than made up for the prolonged hibernation of the umbrella clause in cases and scholarly works. Two decisions of the ICSID in the earlier half of this decade in SGS v. Pakistan[11] and SGS v. Philippines[12] have taken opposite viewpoints on the scope and effect of an umbrella clause. Since these two decisions, the debate in scholarly work has only intensified.[13] Awards of Tribunals continue making attempts to reconcile the two SGS cases; or to establish why one of the two cases should not be followed as a matter of principle.[14] Most recently, in September 2009, the issue was considered in Toto Costruzioni Generali v. Lebanon.[15] But it is not yet possible to say that the debate has been settled conclusively one way or the other. Several issues remain unclear.

 

What is the scope of an umbrella clause? Should there be a presumption for or against the incorporation of contract claims within a BIT? Or can one begin from another presumption altogether in mapping the relationship between contract claims and treaty claims under an umbrella clause? How would the scope of the clause affect jurisdiction of international BIT tribunals over contract and other non-BIT claims? Can parallel proceedings be allowed in both international (BIT) and domestic (contractual) tribunals; and if not, which of the two forums should take precedence?^[16] This note seeks to explore the world of the umbrella clause with an aim to arrive at a richer understanding of some of these issues. In particular, the first issue that the author will examine is whether the umbrella clause converts all contract claims into treaty claims. To answer this question, three views must be considered, that the clause (i) does not convert any contract dispute into treaty disputes;[17] (ii) converts all contract disputes into treaty disputes;[18] or (iii) it converts some contract disputes into treaty disputes.[19] In this last scenario, one must consider how to differentiate between various forms of contractual disputes, and arrive at a workable test to draw the line between the different forms.[20] This note argues that the third possibility is to be preferred, and highlights a workable model of that possibility.

 

      This leads on to the second question. Does the clause allow the BIT dispute resolution provisions to override contractual dispute resolution provisions? Again, one might take three views. First, it is possible that the BIT dispute resolution clause always overrides the contractual clause and contract claims are always admissible before a BIT tribunal.[21] Second, it is possible that the BIT dispute resolution clause never overrides the contractual dispute resolution clause.[22] The third situation would be where the BIT dispute resolution clause may sometimes override the contractual dispute resolution clause. Again, in this third possibility, a workable test needs to be determined. This note argues, endorsing the first of these three views, that there are strong reasons for allowing the ICSID to adjudicate upon disputes arising from that category of contracts which are included within the scope of the umbrella clause, despite the prescription of another method of dispute resolution in the specific contracts.[23]

 

      Part II of this note briefly discusses several of the important decisions on the issues. In this part, the author also considers and rejects the possibility of harmonizing the divergent cases on the point with each other. Part III discusses the question of the scope of the umbrella clause and analyses what obligations the clause would cover. Part IV moves on to the issue of competing jurisdiction between a BIT tribunal and a domestic contractual tribunal.

 

II.   The Arbitral Case Law: Is a Common Thread Visible?

 

      In discussing the arbitral decisions on the point, it will be useful to begin with an overview of the two SGS cases,[24] as these cases are the fulcrum of much of the debate.

 

A.   The Pakistan Case

 

      In SGS v. Pakistan,[25] the claimant had made an investment under the Swiss-Pakistan BIT, and had entered into a pre-shipment inspection contract with the government of Pakistan. The BIT contained a provision referring disputes to the ICSID, and also an umbrella clause that stated, [e]ither Contracting Party shall constantly guarantee the observance of commitments it has entered into with respect to the investments of the investors of the other Contracting Party.^[26] On the other hand, the pre-shipment contract also contained a dispute resolution clause which provided,

 

[a]ny dispute, controversy or claim arising out of, or relating to this Agreement, or breach, termination or invalidity thereof, shall as far as it is possible, be settled amicably. Failing such amicable settlement, any such dispute shall be settled by arbitration in accordance with the Arbitration Act of the Territory as presently in force.^[27]

 

The ICSID Tribunal had to decide as to what kind of disputes it would have jurisdiction over. Would it have jurisdiction over purely treaty-claims; or could it also adjudicate on disputes relating to the pre-shipment contract? Indeed, in the facts of the case, it was not contested that the dispute arose primarily out of the breach of the contract – Pakistan characterized this as a purely contractual matter; while SGS characterized it as an investment-law matter.

 

      Pakistan objected to the jurisdiction of the Tribunal over the disputes relating to the pre-shipment contract. It argued that the parties had freely consented to a more specialized dispute resolution clause in the pre-shipment contract; and that clause should be allowed to prevail over the general clause in the BIT.^[28] In essence, Pakistan was arguing on two levels. First, it was saying that the ICSID tribunal did not have jurisdiction at all, given that the claims were essentially contractual. Secondly, it was also arguing that even if the Tribunal were to find that it did have jurisdiction, it should nonetheless choose to not exercise that jurisdiction in the particular case because of the existence of an alternative dispute resolution procedure.

 

      SGS on the other hand argued that the umbrella clause in the BIT elevated all contractual disputes to the status of investment treaty disputes. It argued that characterization of a dispute was a matter which was to be decided by the claimant; and at the stage of establishing jurisdiction, it was not open to question the claimants characterization. It was only at the stage of deciding on the merits that the correctness of the characterization could be gone into. Further, SGS argued that the dispute resolution clause in the pre-shipment contract could not in any way curtail the jurisdiction of the ICSID tribunal, which was founded on the ICSID Convention and the Swiss-Pakistan BIT.

 

      In deciding on whether the umbrella clause elevated a contractual clam to a treaty claim, the Tribunal properly considered the matter to be one pertaining to the interpretation of treaties. Under settled principles of interpretation of treaties, an interpreter must seek to give effect to the plain meaning of the clause being interpreted, seen in light of the objects and context in which it was enacted.^[29] The Tribunal then went on to hold^[30] that the umbrella clause used the word commitments, and not contractual commitments. Further, the phrase constantly [to] guarantee the observance did not, in the Tribunals opinion, lead to the creation of a new international legal obligation. The Tribunal also felt that adopting the Claimants interpretation would result in an opening of the floodgates – many hundreds of state contracts and commitments would suddenly become international legal obligations. It was felt that this amounted to an imposition of international obligations on the state against its consent. In this light, the Tribunal held that the umbrella clause was merely procedural and did not contain any substantive legal obligation. Therefore, the Claimants argument was rejected; and the Tribunal declined to assume jurisdiction over contractual claims.

 

B.   The Phillipines Case

 

      SGS v. Philippines had similar facts; but involved a different umbrella clause,[31] which stated [e]ach Contracting Party shall observe any obligation it has assumed with respect to specific investments in its territory by investors of the other Contracting Party.^[32] The arguments raised by the parties were similar to those in the Pakistan case, and hence, require no elaboration.

 

      The Tribunal noted that the clause in this case was much more specific and imperative than the clause in the Pakistan case. Further, considering that BITs are entered into for the promotion and protection of investments, it was legitimate to resolve any ambiguities in the wording of the BIT in favour of the investor. The Tribunal also expressed its disapproval of the approach in SGS v. Pakistan, saying, [n]ot only are the reasons given by the Tribunal in SGS v. Pakistan unconvincing: The Tribunal failed to give any clear meaning to the umbrella clause.^[33] Therefore, it was held that the umbrella clause did mean that contractual disputes were elevated to being treaty-based disputes by virtue of the umbrella clause.^[34] Nonetheless, the Tribunal brought in an important refinement – it held that its conclusion did not however mean that the exclusive jurisdiction clause in the contract would be overridden. That clause was also a specific clause entered into by agreement of the parties; and was a specific clause overriding the general clause in the BIT. Therefore, given that there was an alternate forum for the proceedings, the Tribunal stayed its proceedings.

 

It appears that the Tribunal in SGS v. Philippines was relying on the well-known distinction between jurisdiction and admissibility in reaching its conclusion. The Tribunal essentially said that while it did have jurisdiction over the contract-claim; that claim was still not admissible because of the specific dispute-resolution clause in the contract. Thus, the Tribunal found that it did have jurisdiction, but chose not to exercise it. The Tribunal would have exercised the jurisdiction had there been no specific dispute-resolution mechanism in the contract. This approach is distinct from the one taken in SGS v. Pakistan, where the Tribunal refused to find that it had jurisdiction in the first place. In sum, SGS v. Pakistan denied that an umbrella clause effectively converted a contract claim into a treaty claim. SGS v. Philippines allowed such elevation, but maintained that the contractual dispute resolution procedure should be accorded primacy insofar as breaches of contract were alleged.

 

C.   Can a Common Thread be Seen?

 

Given that these two decisions have generated such great debate in international investment arbitration, it is worthwhile to see whether they can actually be reconciled with each other. Indeed, as already noted above, the specific clauses in the two cases were differently worded. It might be contended that the former clause is much weaker than the latter; and the latter is worded so that a strong imperative meaning can be drawn out.^[35] This argument, it must be said, is rather unhelpful. It is hard to see how shall constantly guarantee the observance of commitments is meaningfully weaker than shall observe any obligation. Is observing obligations significantly different from guaranteeing commitments? Certainly not in the context of umbrella clauses – the question is which commitments and obligations are to be guaranteed or observed.[36] Next, does the word any in the Phillipines clause make a meaningful difference? Perhaps, at first glance, this might indeed appear to be the case. In reality, however, this approach is mistaken. Guaranteeing commitments must necessarily mean guaranteeing any commitments unless later read down by the clause itself. The conflict between the SGS cases is best viewed not from the perspective of textual distinctions but as symptomatic of an ideological divide in international investment law which ought to be bridged.^[37]

 

      Two other factors are important in concluding that there is indeed no direct reconciliation between the two cases. First is the fact that the reasoning of the tribunals in both cases is incompatible with each other. Second, mere resolution of the conflict between the two cases will not be sufficient to solve all problems. The question still remains, on law, what is the proper presumption to draw (in the absence of clear evidence either way) as to the scope of the umbrella clause? Should the presumption be that contract claims are covered (following Philippines) or that contract claims are not covered (following Pakistan)? Attempts to reconcile the two cases are therefore unhelpful – they do not answer this basic question; and we must choose between the two options or devise a third one.

 

D.   Subsequent Arbitral Awards

 

      Before proceeding to an analysis of the umbrella clause itself, it will be useful to survey a few other decisions of the ICSID, which shed some light on the interpretative questions under consideration. In particular, the decisions in El Paso, Noble Ventures, and Toto shall be briefly examined.[38]

 

      In El Paso, the key question pertained to whether an umbrella clause had the effect of transforming all contractual undertakings into international law obligations and, consequently, turning breaches of contract into breaches of the BIT.[39] The Tribunal held that:

 

[T]he interpretation given in SGS v Philippines does not only deprive one single provision of far-reaching consequences but renders the whole Treaty completely useless: indeed, if this interpretation were to be followed – the violation of any legal obligation of a State, and not only of any contractual obligation with respect to investment, is a violation of the BIT, whatever the source of the obligation and whatever the seriousness of the breach[40]

 

Premised on this, the Tribunal went on to draw a distinction between a State acting as a merchant and the state acting as a sovereign. It was held that the umbrella clause was relevant only insofar as the state was acting in its sovereign capacity. Accordingly, purely commercial contractual obligations would not be elevated to treaty claims.[41]

 

Accordingly, El Paso advocated the middle approach. It is submitted that while the Tribunal reached the correct conclusion that not all contractual disputes are elevated to treaty disputes, its reasoning is open to criticism. Additionally, the acting as a sovereign test, which the Tribunal adopted does not have a sound legal basis.[42] The positive case for the correct legal test will be made subsequently in this note; for now, it will suffice to state in brief why the reasoning of El Paso is not entirely convincing.[43] First, one of the major criticisms which may be levied at El Paso is that its sovereignty test is both, ambiguous and unsupported in public international law;[44] it has no textual warrant and is capable of producing arbitrary results.[45] Second, in El Paso, the Tribunal noted that if an umbrella clause were to be given a wide meaning, it would be sufficient to include a so-called umbrella clause and a dispute settlement mechanism, and no other notes setting standards for the protection of foreign investments.[46] This view, it is submitted, is incorrect. The umbrella clause may have the effect of converting certain obligations into breaches of the treaty; that still requires some obligation to exist outside the clause. For instance – and given that the El Paso Tribunal specifically adverted to this – one may consider the standard of fair and equitable treatment (FET). The Tribunal suggests that the FET standard would be useless; as anyway, the obligation would be subsumed by the umbrella clause. This is true only if the FET standard represents a customary obligation. Only then would it amount to an obligation in the first place. There is great debate as to whether FET is customary.[47] Whatever be its customary status, the content of the FET obligation turns on the wording of the FET clause.[48] Thus, whatever be the interpretation of the umbrella clause, the content of the FET standard depends on the FET clause itself. Clearly – and one would have thought, obviously – the other clauses do matter even with the umbrella clause being widely interpreted. The El Paso Tribunals concerns that a wide umbrella clause would render all other BIT clauses unnecessary, is unpersuasive. Finally, the El Paso decision has been supported in some commentaries on the ground that a wider interpretation of the umbrella clause would leave the investor open to make treaty claims in purely commercial contracts.[49] This is true only if an alternative, stronger test cannot be found. As shall be demonstrated subsequently, such a test does exist – and not all contractual matters come within the fold of BIT protection.

 

      At the same time however, it cannot be said that the entire approach of El Paso was incorrect – El Paso appears to have been motivated with the right concerns but failed to establish a sufficiently strong legal basis for that approach. The reason why this extremely liberal view is unwarranted will be evident on considering one of the broadest interpretations of an umbrella clause – Noble Ventures.[50] This case held that:

 

in the interest of achieving the objects and goals of the treaty, the host state may incur international responsibility by reason of a breach of its contractual obligations the breach of the contract thus being internationalized.[51]

 

These observations by the majority have been taken to imply that all contractual obligations are converted into treaty obligations.[52] Additionally, this result ignores the wording of most umbrella clauses, which refer to obligations in relation to investments.[53]

 

Furthermore, in the facts of Noble Ventures, the contract was between the claimant and the Romanian State Ownership Fund. The Tribunal failed to consider whether the obligations under the umbrella clause pertained only to contracts entered into with the State itself, or whether it also included contracts entered into with public companies or local subdivisions/governmental units of the State. On the face of it, that issue would be simple as the acts of the local subdivisions/units would be attributable to the State itself, on the public international law principles of attribution.[54] Indeed, this was mentioned in SGS v. Pakistan itself.[55] However, a deeper examination would reveal that attribution is not relevant in these cases.[56] It is only contracts with the State itself which would be covered.

 

It must be noted that the umbrella clause is used in these cases to assert jurisdiction over contract claims. Typically, a BIT dispute resolution mechanism is seen as an offer to arbitrate by the State, which is accepted by the investor.[57] In deciding issues of jurisdiction, the use of attribution to include contracts by local units of the State would imply that the States offer to arbitrate is being attributed downwards onto the local units.

 

This would in fact be a case of attribution of a States act to its local unit, thereby subjecting the contract with the local unit to the States offer to arbitrate. This is clearly an incorrect application of the principles of attribution.[58] As Professor Crawford notes:

[T]he issue of attribution arises when it is sought to hold a state responsible for some breach of an international obligation The problem here concerns jurisdiction, not merits; the formation of a secondary agreement to arbitrate, not the breach of a primary obligation concerning the protection of investments the question of the scope of a commitment to arbitrate made by the State is a matter of interpretation and has nothing to do with attribution. International law does not treat separate entities with their own legal personality as part of the State for all purposes[59]

 

On a wording of the BIT, it would be only those contracts made by the State which are subject to the States offer to arbitrate. The wide reading of the umbrella clause premised on Noble Ventures would err in not considering this point on the irrelevance of attribution – the umbrella clause would in reality cover only those contracts which are entered into with the State itself.

 

      This then brings us to the decision in Toto;[60] which is amongst the latest pronouncements on the point. Toto follows the Philippines logic. Toto is most relevant for its discussion on whether a contractual dispute resolution clause would prevail over the BIT arbitration clause. Here, the Toto Tribunal followed the majority opinion in the Philippines case, and held:

 

The contractual claims remain based upon the contract; they are governed by the law of the contract and may be affected by the other provisions of the contract. In the case at hand that implies that they remain subject to the contractual jurisdiction clause and have to be submitted exclusively to the Lebanese courts for settlement. Because of this jurisdiction clause in favor of Lebanese courts, the Tribunal has no jurisdiction over the contractual claims arising from the contract referring disputes to Lebanese courts.[61]

 

      Interestingly, the Tribunals observations are couched in terms of lack of jurisdiction itself. The Philippines Tribunal did not rule that it had no jurisdiction at all; rather, it found it fit to stay proceedings in view of the contractual dispute resolution clause.[62] The observations in Toto seem to have gone further, in classing the matter as one of jurisdiction itself.[63] The correctness of this approach – the question of whether a contractual dispute resolution clause prevails over the BIT dispute resolution cause – will be examined subsequently; for the moment, this note will turn its attention back to the scope of the umbrella cause itself.

 

III.  The Scope of the Umbrella Clause

 

      In attempting to resolve the difficulties surrounding the application of the umbrella clause, it is essential to take into account the fact that the umbrella clause is part of an investment treaty. Therefore, it is the principles of interpretation of enterprises which must serve as guideposts to our future journey.

 

A.   The Basic Principles of Interpretation

 

      The customary international law principles dealing with the interpretation of treaties find themselves codified in the Vienna Convention on the Law of Treaties.[64]These principles have achieved almost universal consensus in international law. The general rule of interpretation, Note 31 of the Convention, is born out of a compromise between two competing schools of legal interpretation – the literal and the teleological. The note privileges neither the strict meaning of the words in the treaty, nor does it leave the words meaningless. Thus, Note 31 requires a treaty to be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.[65] Thus, our interpretation of the umbrella clause must be guided not simply by textual distinctions between obligations and commitments, but should take into account the context in which the umbrella clause operates and the object of the clause. An umbrella clause of the type with which we are concerned with at the present is invariably in the context of an investment treaty – a treaty between two nations to promote and protect foreign investment between them. In order to appreciate how a general umbrella clause operates within this context, one will have to appreciate the object behind the introduction of the umbrella clause in investment treaties. A brief historical detour is therefore necessary.

 

B.   A Short Look at the History of the Umbrella Clause

 

      A concrete formulation of the umbrella clause appeared in the Abs-Shawcross Draft Convention of International Investments abroad, 1959.[66] This Draft Convention used the term any undertakings, and scholars even in the early 1960s understood the importance of these words. A leading international law scholar expressed the view that the clause covered undertakings by contracting parties both to subjects and objects of international law.[67] Professor Fatouros wrote more clearly that the umbrella clause was intended to cover cases of contractual commitments of states to aliens.[68] Further, it was stated by another leading scholar at that time that the idea behind the inclusion of the umbrella clause was to dispel whatever doubts may possibly exist as to whether a unilateral violation of a concession contract is an international wrong.[69] These statements clearly went on to influence the next important draft convention – the OECD Draft Convention on the Protection of Foreign Property. The commentary to that Convention clearly formulates the idea that the umbrella clause was meant to elevate any right in a contract or a concession between a state and a private investor to an international right.[70]

 

 

C.   The Relevance of the History

 

      If one were to look merely at these drafts, the conclusion seems inescapable that the clause was meant to elevate all contractual disputes into treaty disputes.[71] A few problems remain unsolved before one actually reaches this conclusion. These are merely draft conventions. How appropriate is it to rely on the commentaries of a draft convention – influenced undoubtedly by the political and economic agenda of the drafters – in interpreting actual treaty provisions? It might well be contended that these drafts are written from a developed-state perspective. In the world of investment disputes, it is more likely than not that the investor will be from a developed nation, and the investment will be made to a developing nation.[72] It is not too fanciful to imagine that the drafts would favour extending investment protection. When actual BITs are signed, however, such an understanding may not actually exist between parties to the BIT. What we therefore need to do is to achieve some interpretative balance which would enable us to best approximate the common intention of the two parties to the BIT.[73]

 

      How then can we proceed to balance the interests of the parties involved so as to best approximate the intention of the parties to the BIT? The logical first step must be to identify the various interests. There are at least three interests involved – the interest of the investor, the allied interest of the State to which the investor belongs, and the allegedly conflicting interest of the State in which the investment is made. It would appear at first glance that the interest of the investor is to allow for a wide-ranging ICSID jurisdiction, while that of the Respondent State would be to limit jurisdiction.[74] This formulation is, however, a bit simplistic – it does not take into account the fact that the Respondent State (even if it is a developing state) will almost always enter into a BIT in order to promote investments within its territory. As a long-term strategy for such a state, it might well be unwise to adopt on principle a restrictive model of dispute settlement. Therefore, the fact that the drafts mentioned above were written from the perspective of developed states does not necessarily mean that the interests of developing nations in this regard are necessarily conflicting.[75] And if any such interests were actually conflicting, the state could always rely on the intention in the specific BIT in question. As a matter of presumption, however, the fact that all interests may not have been adequately represented is not sufficient to ignore the existence of the several draft investment treaties. The drafts recognize the thought that BITs are a touchstone of international relations[76] and do play the valuable role of highlighting the introduction of the concept of the umbrella clause in international investment arbitration.[77] That role ought not to be unnecessarily ignored on an alleged conflict of interests between developing and developed countries.

 

D.   Interpreting the Clause

 

      Thus, an examination of the objects of the typical umbrella clause reveals that the clause is concerned with extending the scope of investment protection. It still needs to be established, however, that this object and context must translate into a wide-ranging interpretation of the umbrella clause. Indeed, the narrow interpretation will lead to the umbrella clause being rendered useless, in violation of the principle of effet utile which suggests that every provision of a treaty must be given some meaning. If the umbrella clause covers only treaty-claims, then effectively, it is rendered meaningless – the treaty claims must, by definition, anyway be covered under another specific provision of the BIT. At the same time, the wording of the umbrella clause typically talks about all obligations in relation to the investment. Just as the word all cannot be ignored to include solely treaty-based obligations; so too, the words in relation to the investment cannot be ignored to include all contract-based obligations.

 

E.   A Conflict in Methodologies

 

      The issues which need to be addressed can perhaps be looked at as a conflict between two methodologies or approaches to international investment law.[78] The Pakistan approach is an example of a disintegrationist approach. This approach maintains clear lines between different types of legal norms. Consequently, it interprets BITs as self-contained codes of investment protection. The integrationist approach on the other hand seeks to harmonize interlocking legal norms. It favours looking at BITs as open-ended instruments.[79] Can one interpret an umbrella clause so as to impute an arbitration agreement, without that imputed agreement suffering from vagueness or from unclear consent? Indeed, the very notion of imputing an agreement indicates that there was no express consent for the agreement. In such a scenario, how far can states be assumed to have consented to adoption of an integrationist viewpoint? Clearly, states must attach importance to their own domestic legal systems. They cannot be taken to have consented to an integrationist approach to all legal obligations.[80] By entering into the BIT, they could have intended – at best – an integrationist approach to investment protection. Reading this in light of the objects and the context of the typical umbrella clause, it appears that the correct view to take of the matter would be the middle path.

 

      The object of an umbrella clause is to promote the scope of investment protection. Indeed, the focus is not on promoting the investor at the expense of the host state. Instead, the focus is on promoting the investment. With respect to states too, the act of entering into a BIT might be seen as having accepted an integrationist view towards investment protection – they can have hardly approved of an external legal system to govern all aspects of their relationship with investors. The extremely wide view of the umbrella clause including within its scope all contracts which are likely to give rise to unintended consequences with the state being held to account for the contractual performance of entities over which it has little or no practical control[81] runs the danger of driving states away from BIT-related systems, thereby seriously hampering investor-state relations.[82] Of course, this argument does not appreciate the non-application of the rules of attribution, discussed previously in this note in discussing Noble Ventures.[83] Yet, the contrary argument does run the risk of ignoring the words in an umbrella clause which refer to obligations in respect of an investment. Additionally, given that a great deal of jurisprudence has developed around the notion of investment, risks of uncertainty are consequently lower.[84] One of the main criticisms of the sovereignty limitation of El Paso, we have seen, is that there is no determinate legal test or textual support for the test.[85] Both these difficulties do not arise when the investment-contract test is adopted. That being the case, the most appropriate solution – both in law and on policy – seems to be that the umbrella clause would cover those disputes which have a nexus with the investment – in other words, investment related contracts will get protection under the umbrella clause.

 

F.   The Investment Test to Limiting the Scope of the Umbrella Clause – Reconciling Object, Context and Text

 

      This approach also best integrates the object and context of the umbrella clause with the text.[86] The focus is – so to say – on neither the State nor the investor. The focus is on the investment. Thus, the test to differentiate between contracts protected under the BIT and contracts not protected under the BIT must depend on the nature of the particular contract. Is the particular contract closely linked to the investment? Is it a contract on which the rights of the investor in his capacity as an investor are affected? In some cases, the contract will clearly be closely linked to the investment – for instance, contracts with the government to take over the investment; or contracts essential for the continuance and survival of the investment. These contracts affect the investor as an investor and not just as any other contracting party. In these cases, the contract will be within the scope of the BIT. In other cases, the contract will be purely commercial.[87] Take for instance a simple contract entered into by the investor with a government agency for the supply of certain goods. In such cases, unless the supply of those goods is vital for the continuance of the investment, it is not easy to say that the contract is closely linked to the investment.[88] Therefore, there is no reason why such a contract should also be entitled to BIT protection. The test is as follows – is the contract one which is entered into in relation to the investment and in the capacity of an investor, or is the contract one entered into simply as a commercial contract without particular focus on the investment itself? The former category will be governed under the scope of BIT protection; while the latter will not.

 

      It would then appear that this is the most satisfactory solution to the question of the scope of the umbrella clause. That still leaves open the second question. Assuming that there is a dispute-settlement provision in an investment-related contract, will the ICSID tribunal be justified in assuming jurisdiction ignoring the particular dispute-resolution mechanism?

 

IV.  Umbrella Clauses and Contractual Forum-Selection Clauses

 

      In support of the contention that the ICSID tribunal should not ignore the specific dispute-resolution mechanism, one strong argument which can be raised is that the specific mechanism in the contract is the clearest indication of the intent of the parties and must prevail. Under international law, the maxim of lex generalis non derogat lex specialis entails that a general provision cannot override a specific provision.[89] A BIT dispute redressal mechanism is a general mechanism in relation to investments, while a contractual mechanism is a special mechanism in relation to a particular contract. As such, it can be argued that the BIT mechanism cannot override the contractual mechanism. The investor concerned has consented to a separate dispute resolution mechanism. In such a case, why should that choice not be given effect to? To this question, there are several answers possible.

 

      At the outset, it is essential to note that we are already dealing with a very limited class of contracts – those dealing with investment protection. In the case of such contracts, the existence of umbrella clauses means that those investment contracts are in fact so closely linked to the BIT that they can be considered to be obligations arising from the BIT itself. In such a case, saying that contractual dispute resolution provisions would prevail would mean that even in the case of a dispute which is closely linked to the BIT and arises out of an obligation from the BIT, the BIT dispute resolution provisions would be rendered ineffective.[90] In the case of investment-contracts of the type which are seen to be covered under the umbrella clause, the maxim of lex generalis non derogat lex specialis cannot apply. This is because in the case of investment contracts, the nature of the contract is such that it is effectively similar to a BIT in terms of substance. As such, it can hardly be considered to be of a genus different from the BIT.[91] Indeed, a violation of that contract would amount to a violation of the BIT itself. While ICSID arbitration is dependent on consent of the parties,[92] the proliferation of BITs suggests that it is the preferred mode of arbitration of investment-related disputes. In such a case, a different dispute resolution mechanism under another agreement cannot be said to affect or vitiate this over-arching consent to submit investment disputes to ICSID arbitration.[93]

 

A.   Violations of BITs as well as Contracts

 

      When one says that investment-contracts are covered within the scope of the umbrella clause, this means that the violation of the contract will amount to a violation of the umbrella clause of the BIT. Thus, the BIT is violated as well as the contract. Allowing the contract forum selection clause to prevail would imply that the two causes of action are identical.[94] A breach of a treaty is conceptually distinct from a breach of a contract – even when the breach of a contract is an ingredient in establishing a breach of the treaty.[95] In such circumstances, allowing a contractual dispute resolution provision to prevail would be tantamount to allowing a contractual mechanism to supersede a BIT mechanism even in situations where the BIT has been violated. It would mean that the contractual dispute mechanism has the authority to decide on BIT disputes. Or, looking at it from another point of view, when a BIT Tribunal is held to have no jurisdiction pending a contractual mechanism, it would mean that the violation of the treaty arises not from a failure of the State to observe its obligations, rather from a determination by the domestic Tribunal of such failure. But, the typical umbrella clause is clear – the clause will be breached when the individual contract itself is breached; not just when the individual contract has been held to have been breached.[96]

 

      Thus, giving preference to a contractual dispute mechanism runs the risk of conflating two different things – the breach of a contract and a determination of the breach of a contract.[97] Such a reading is clearly impermissible under the text of the typical umbrella clause and will lead to a legitimacy crisis in the field of investment protection[98] Thus, if an investment contract prescribes a mode of dispute settlement, that mode must be considered as giving an additional choice to the aggrieved investor, but not as doing away with an established BIT mechanism.[99] The BIT mechanism is the result of a treaty between two sovereign nations; this cannot in international law be overridden or allowed to be overridden by a contract between a state and a private party. Indeed, allowing such an override will be the international investment law analogy of contracts against public policy. It is clearly the policy of international investment law to avoid excessive fragmentation of investment law by leaving the settlement of investment disputes to an excessive number of local bodies.[100] The object of BITs and of international investment law – the very policy behind the emergence of international investment protection – will be defeated by allowing states to subvert internationally agreed procedures.[101]

 

B.   Countering the Counters

 

It has been contended that there is little evidence that by increasing the role of the ICSID to cover contract claims, more certainty is provided in the international investment regime.[102] However, under the reading of the umbrella clause adopted by this note, it is not every contract claim which will be covered under the ICSID umbrella – it is only those contracts which have a nexus with the investment. In the case of this specific category of contracts, clearly certainty will be promoted by preferring ICSID arbitration to localized remedies.[103]

 

At this juncture, it is worth appreciating the main reasons given by the ICSID Tribunal in Philippines in refusing to decide on the merits of the contract claims-this despite holding that it did have the jurisdiction to do. The Tribunal considered that the fact that the umbrella clause elevated the contract claim in the case to a treaty claim could not hide the origins of the obligation. This approach is not particularly convincing; the Tribunals reasoning as to why once the claim is elevated to a treaty claim the incidents of the BIT should not follow is unclear. Further, on the facts of the case, the Tribunal was clearly expecting this reasoning to apply to pure contractual claims. But, it has already been noted in this note that pure contractual claims are not correctly elevated to treaty claims; it is only investment-related contract claims which are elevated. The source of obligations in these contracts is closely connected to the investment, and is a subject-matter of the BIT. As such, the reasoning as applied to pure contractual claims may not be appropriate in the specific class of investment-related contracts.

 

Indeed, the ICSID seems to have recognized this aspect in an earlier decision itself, as it took pains to point out that the contract forum-selection clause would override in cases where the essential basis of the claim is contractual.[104] In investment-related contracts, the essential basis of the claim is still tied to the BIT itself. Although the form of the obligation is contractual, the substance is related to the investment protection structure.[105] Therefore, there is no compelling reason for the ICSID to defer to the contractual dispute settlement mechanism as a matter of jurisdiction; and on balance, it seems to be well founded to suggest that the ICSID should not refuse to hear disputes arising out of investment-contracts even if those contracts provide for other means of dispute settlement.

 

C.   A Theory of Deference on Merits?

 

      At the same time, practical and efficiency concerns ought not to be undermined.[106] Theoretically, both the BIT Tribunal as well as the domestic tribunal might have jurisdiction as the two causes of action are separate – breach of treaty and breach of contract respectively. Yet, it is hard to deny that similar factual determinations would have to be made in both cases, where the breach of a contract is an ingredient in the breach of the treaty. The stronger reasoning in Philippines, thus, would have been to stress that the investor had invoked the contractual remedy; and that repetitive adjudication should be avoided.[107]

 

      On the face of it, this begs the question of why that should matter when the causes of action are wholly different. The answer lies, perhaps, in determining a theory of deference, rather than treating the question as one of jurisdiction and/or maintainability. How much should a BIT Tribunal defer to the conclusions of a domestic tribunal, when the breach of a contract – adjudicated upon by the domestic tribunal – is at issue in determining BIT breaches before the BIT tribunal? This is – it will be noticed – not a question which requires an answer at the threshold jurisdictional stage. It might be argued that the matter is essentially similar to the concept of res judicata. Arguably, res judicata is a general principle of law under Art.38 of the Statute of the International Court of Justice; and would preclude jurisdiction itself. It is submitted that res judicata would not apply in a situation where the two causes of action are essentially different, as in the case of a breach of treaty versus a breach of contract.[108] The question here – given the different causes of action – is more of issue preclusion rather than claim preclusion; and civil law countries typically recognise only a narrow version of res judicata which would not encompass pure issue preclusion.[109] If res judicata is then a part of the general principles of law under Art.38, it cannot encompass issue preclusion. Hence, the matter here is not resolvable as a matter of jurisdiction on the basis of res judicata.

 

      In any event, the arguments made thus far in the paper – pointing towards the objects of BIT protection, would indicate that jurisdiction cannot be barred. Allowing a res judicata type argument would only enable a result being reached contrary to what the parties have agreed. Hence, there are strong theoretical difficulties in treating the matter as one of jurisdiction. Rather, the matter is best treated as one pertaining to the level of deference to be given to the domestic Tribunal in a matter of merits.[110] This will turn on appropriate theories of deference; and it is hoped that such a theory could be developed in this context.[111] The formulation of such a theory would be a better approach rather than treating the questions as a matter to be decided at the jurisdictional threshold.[112]

 

V.   Conclusion

 

      The role of the umbrella clause is bound to continue giving rise to interpretative difficulties. This note, it is hoped, points the way to a tenable solution keeping in mind the principles of treaty interpretation. Beginning with a survey of some of the conflicting approaches taken by tribunals on the issues under consideration, this note rejects any attempt to find an all-encompassing theory to reconcile the divergent streams of thought. The problem, as the note postulates, needs to be solved by choosing between different methodologies and not by making attempts to bring together all differences under a single umbrella explanation. Having considered the different streams of decisions, it is suggested that an umbrella clause cannot be so wide in scope as to encompass all obligations; yet, the clause must also not be rendered meaningless. The question then turns to finding a mechanism – a legal test – to understand what obligations are covered under the umbrella clause and what are not. A test proposed by some tribunals – the sovereignty test – stands rejected by this note; and a test based on the meaning of investment appears to be the most robust alternative. At the same time, an allied question pertaining to the scope of jurisdiction upon an entity by whom the obligation was entered into – the State itself or even the domestic sub-units of the State – was sought to be addressed. It is suggested that in dealing with questions of jurisdiction, particularly given the process by which the BIT agreement to arbitrate is formed, principles of attribution have no role to play. Thus, a two-fold limitation is seen on the scope of the umbrella clause – the obligation must be entered into by the State itself; and the obligation must be related to investment. Both these limitations, it is suggested, are compatible with the applicable principles of public international law.

 

      Given these two limitations, however, this note does not see any need for restricting the jurisdiction of BIT Tribunals on the basis of a contractual dispute resolution mechanism. This does, admittedly, raise questions of efficacy and duplication of judicial decision-making. The road towards the solution of these issues might appear – it is tentatively proposed – not in a jurisdiction/maintainability analysis of BIT Tribunals, but in developing a theory of judicial deference to domestic (contractual tribunal) decision-making as a matter of merits.

 

      Deference, generally, depends on the expertise of the authority which took the prior decision. In the case of ad-hoc domestic arbitrations, that aspect could become a very fact-specific matter; and conceivably, it would be a fruitful exercise for a general theory of deference to be concretized to aid in this factual analysis. Perhaps, seen in that perspective, international investment protection will achieve its real goal and object; while at the same time, taking into account practical efficiency concerns. Consequently, such a theory may light up the path towards further refinement of this part of the law governing international investments.