1. THE ‘CREATIVE’ LAW INTERPRETATION BY INVESTMENT TRIBUNALS

The present panel discusses how the arbitrators tend to reconcile the strict application of the law while giving consideration to the expectations of the trading community. Our moderator has outlined four sub-topics, the first of which is worded as follows:

“Interpretation of the rule (possibly tendentious) or characterization of the situation (possibly tendentious) in order to justify the result which the arbitrator wants to reach (maybe for good reasons)”.

He invited me to examine whether the above quoted proposition also applies in investment arbitrations.

Based on the case law, the answer is rather easy: it is undeniable that arbitral tribunals constituted under investment treaties have reached opposite results when applying the same or similar rule of law, or that annulment committees have endorsed a completely different outcome, or that the majority view has been strongly criticized by dissenters.

In investment disputes, the applicable rules of law are primarily the provisions of the relevant bilateral or multilateral treaty, then the pertinent rules of customary international law and then, to a more limited extent, the law of the State which is a party to the dispute.

It is the divergent interpretation of identical treaty provisions that, in particular, induces tribunals to reach opposite results, each tribunal defending its views as the most correct in law and the most apt to capture the parties’ real intentions.

The jurisprudential divide has emerged with respect to several crucial matters. However, I will focus my analysis on those that seem to be the most exemplary, and still unsettled, such as:

(i) whether a Most Favoured Nation (MFN) clause may be applied as a source of jurisdiction for the arbitral tribunal;

(ii) whether the ‘umbrella’ clauses convert a contract claim into a treaty claim and make it arbitrable.

I will then more succinctly address some other matters with respect to which investment tribunals are equally unable to reach consensus or jurisprudence constante, such as the extent to which the host State may validly raise the ‘necessity defence’ to avoid liability for breach of the investment treaty, and the precise notion of ‘protected investment’ as a condition for attributing jurisdiction to investor-State tribunals.

2. THE MFN CLAUSE AS A MEANS TO ASSUME JURISDICTION

Several Bilateral Investment Treaties (BITs), while providing for access to arbitration, including ICSID arbitration, make the investor’s right to arbitrate conditional on the previous resort to local courts, or limit the scope of arbitration to a restricted category of disputes.

A good example is given by Argentinian BITs, which typically provide for an 18-month local litigation as a precondition to commence ICSID (or UNCITRAL) arbitration. Certain tribunals have bypassed the condition precedent by importing, via the MFN clause contained in the applicable BIT, a different dispute provision found in another BIT in force between Argentina and a third State. Such other provision provides no condition precedent to pursue arbitration and is therefore viewed as ‘more favourable’ and thus incorporated by reference into the basic BIT by effect of its MFN clause. Following such an approach, these tribunals have declared the claimant’s claim immediately arbitrable and assumed jurisdiction despite the claimant’s failure to refer the matter to Argentinian courts for at least 18 months prior to commencing arbitration.1

One might question whether the difference between the basic and the third party BIT really affects the quality of the procedural treatment offered to foreign investors in one case and in the other. Consenting access to arbitration immediately or after exhaustion of the 18-month court litigation in the host country, does not drastically differentiate the procedural treatment reserved to foreigners in the dispute resolution. Both treaties provide for the investor’s right to eventually arbitrate and it seems implausible to argue that one treatment is less favourable than the other because arbitration is postponed by 18 months in order to allow an attempt of judicial resolution in such a short time. What matters in terms of substantive protection, is that the investor benefits from the right to arbitrate in both cases and is able to do so within a reasonably short time after the emergence of the dispute.

Therefore, the cases discussed here have perhaps magnified the difference in treatment and thus incorrectly triggered the application of the MFN clause where it was not strictly warranted. The 18-month waiting period was clearly required in an unambiguous treaty provision, substantially favourable no less than other provisions, granting access to arbitration without subjecting the investor to arduous or discriminatory conditions. In my view, the contracting States’ intention should be respected.

Be that as it may, the above-mentioned tribunals have noticed a substantial difference in treatment, thus looking to the most favourable treatment. They did so in order to assume immediate jurisdiction and decide the merits considering this approach wiser, or less disruptive in the economy of the process, rather than declining jurisdiction and sending the claimant back to local courts temporarily and with no hope of having the dispute decided in 18 months.2

Other tribunals followed the opposite line, giving to the treaty provision a literal interpretation and rejecting the claimant’s attempt to invoke the MFN clause to disregard a restriction or a condition precedent agreed by the two contracting States.3 They were keen not to ‘re-write’ the treaty provision and were not persuaded by possible concerns of procedural economy or futility of the 18-month attempt. Moreover, they gave no relevance to the alleged different nature of the dispute (domestic dispute before the Argentinian courts versus international dispute before the investment tribunal), a difference that was instead accepted by other tribunals.4 They simply wished to enforce the treaty as unambiguously drafted.

In brief, 10 investor-State tribunals have overcome the Argentinian BIT requirement for an 18-month prior litigation treating this clause as a mere procedural directive, falling within the discretionary power of the tribunal to observe or discard,5 whereas five other investor-State tribunals have treated the same clause as a mandatory jurisdictional requirement and a precondition to be satisfied in order to allow formation of the State’s consent to arbitrate.6 The jurisprudential divide is therefore 10-to-five.7

In BG v. Argentina, the tribunal rejected Argentina’s objection that the tribunal lacked jurisdiction due to the claimant’s failure to first bring the dispute before the State courts. The tribunal evaded the objection asserting that the State attitude made its own court system unreliable and the option futile, so that under international law principles and the Vienna Convention on the interpretation of international treaties non-compliance with the local litigation condition was ‘excused’.8 The seat of the (UNCITRAL) proceedings being Washington, Argentina challenged the arbitral decision before the competent district court, where the challenge was rejected.9 However, the challenge was upheld by the competent appeal court, which set aside the award on the ground that the two contracting States had not intended to assign the decision on arbitrability to the arbitrators. They had rather set forth a condition precedent to the consent to arbitrate, that the arbitrators had no power to disregard.10 Recently, the challenge has been decided by the U.S. Supreme Court, which re-validated the arbitral decision, ruling that ‘The local litigation requirement is a matter for arbitrators primarily to interpret and apply. Courts should review their interpretation with deference’.11

In Plama v. Bulgaria, the matter in dispute was whether the MFN clause might be invoked to import into the basic BIT a broader or unrestricted arbitration clause taken from another BIT. The applicable BIT was the treaty in force between Cyprus and Bulgaria, the dispute provision of which confined arbitration only to the quantum of the compensation due in case of expropriation, while reserving the legality of expropriation to local courts. Relying on a BIT made by Bulgaria with a third country, the claimant requested the tribunal to assume jurisdiction on the basis of the dispute provision found in another BIT, pursuant to which arbitration was made available with no restriction as to the arbitrable matters. The tribunal rejected the claimant’s request and declined jurisdiction. It observed that the basic BIT clearly limited Bulgaria’s consent to arbitration and that it would be ‘exorbitant’ for an investor to ‘seek to replace’ the applicable BIT clause by cherry picking a totally different dispute mechanism from another treaty. It thus declined jurisdiction distancing itself from Maffezzini and its followers and criticized them for ‘neutralizing’ the treaty requirement for a prior recourse to local courts.12

In brief, the same rule of law was given two opposite meanings. As the reasoning of the tribunals show, each interpretation was justified by the desire to reach a predetermined result that, in the views of the tribunals, was the most appropriate: on the one side, giving prevalence to the investor’s immediate access to arbitration rather than to a futile or ‘nonsensical’ pre-arbitration litigation; on the other side, giving prevalence to the contracting State’s intention as reflected in the BIT, thus admitting arbitration only after that the State consent to arbitrate was validly formed.

Who is right, who is wrong? Were the tribunals deciding their jurisdiction under the competenz/competenz doctrine? This doctrine applies when mutual consent to arbitration is established and the tribunal must determine the precise scope of its jurisdiction. However, it might be argued that in the instant cases where the dispute had not been first referred to domestic courts Argentinian consent to arbitration was not yet perfected and, therefore, what was in question was arbitrability rather than competenz/competenz.

In conclusion, although the cases of the second category are a minority (five-to-10, as seen above) I would tend to side with them rather than with the majority of the decisions.

3. UMBRELLA CLAUSES

In addition to the question of whether MFN clauses are a means to validly establish or enlarge the jurisdiction of investment tribunals, the question of whether the ‘umbrella clauses’ confer upon investment tribunals jurisdiction over contract claims is the second-most controversial issue within the world of investment law.13

This clause, also known as the ‘observance of undertakings’ clause, is contained in several BITs.14 In essence, the host State guarantees to the foreign investor that it shall not only comply with the treaty provisions, but also with any other obligation undertaken by the State to benefit the investor and its investment outside the treaty, i.e. by contract, unilateral act, or unilateral promise. The umbrella clause is generally described as having a double effect: first, it allows elevation of a State contract breach to a State treaty breach; second, it confers upon the tribunal the jurisdiction to sanction a contract breach which, based on the first effect, has become a treaty breach.15 As seen below, however, in order for the above proposition to be correct it should be complemented by certain qualifications.

The issue relating to the meaning and legal effect of similar clauses was firstly faced by the ICSID tribunal in the case SGS v. Pakistan. That tribunal considered the clause too broad and too vague to produce an unlimited expansion of the jurisdiction of investment tribunals and refused to assume jurisdiction over contract claims based on such a clause.16 However, a few months later, another ICSID tribunal came to the opposite conclusion, admitting that, depending on their language, umbrella clauses may well have the effect of expanding the arbitral jurisdiction over contract claims.17

The matter was again discussed in Salini v. Jordan,18 where the tribunal was not satisfied that the clause at issue was an umbrella clause at all and held that it did not involve any guarantee by the host State with regard to specific transactions.

A balanced position on the umbrella clauses is offered by Professor Crawford, who — based on the discussion provoked by the three above mentioned cases asserts that the precise meaning of these clauses depends on their actual language and in order to be meaningful for the resolution of a case, they must be unambiguous. He identifies four schools of thought:

(i) a school advocating a narrow interpretation, holding that these clauses are operative only to the extent it is possible to discern a shared parties’ intent to make a contract breach equivalent to a treaty breach, which would deprive the clause of any content;

(ii) a school seeking to limit these clauses to breaches of contract committed by the State in the exercise of sovereign authority, which imposes an unwarranted characterization test for the level of breach, thus producing arbitrary results;

(iii) a third view adopting an extreme solution, namely internationalization tout court of investment contracts thereby transforming contract claims into treaty claims directly subject to treaty rules and treaty jurisdiction, thus allowing the investor to evade the exclusive dispute forum possibly agreed upon in the investment contract; and

(iv) a last school, according to which an umbrella clause is operative and may constitute the basis for a substantive treaty claim, however without automatically converting a contractual claim into a treaty claim, or changing the proper law of the contract, including its own mechanism for dispute resolution.

Professor Crawford endorses the fourth viewpoint, in which the claims remain contractual in nature and governed by their own applicable law, but the umbrella clause allows enforcement of the State’s undertakings without transforming the nature and content of the underlying obligations.19

The debate has remained unsettled and, more recently, two other divergent decisions were rendered in 2009 and 2010. In both cases, the dispute arose from contracts having the same object and nature, namely public services contracts entered into between Paraguay and two foreign companies specializing in pre-shipment inspections of goods to be unloaded in Paraguay as a form of assistance to the national customs authorities in fixing entry tariffs. In both cases, even the object and nature of the dispute was the same, namely the inexcusable failure by Paraguay to pay a rather significant amount of unpaid invoices issued by the services providers under the respective contracts. Finally, both cases were founded on BITs containing an umbrella clause.

Notwithstanding the similarity of the matters, the tribunals gave two opposite interpretations to the umbrella clauses.

In SGS v. Paraguay, the dispute arose from Paraguay’s arbitrary failure to pay a number of pending invoices totalling US$40 million payable under a customs assistance contract. Notwithstanding the underlying services contract provided that disputes arising thereunder should be submitted to Paraguayan courts, SGS invoked several breaches by Paraguay of the applicable BIT, such as violation of the duty of fair and equitable treatment, adoption of discriminatory measures and breach of the umbrella clause obliging Paraguay to perform contract commitments undertaken in respect of a specific investor and its investment. Paraguay objected that, in essence, SGS’s claim was purely contractual in nature, but the tribunal dismissed the objection noting that it had been called upon to resolve several treaty claims and assumed jurisdiction.20

However, when the tribunal examined the merits, it followed a strictly contractual approach and exclusively held Paraguay liable for breach of the contractual obligation to pay and, consequently, for breach of the BIT umbrella clause. It considered that it had no need to take position on the outstanding treaty claims because their potential effects were absorbed by the breach of the umbrella clause, sufficient to establish Paraguay’s liability under the treaty and the contract, and because consideration of the other treaty claims would not modify the quantum compensation.21

The approach followed by the SGS v. Paraguay tribunal was not entirely consistent. When assuming jurisdiction, it took distance from the contract and based its decision not only upon the umbrella clause, but upon the entirety of the claims for breach of the treaty raised by the claimant, the violation of the umbrella clause being one of the several BIT breaches alleged. Apparently, this approach was necessary in order to justify departure from the forum provided in the contract for contract breaches. In the words of Professor Crawford, such an approach would correspond with what he defines as the ‘fourth’ school of thought, pursuant to which the umbrella clause takes relevance in the context of the other treaty breaches involved. However, when the tribunal determined the merits, it changed its mind and followed what Professor Crawford calls the ‘third’ theory, according to which the umbrella clause per se transforms a contract breach into a treaty breach and the contract breach suffices to confer jurisdiction upon a BIT tribunal.

The other case involving Paraguay in an almost identical dispute was BIVAC v. Paraguay.22 This tribunal did not give relevance to the distinction between contract and treaty claims. It rather seemed to endorse the theory according to which the umbrella clause expands the jurisdiction of the treaty tribunal to include contract matters. The tribunal did indeed ‘conclude that Article 3(4) of the BIT [the umbrella clause] does have the effect for which BIVAC argues, namely that it gives the Tribunal jurisdiction over a claim that arise from or is produced directly in relation to the Contract’.23 It was even clearer when it specified that ‘In the present case, in relation to Article 3(4) we do not see how it could be concluded that the fundamental basis of the claim was the BIT rather than the contract. Any other approach strikes us as being so artificial as to be reasonable’24 and that ‘Article 3(4) of the BIT provides a basis for the exercise of jurisdiction by the tribunal over claims relating to the Contract’.25

By so reasoning, the BIVAC tribunal risked competing against the exclusive jurisdiction of the Paraguayan courts in relation to contract disputes, to a far greater extent than the SGS v. Paraguay tribunal. In any case, notwithstanding its unambiguous premises on the implications of the umbrella clause and its own jurisdiction, the tribunal eventually found that the claim was inadmissible on the ground that the relevant BIT had been concluded well before the contract and the contractual forum clause was therefore subsequent to the umbrella clause. It inferred that, when agreeing to refer contractual disputes to the Paraguayan courts, the parties were presumably aware that the BIT was in force, but agreed to waive any possible reliance on the BIT umbrella clause, thus conferring to the State courts exclusive jurisdiction.

I find it difficult to share the above reasoning. It is unrealistic to ‘presume’ that the parties be aware of the existence of a BIT and, in particular, of an umbrella clause when they execute a trade contract referring the determination of the resulting disputes to an agreed forum different from the treaty forum. As the above unsettled debate shows, the existence and significance of umbrella clauses is such a mysterious and controversial matter for specialists that it can hardly be assumed that even the most sophisticated parties might have it in mind when drafting a commercial contract. The tribunal had just one way available if it intended not to judge upon a dispute for mere non-payment of commercial invoices, that of dismissing the claim on the ground that exclusive reliance on an umbrella clause, with no further reliance on treaty protection, cannot form the basis of jurisdiction of a treaty tribunal over allegations of contract breaches which the parties had agreed to submit to State courts. Alternatively, to be consistent with its premise that the umbrella clause attributed to it jurisdiction over contract disputes, the tribunal should have accepted jurisdiction forthwith.

The above cases did not put an end to the discussion. In more recent cases, the relevance of an umbrella clause was again at issue, as follows:

(i) An ICSID tribunal found that a BIT provision invoked by the claimant as having the significance of an umbrella clause did not qualify as such, but wondered whether the MFN clause inserted in the BIT might be used to import a proper umbrella clause found in a third BIT concluded by the host State. The tribunal admitted this possibility, found the proper umbrella clause in another BIT, imported it into the basic BIT and retained jurisdiction.26

(ii) Another ICSID tribunal refused to rely on an umbrella clause since the contract from which the dispute arose was made between the investor and the University of Kiev, not the Republic of Ukraine. However, the tribunal deemed it necessary to explain that, even in the contrary case, it would have been bound to decline jurisdiction in the specific circumstances of the case, because the same contractual dispute had already been brought before and decided by the Kiev courts. Consequently, assuming that the umbrella clause would have been theoretically applicable, it could not have had the effect of subtracting the jurisdiction from the Ukrainian courts and shifting it to the BIT tribunal.27

(iii) Finally, in Burlington v. Ecuador, an ICSID tribunal held that the foreign shareholder of a totally owned local subsidiary was not entitled to rely on the BIT umbrella clause to enforce against the host State the rights of the subsidiary under the contract executed between the subsidiary and the State. In other words, in order for an umbrella clause to be applicable, the parties to the BIT arbitration should be the same parties to the contract.28 A similar decision had already been adopted by other ICSID tribunals in Azurix v. Argentina29 and Siemens v. Argentina30 and by the ad hoc annulment committee in CMS v. Argentina.31

As the above discussion outlines, this is a matter in which uncertainty of the law is still heavily prevalent.

4. OTHER EXAMPLES OF DIVERGENT INTERPRETATION

4.1 The ‘Necessity Defence’

Pursuant to one of the most common BIT provisions, the host State cannot be held liable for mistreatment of the investor under the BIT if it proves that it has acted under a state of necessity or emergency in protection of essential security interests and public order of the country. The principle is also known in customary international law as the ‘state of necessity defence’ or the ‘essential security interests exception’.32

Argentina raised this objection in practically all arbitral disputes triggered by its 2001/2002 emergency legislation, which proved seriously detrimental to several foreign investors. In all such first cases, the ICSID tribunals were constituted on the basis of the BIT in force between Argentina and the US, Article XI of which provided as follows:

“This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfilment of its obligations with respect to the maintenance or restoration of international peace or security, or the Protection of its own essential security interests”.

All tribunals set up on the basis of this BIT interpreted Article XI in the light of the principles of international customary law governing the necessity defence, as reflected in Articles 25 and 27 of the ILC Articles.33 The Articles in question entitle a State to invoke necessity to excuse its violations of international treaties only if it demonstrates that its conduct meets the following elements: (i) the wrongful act was the ‘only way’ to safeguard the State’s essential interest against a grave and imminent peril; (ii) the State did not itself contribute to the situation of necessity; and (iii) the State complied with its obligations as soon as necessity no longer existed.

In consideration of the factual circumstances of the disputes brought before them, all above tribunals held that Argentina was not entitled to rely on the necessity defence, because the disputed State measures were not the only available way to overcome the crisis and Argentina had itself contributed to the crisis.34

These awards received severe criticisms by ad hoc annulment Committees and some of them were actually annulled. The ad hoc Committee called upon to rule on Argentina’s application for annulment of the CMS award dismissed the application for annulment, however criticizing the award for an erroneous interpretation of Article XI:

“[…] the Tribunal gave an erroneous interpretation to Article XI […] the Committee cannot simply substitute its own view of the law and its own appreciation of the facts for those of the Tribunal […] it is the case in the end that the Tribunal applied Article XI of the Treaty. Although applying it cryptically and defectively, it applied it. There is accordingly no manifest excess of powers”.35

Differently from the CMS ad hoc Committee, which dissented with the merits of the decision but admitted that Article XI had been applied, although allegedly wrongly, the Sempra ad hoc Committee concluded that the tribunal had ‘failed to apply’ Article XI of the BIT by exclusively relying on ILC Article 25 and ignoring certain alleged ‘differences’ between the two provisions. It thus annulled the award for ‘manifest excess of powers’.36 In Enron, the ad hoc Committee blamed the tribunal for having endorsed a financial expert’s report indicating the existence of alternative ways for redressing Argentina’s crisis without further analysis. It thus annulled the award on the ground that the tribunal’s interpretation of the applicable law (Article XI of the BIT) would lack ‘sufficient reasoning’.37

I have already had the opportunity to disagree with the annulment decisions in Sempra and Enron.38 Those two awards were indeed wrongly annulled. The tribunals had not failed to apply the law, be it represented by Article XI of the BIT or ILC Articles 25 or 27. The treaty and customary law provisions were diligently examined both individually and through mutual comparison. Their implications and significance were addressed through a lengthy and well-articulated reasoning. The Committees’ finding for a manifest excess of powers has no justification and is unacceptable. In the reality, and this is impermissible under Article 52 of the ICSID Convention, the two Committees assumed the role of an appeal court substituting their views for those of the tribunals in questions of merits.

I still consider that these two Committees went far beyond the limits set in Articles 52 and 53 of the ICSID Convention, which expressly prohibit that an ICSID award be subject to appeal or to revision of the merits. They clearly confine the review of the awards to pure procedural legitimacy, with no intervention on the substance already decided.

Fortunately, the above annulments did not discourage subsequently appointed ICSID tribunals to continue to apply the proper law governing the States’ necessity exception or defence. One such case is, for instance, a recent ICSID award, which again applied ILC Article 25 to resolve a dispute based on the BIT in force between Argentina and France. It rejected Argentina’s defence, seeking justification for its actions on the basis of the necessity defence. Holding that Argentina had failed to substantiate the key elements required by Article 25 for the successful invocation of a necessity defence, the tribunal granted substantial recovery of claimant’s claims for expropriation.39

4.2 The Notion of ‘Protected Investment’

Other cases where investor-State tribunals have greatly dissented, concern determination of the scope of the protected investment, i.e. definition of the investment eligible for treaty protection.

On the one hand, a number of tribunals have adopted an objective or restrictive definition of investment, in line with the widely accepted ground rules fixed by the Salini v. Morocco tribunal for the definition of covered investment.40 That ruling is still considered the most prominent example of such a consistent approach. In defining the protected investment within the meaning of Article 25 of the ICSID Convention, the tribunal adopted what became later known as the Salini test. According to this test, an investment is internationally protected within the terms of the Convention and applicable BITs when it satisfies the following requirements: contributions in capital, a certain duration of performance of the contract, participation in the risks of the transaction and (somewhat contested41) contribution to the economic development of the host State. More recent cases tend to add some new criteria, especially the legality and bona fide nature of the investment, which must have been made ‘in accordance of the laws of the host State’.42 The Salini criteria have been endorsed by several investment tribunals.43

On the other hand, other tribunals seem to adopt a more liberal approach consisting in enlarging the investment definition so as to include ventures that may prima facie appear more questionable. For instance, a number of tribunals have accepted as protected investments loans,44 promissory notes,45 minority shareholdings,46 the possibility to access an international market,47 arbitral awards48 and, more recently and controversially, the purchase of security entitlements in government bonds.49

The ‘expansive’ definition of investment, adopted by the first BIT tribunals in the 1990s and early 2000s, was magisterially discussed by Professor Charles Leben in 2003.50 Several years lapsed and the state of jurisprudence is still unsettled also in this matter, although much less divided in comparison to the deep divide found in relation to the matters discussed herein above.

5. GENERAL REMARKS

Investment arbitrations are not conducted under the guidance of a permanent court. Every case is assigned to an individual tribunal. In the absence of a doctrine of binding precedent (stare decisis), arbitral interpretation exclusively binds the parties to the case at hand. The individual award is not intended to establish rules that reach beyond this fundamental limit. Tribunals have the tendency to cite previous awards and confront their views with those expressed in past decisions, but have also the tendency to (correctly) make it clear that they are not bound by, and free to diverge from, their predecessors.

This may well explain how the same legal issue receives different, and sometimes opposite, solutions.

The same circumstance may also help understand why certain tribunals, and especially certain annulment committees, have appeared particularly ‘activist’, in contrast with others that have instead shown to be more ‘restrained’ or disciplined. The activism displayed in certain renowned cases has been sometimes depicted as the portrayal of an ‘expansionary trend’ in international arbitration.51

There might probably be a certain degree of overstatement in similar expressions, but it is a fact that BITs are too often remarkably laconic and vague. Understandably, the tribunals’ tendency is filling in the treaty lacunae and the inevitable interpretative ‘law-making’ gives rise to a variety of results. The same is true for any form of jurisprudential expansive interpretation of the underlying law standards, for which the judge or arbitrator lacks unequivocal interpretive guidance.

It must indeed be underlined that this phenomenon is not an exclusive feature of investment arbitration. Other judicial bodies especially international courts are in place that often shape the law through a process of creative interpretation to cope with the circumstances prevailing in their respective domain. For sure, one such forum is the European Court of Justice. Its decisions have undoubtedly forged the UE law in all relevant community sectors, sometime beyond the basic treaty principles, which required to be implemented in unexpected social and economic scenarios.52

In conclusion, the uncertain and inconstant evolution of international investment law poses a serious risk of unpredictability in the outcome of dispute resolution, which is highly undesirable. However, it is my view that the process will inevitably continue and that we should accept the fact that we will have to live with this situation for years to come.

I see only two alternative remedies, one good, one bad.

The first remedy is to advise the contracting States to better draft their BITs, abandoning vague notions and defining more precisely what they expect investment arbitrators should decide in case of disputes. In the light of the many lessons imparted from the past cases, the States should now be able to provide the tribunals with unequivocal and exhaustive provisions.

The second (but unwelcome) alternative remedy, several times proposed and rejected, would be the creation of a permanent appellate body charged with the task to revise the substance of the arbitrators’ decisions and correct errors in law, if any, thus progressively creating a uniform law. However, this remedy is generally considered a cure that is worse than the disease.53 Apart from the fact that no guarantee exists that the appellate body would be more correct than the tribunals in the application of the law or fail-safe from any miscarriage of justice, the duration of the proceedings would unacceptably be prolonged: first arbitration, then appeal, then the annulment phase.

Therefore, I do not intend to recommend it.



1
The first such case was Maffezzini [an Argentinian national] v. Kingdom of Spain, ICSID Case No. ARB/97/7, Decision on Jurisdiction, 25 January 2000, followed by Siemens A.G. v. Argentine Republic, ICSID Case No. ARB/02/8, Decision on Jurisdiction, 3 August 2004; Camuzzi Int’l S.A. v. Argentine Republic, ICSID Case No. ARB/03/2, Decision on Jurisdiction, 11 May 2005; Gas Natural SDG, S.A. v. Argentine Republic, ICSID Case No. ARB/03/10, Decision on Jurisdiction, 17 June 2005; Suez, Sociedad General de Aguas de Barcelona S.A. and InterAguas Servicios Integrales del Agua S.A. v. Argentine Republic, ICSID Case No. ARB/03/17, Decision on Jurisdiction, 16 May 2006; National Grid plc v. Argentine Republic, UNCITRAL, Decision on Jurisdiction, 20 June 2006; Suez, Sociedad General de Aguas de Barcelona S.A. and Vivendi Universal S.A. v. Argentine Republic/AWG Group Ltd. v. Argentine Republic, ICSID Case No. ARB/03/19, Decision on Jurisdiction, 3 August 2006; Impregilo S.p.A. v. Argentine Republic, ICSID Case No. 07/17, Award, 21 June 2011; Hochtief AG v. Argentine Republic, ICSID Case No. ARB/07/31, Decision on Jurisdiction, 24 October 2011. A similar decision was rendered in a case where the applicable BIT was the BIT in force between the United Kingdom and Turkmenistan and the issue was related to the MFN clause inserted therein: see Garanti Koza LLP v. Turkmenistan, ICSID Case No. ARB/11/20, Decision on Jurisdiction, 3 July 2013.


2
See for instance, BG Group Plc. v. The Republic of Argentina, UNCITRAL, Final Award, 24 2007.


3
Salini Costruttori S.p.A and Italstrade S.p.A. v. The Hashemite Kingdom of Jordan, ICSID Case No. ARB/02/13, Decision on Jurisdiction, 9 November 2004; Plama Consortium Limited v. Republic of Bulgaria, ICSID Case No. ARB/03/24, Decision on Jurisdiction, 8 February 2005; Telnor Mobile Communications A.S. v. The Republic of Hungary, ICSID Case No. ARB/04/15, Award, 13 September 2006; Wintershall Aktiengesellschaft v. Argentine Republic, ICSID Case No. ARB/04/14, Award, 8 December 2008; ICS Inspection and Control Services Limited v. Argentine Republic, UNCITRAL, Decision on Jurisdiction, 10 February 2012; Daimler Financial Services AG v. Argentine Republic, ICSID Case No. ARB/05/1, Award, 22 August 2012.


4
See for instance, Daimler v. Argentina, §§ 179 ff., criticizing Maffezzini and its followers.


5
See the nine ICSID cases cited in footnote n. 1 and the UNICTRAL case cited in footnote n. 2.


6
See the four ICSID cases and the UNCITRAL case cited in footnote n. 3.


7
For an overview of the divided case-law, see J. A. Maupin, ‘MFN-based Jurisdiction in Investor-State Arbitration: Is There Any Hope for a Consistent Approach?’, in Journal of International Economic Law, 14(1), March 2011, pp. 157 ff. and W. Ben Hamida, ‘MFN and Procedural Rights: Solutions from WTO Experience?’ in Transnational Dispute Management (TDM) Journal, Vol. 8, Issue 3, September 2011, p. 231 ff.


8
BG . v. Argentina, §§ 147 ff.


9
D.C. District Court, 7 June 2010.


10
United States Court of Appeals for the District of Columbia Circuit, 17 January 2012.


11
Supreme Court of the United States, BG Group plc v. Republic of Argentina, Argued 2 December 2013-Decided 5 March 2014.


12
Plama v. Bulgaria, see references in footnote n. 3 above. The case was commented, among others, by A.R. Parra, The History of ICSID, Oxford University Press, 2012, p. 294 ff.


13
J. Crawford, ‘Treaty and Contract in Investment Arbitration’, in Arbitration International, Vol. 24, Issue 3, pp. 351-374.


14
According to Prof. Crawford ‘Around 40 per cent of BITs contain some version of an umbrella clause’: see J. Crawford, cit., p. 366.


15
See A. Sinclair, ‘The Origins of the Umbrella Clause in the International Law of Investment Protection’, in Arbitration International, Vol. 20, n. 4; A. Gardina, ‘Les aspects juridiques du recours à l’arbitrage par un investisseur contre les autorités de l’Etat hôte en vertu d’un traité interétatique’, Report to the 18th Committee of the Institut de Droit International, Rhodes Session, 2011, pp. 5-16; I. Fadlallah, ‘La distinction “Treaty Claims — Contract Claims” et la compétence de l’arbitre (CIRDI: faisons-nous fausse route?)’, in Les Cahiers de l’arbitrage, Vol. III, Gazette du Palais, July 2006, p. 126 ff.; I. Fadlallah and P. Mayer, ‘Contract claims et clauses juridictionnelles des traités relatifs à la protection des investissements’, in Journal du Droit International, 2009, pp. 71-96.


16
SGS Société Générale de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID Case No. ARB/01/13, Decision on Objections to Jurisdiction, 6 August 2003.


17
SGS Société Générale de Surveillance S.A. v. Republic of the Philippines ICSID Case No. ARB/02/6, Decision on Objections to Jurisdiction, 29 January 2004.


18
Salini Costruttori S.p.A. and Italstrade S.p.A. v. The Hashemite Kingdom of Jordan, ICSID Case No. ARB/02/13, Decision on Jurisdiction, 9 November 2004, §§ 126 to 127


19
J. Crawford, cit., pp. 366 to 369.


20
SGS Société Générale de Surveillance S.A. v. The Republic of Paraguay, ICSID Case No. ARB/07/29, Decision on Jurisdiction, 12 February 2010.


21
SGS Société Générale de Surveillance S.A. v. The Republic of Paraguay, ICSID Case No. ARB/07/29, Award, 10 February 2012.


22
Bureau Veritas, Inspection, Valuation, Assessment and Control, BIVAC B.V. v. The Republic of Paraguay, ICSID Case No. ARB/07/9, Decision on Jurisdiction, 29 May 2009.


23
Ibidem, § 142.


24
Ibidem, § 142.


25
Ibidem, § 159.


26
Mr. Franck Charles Arif v. Republic of Moldova, ICSID Case No. ARB/11/23, Award, 8 April 2013.


27
Bosh International, Inc and B&P Ltd Foreign Investments Enterprise v. Ukraine, ICSID Case No. ARB/08/11, Award, 25 October 2012.


28
Burlington Resources Inc. v. Republic of Ecuador, (formerly Burlington Resources Inc. and others v. Republic of Ecuador and Empresa Estatal Petróleos del Ecuador (PetroEcuador)), ICSID Case No. ARB/08/5, Decision on Liability, 14 December 2012.


29
Azurix Corp. v. The Argentine Republic, ICSID Case No. ARB/01/12, Award, 14 July 2006, § 384.


30
Siemens A.G. v. The Argentine Republic, ICSID Case No. ARB/02/8, Award, 17 January 2007, § 204.


31
CMS Gas Transmission Company v. The Republic of Argentina, ICSID Case No. ARB/01/8, Decision of the ad hoc Committee on the Application for Annulment of the Argentine Republic, 25 September 2007, §§ 97-98.


32
See C. Titi, ‘The Arbitrator as Lawmaker: Jurisgenerative Processes in Investment Arbitration’, in The Journal of World Investment & Trade, Vol. 14, No. 5, 2013, p. 833.


33
Draft Articles of the International Law Commission on Responsibility of States for Internationally Wrongful Acts.


34
The cases were the following: CMS Gas Transmission Company v. The Republic of Argentina, ICSID Case No. ARB/01/8, Award, 12 May 2005; Sempra Energy International v. The Argentine Republic, ICSID Case No. ARB/02/16, Award, 28 September 2007; Enron Corporation and Ponderosa Assets, L.P. v. Argentine Republic (also known as: Enron Creditors Recovery Corp. and Ponderosa Assets, L.P. v. The Argentine Republic), ICSID Case No. ARB/01/3, Award, 22 May 2007; El Paso Energy International Company v. The Argentine Republic, ICSID Case No. ARB/03/15, Award, 31 October 2011.


35
CMS Gas Transmission Company v. The Republic of Argentina, ICSID Case No. ARB/01/8, Decision of the ad hoc Committee on the Application for Annulment of the Argentine Republic, 25 September 2007, §§ 135-136.


36
Sempra Energy International v. The Argentine Republic, ICSID Case No. ARB/02/16, Decision on the Argentine Republic’s Application for Annulment of the Award, 29 June 2010.


37
Enron Creditors Recovery Corp. and Ponderosa Assets, L.P. v. The Argentine Republic, ICSID Case No. ARB/01/3, Decision on the Application for Annulment of the Argentine Republic, 30 July 2010.


38
A. Crivellaro, ‘Annulment of ICSID Awards: Back to the “First Generation”?’, in Liber Amicorum en l’honneur de Serge Lazareff, Editions Pedone, 2011, p. 169 ff.; A. Crivellaro, ‘The Failure to State Reasons in ICSID Awards’, in The Paris Journal of International Arbitration, 2012, 4, pp. 865ff.


39
EDF International S.A., SAUR International S.A and Leòn Partcipaciones Argentinas S.A. v. Argentine Republic, ICSID Case No. ARB/03/23, Award, 23 July 2012.


40
Salini Costruttori S.p.A. and Italstrade S.p.A. v. Kingdom of Morocco, ICSID Case No. ARB/00/4, Decision on Jurisdiction, 31 July 2001.


41
For instance, in Saba Fakes v. Turkey, the tribunal did not consider the contribution to development as part of the investment definition within the meaning of the ICSID Convention: Saba Fakes v. Republic of Turkey, ICSID Case No. ARB/07/20, Award, 14 July 2010.


42
Phoenix Action Ltd. v. The Czech Republic, ICSID Case No. ARB/06/5, Award, 15 April 2009.


43
Jan de Nul N.V. and Dredging International N.V. v. Arab Republic of Egypt, ICSID Case No. ARB/04/13, Decision on Jurisdiction, 16 June 2006; Joy Mining Machinery Limited v. Arab Republic of Egypt, ICSID Case No. ARB/03/11, Award on Jurisdiction, 6 August 2004; Malaysian Historical Salvors, SDN, BHD v. The Government of Malaysia, ICSID Case No. ARB/05/10, Award on Jurisdiction, 17 May 2007; Helnan International Hotels A/S v. Arab Republic of Egypt, ICSID Case No. ARB/05/19, Decision on Objection to Jurisdiction, 17 October 2006; Biwater Gauff (Tanzania) Ltd. v. United Republic of Tanzania, ICSID Case No. ARB/05/22, Award, 24 July 2008; Electrabel S.A. v. Republic of Hungary, ICSID Case No. ARB/07/19, Decision on Jurisdiction, 30 November 2012; Deutsche Bank AG v. Democratic Socialist Republic of Sri Lanka, ICSID Case No. ARB/09/2, Award, 31 October 2012; Quiborax S.A., Non Metallic Minerals S.A. and Allan Fosk Kaplún v. Plurinational State of Bolivia, ICSID Case No. ARB/06/2, Decision on Jurisdiction, 27 September 2012.


44
Ceskoslovenska Obchodni Banka, A.S. v. The Slovak Republic, ICSID Case No. ARB/97/4, Decisions on Objections to Jurisdiction, 24 May 1999.


45
Fedax N.V. v. The Republic of Venezuela, ICSID Case No. ARB/96/3, Decisions on Objections to Jurisdiction, 11 July 1997.


46
Lanco International Inc. v. The Argentine Republic, ICSID Case No. ARB/97/6, Preliminary Decision on Jurisdiction, 8 December 1998; Continental Casualty Company v. The Argentine Republic, ICSID Case No. ARB/03/9, Decision on Jurisdiction, 22 February 2006.


47
Pope & Talbot Inc. v. The Government of Canada, NAFTA Arbitration, Interim Award, 26 June 2000.


48
Saipem S.p.A. v. The People's Republic of Bangladesh, ICSID Case No. ARB/05/07, Decision on Jurisdiction, 21 March 2007.


49
Abaclat and Others v. Argentine Republic (formerly Giovanna a Beccara and Others v. The Argentine Republic), ICSID Case No. ARB/07/5, Decision on Jurisdiction and Admissibility, 4 August 2011.


50
C. Leben, ‘La théorie du contrat d’Etat et l’évolution du droit international des investissements’, in Recueil des Cours de l’Académie de Droit International, Vol. 302, 2003, pp. 245ff.


51
M. Sornarajah, ‘A Coming Crisis: Expansionary Trends in Investment Treaty Arbitration’, in , K.P. Sauvant and M. Chiswick Patterson, eds., Appeals Mechanism in International Investment Disputes, New York, 2008; A. Reinisch, ‘The Proliferation of International Dispute Settlement Mechanisms: The Threat of Fragmentation vs. the Promise of a More Effective System? Some Reflections from the Perspective of Investment Arbitration’, in I. Buffard, J. Crawford, A. Pellet, eds., International Law Between Universalism and Fragmentation-Festchrift in Honour of Gerhard Hafner, Leiden-Boston, 2008.


52
H.W. Micklitz, Judicial Activism of the European Court of Justice and the Development of the European Social Mode in Anti-Discrimination and Consumer Law, European University Institute Working Paper, Law 2009/19; J. Bengoetxea, N. MacCormick, L.M. Soriano, ‘Integration and Integrity in the Legal Reasoning of the European Court of Justice’, in G. de Burca & J.H.H. Weller, eds., The European Court of Justice, Oxford University Press, 2001.


53
C. Tams, ‘Is There a Need for an ICSID Appellate Structure?’, in R. HOfmann & C. Tams, eds., The International Convention on the Settlement of Investment Disputes (ICSID): Taking Stock After 40 Years, 2007; A. H. Qureshi, ‘An Appellate System in International Investment Arbitration?’, in P. Muchlinski, F. Ortino, C. Schreuer, eds., The Oxford Handbook of International Investment Law, 2008; F. Ortino, A. Sheppard, H. Warner, eds., Investment Treaty Law, Current Issues, Volume 1, Appeals and Challenges to Investment Treaty Awards: Is it Time For an International Appellate System?, 2006, with contributions by V.V. Veeder, D. Bischop, J. Jill, N. Blackaby and C.I. Suarez Anzorena.