Full Contents List

• Introduction
  1. Preface
  2. Foreword
  3. Introduction
• Regional Overviews
  1. Recent Developments in Investment Treaty Arbitration
  2. Challenges to Arbitrators
  3. State of Necessity in Investment Arbitration
  4. Arbitration of Intellectual Property Disputes
  5. Provisional Application of the Energy Charter Treaty
• Country Overviews
  1. Austria
  2. Czech Republic
  3. Denmark
  4. Finland
  5. France
  6. Greece
  7. Ireland
  8. Netherlands
  9. Poland
  10. Portugal
  11. Spain
  12. Ukraine
  13. United Kingdom
  14. Middle East Overview
  15. Lebanon

Page Options

print this chapter

Purchase This Special Report

buy online

Related Special Reports

buy online

buy online

Introduction

The term ‘state of necessity’ under international law describes a situation in the presence of which a state is excused from performing an international obligation. Such situation is generally believed to be an actual threat or a prospective peril to a state’s essential interest. From an operative point of view, a state of necessity has the ability to change to a legitimate action a conduct that would otherwise be considered wrongful.
The ‘state of necessity’ plea has been known for many years. However, it has recently come under detailed scrutiny in the area of foreign investment protection. The crisis endured by Argentina between 2001 and 2002 and the measures adopted by the Argentine government to tackle such a severe crisis have given rise to the compelling need both to reassess the actual characteristics of such defence under customary international law and to analyse the relationship of such defence with similar provisions contained in bilateral investment treaties (BITs).
Two well-known ICSID arbitrations that stemmed from the application of the 1991 US-Argentina BIT have approached the issue from two rather different angles and have therefore sparked a debate as to how such defence should be entertained by arbitral tribunals in investment arbitration.
This paper will briefly illustrate how those two cases have treated some of the central issues relating to the defence of ‘state of necessity’ in investment arbitration.

‘State of necessity’ defence in customary international law

The ‘state of necessity’ defence has been pleaded in international disputes on a number of occasions. In most cases the existence of such defence was not challenged.
An oft-quoted example is the dispute between the British and Portuguese governments in 1832. The Portuguese government had a pressing need to provide for the subsistence of some of its troops fighting in Portuguese territory. Therefore, despite the existence of a treaty obligation to the contrary, the government confiscated property owned by British subjects. Given the circumstances, it was recognised in the legal opinion sought by the British government that the treaty entered into by the two countries was not:
of so stubborn and unbending a nature, as to be incapable of modification under any circumstances whatever, or that their stipulations ought to be so strictly adhered to, as to deprive the Government of Portugal of the right of using those means, which may be absolutely and indispensably necessary.¹

More recently, in 1967, the Liberian Torrey Canyon tanker went aground off the coast of Cornwall outside British waters, spilling large amounts of oil that put the English coastline under an immediate threat of pollution. The British government bombed the ship to burn the remaining oil. While the government did not formalise any legal justification for its conduct, it claimed that the operation had been decided because of the situation of extreme danger that had not been possible to successfully tackle by alternative means. The conduct of the British government was considered legitimate and no international protest followed.²

Codification of customary international law by the ILC

The substantial authority in support of the existence of a doctrine of ‘state of necessity’ led to the inclusion of such defence in a well-known attempt to codify the customary international law relating to state responsibility. This was carried out by the International Law Commission in the so-called Draft Articles on Responsibility of States for Internationally Wrongful Acts (ILC Articles).³
In the rather narrow definition provided under ILC Article 25, a ‘state of necessity’ may not be invoked by a state as a ground for precluding the wrongfulness of an act not in conformity with an international obligation unless the act (i) is the only way for the state to safeguard an essential interest against a grave and imminent peril; and (ii) does not seriously impair an essential interest of the state towards which the obligation exists. In any case, under ILC Article 25, ‘necessity’ may not be invoked by a state as a ground for precluding wrongfulness if the international obligation in question precludes the possibility of invoking necessity, or the state has contributed to the situation of necessity.
Whether an actual ‘state of necessity’ defence exists under customary international law and whether such rule has been correctly codified in the ILC Articles is not unanimously accepted. Some doubts in this respect were expressed in the famous Rainbow Warrior arbitration, where the arbitral tribunal described the ILC Articles as controversial.⁴ Despite some isolated claims to this extent, it is generally believed that such defence does indeed exist under customary international law and it does so in the form codified under ILC Article 25. The International Court of Justice supported this argument in Gabcíkovo-Nagymaros Project.⁵ In that case, the parties in dispute were in agreement upon the existence of a ‘state of necessity’ defence as well as upon the fact that the defence should be evaluated in the light of the criteria laid down by the ILC Articles. The ICJ endorsed such view noting that the ‘state of necessity’ is indeed a ground recognised by customary international law for precluding the wrongfulness of an act not in conformity with an international obligation and that such defence has been codified in the ILC Articles.⁶

‘State of necessity’ in foreign investment arbitration

The ‘state of necessity’ defence has been the object of controversial interpretation in several cases brought against Argentina following the crisis that the South American state endured in the early 2000s. Of particular interest are the two conflicting decisions rendered in the cases of CMS v Argentina⁷ and LG&E v Argentina.⁸ Both cases arose in connection with the 1991 US-Argentina BIT.

Factual background

The factual background of the two cases is similar. Some foreign investors had participated in a privatisation program of the Argentine gas distribution industry. This was mainly done by foreign investors through the purchase of shares of local companies following a public bidding process. The local companies were granted long-term licences with the right to calculate gas tariffs in US dollars and to convert them into pesos at the prevailing exchange rate. The implemented tariff regime also provided the right to have such tariffs adjusted every six months in accordance with the so-called United States Producer Price Index (US PPI). It is important to remember that at that time, in a more general effort to create stability for the local currency, Argentina had also pegged the value of the Argentine peso to the value of the US dollar.
In the late 1990s, when the Argentine economy had already started to show signs of recession, it was decided that any increase in gas prices would be detrimental to the national economy and social peace in Argentina. During that time the financial and social situation worsened to a worrying level. Riots and unrest took place in the streets of the major Argentine cities where the population were denied access to their bank accounts. The political structure of the country very nearly collapsed, with several presidents of the Republic elected and removed in the space of few weeks. As a result, and in order to tackle the situation, a number of measures affecting the value of the foreign investments were adopted (the so-called ‘pesification’ process).
The action taken by the Argentine government gave rise to a considerable number of ICSID claims under the several BITs that had been entered into by Argentina. One of the defences raised by Argentina in those arbitrations was that Argentina could not be held liable for its conduct as it had acted under a ‘state of necessity’. Argentina in raising such defence was mainly relying on article XI of the US-Argentina BIT, according to which:
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.

Conflicting decisions

The arbitral tribunal in CMS, amongst other things, rejected in toto the ‘state of necessity’ defence that had been raised by Argentina. The decision and the method of analysis advocated by the CMS tribunal were followed in the decision taken in September 2007 by the ICSID tribunal in Sempra v Argentina.⁹ The award rendered in CMS recently survived almost unscathed - even though it did not escape a remarkable amount of criticism - an application for annulment under article 52 of the ICSID Convention due to the limited powers allowed by the ICSID Convention to annulment committees.¹⁰
In LG&E the tribunal found, contrary to the CMS award, that Argentina was indeed in a ‘state of necessity’ that excused non performance of the BIT, even though according to the tribunal such ‘state of necessity’ only lasted for 18 months.
Two of the very many issues debated in those arbitrations were (i) the meaning and scope to be given to a specific provision dealing with ‘state of necessity’ that was contained in the BIT as well as the relationship between this provision and the above-mentioned ILC Article 25 and (ii) the degree of analysis that arbitral tribunals should undertake in deciding whether a defence of ‘state of necessity’ has been legitimately called by a defendant state. This two issues will be briefly analysed in turn.
The CMS tribunal primarily considered ‘necessity’ under customary international law, as expressed in ILC Article 25. Even though the Tribunal also analysed article XI of the BIT, it is clear that the tribunal considered ILC Article 25 as the predominant source of law. Indeed the ‘state of necessity’ defence raised by Argentina was rejected for the very reason that it did not comply with the requirements imposed by such rule of customary international law.
Unlike the CMS tribunal, the tribunal in LG&E started its assessment of Argentina’s ‘state of necessity’ defence by applying the BIT’s emergency clause. It also, however, analysed the concept of ‘necessity’ under customary international law, though only as a subsidiary point in supporting its conclusion.

Problematic issues

It is possible to detect in the above-mentioned awards the inclination to consider the rules of customary international law and international treaty law as interchangeable or, in any case, to apply customary international law as a subsidiary source of interpretation of treaty law.
The approach taken by the two tribunals is perhaps not entirely satisfactory. It is widely accepted that in accordance with the general rule of law lex specialis derogat generali, the provisions contained in an international treaty should take precedence over the application of provisions of a more general nature such as customary international law. This is not just an issue of formalistic precedence of one source of law over another. The problem becomes clearer through a brief comparison between the scope and nature of the provisions contained in BITs and the provisions of customary international law.
As is well known, the provisions of customary international law, by definition, are necessarily formed over a long period of time and need to be recognised by the majority of the community that they are intended to regulate in order for such rules to be considered as legally binding (the so-called opinio juris). The situation is diametrically different with BITs as they serve a much narrower scope and are intended to last for a comparatively short period of time. BITs are entered into between two states and serve a very narrow, sometime regional, scope. They apply to very limited categories of entities (ie, investors) and are aimed at providing protection to very limited subject matters (ie, investments as defined in the relevant BIT).
Because of the exceptional nature of BITs, and therefore their limited scope of application, it is not unlikely that states may be prepared to undertake, by signing BITs, obligations which might be regarded by the same signatory parties as diverging from the existing rules of customary international law. In other words, and more specifically to our discussion, sovereign states may be inclined, in order to serve a specific and time-limited purpose, to regulate the ‘state of necessity’ defence in a way which may be completely inconsistent with the rules of customary international law.
It is for this reasons that whether customary international law rules and international treaty rules are actually interchangeable should be ascertained carefully. Furthermore, it should also be carefully assessed whether it is appropriate, given the differences in terms of scope and background that might exist between the ‘state of necessity’ defence under customary international law and the same defence under BITs, to use such different sources of law as a supplementary source of interpretation. To avoid any risk of departing from the intention of the signatory parties, it would therefore appear rather more appropriate if BIT provisions, such as those dealing with ‘state of necessity’, were interpreted independently from other more general sources of law. This would seem particularly appropriate where there is a clear indication that the two sets of provisions may have different background and scope of application. Perhaps, in these circumstances, it would be more appropriate to use means of interpretation that are available under international law but that, for some reasons, are not frequently used.
As is well known, interpretation of international treaties is to be carried out in compliance with the provisions laid down by the Vienna Convention 1969 on the Law of Treaties. Most often treaties are interpreted by using the provisions contained under article 31 of the Vienna Convention. Very little use is usually made of the supplementary means of interpretation provided under article 32 of the same Convention, according to which recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31, either leaves the meaning ambiguous or obscure, or leads to a result that is manifestly absurd or unreasonable.
The supplementary means of interpretation could also help ascertain other aspects of the ‘state of necessity’ defence which have proved particularly problematic. Two of this issues are whether the defence of ‘state of necessity’ is self-judged and which party should bear the financial burden of a situation of ‘necessity’.
As regards the first issue, both the CMS tribunal and the LG&E tribunal concurred that a ‘state of necessity’ is not self-judged. The conclusion adopted by the two tribunals seems to have been generally welcomed as correct since it would be unfair if states were left with the unqualified right to escape the performance of their obligations under international law by simply claiming to be in a ‘state of necessity’. However, both the CMS and the LG&E tribunals dealt with the issue with some uneasiness, as if mindful that sovereign states should be allowed a degree of autonomy in the implementation of their policies and that such autonomy should not be subjected to the scrutiny of arbitral tribunals. The LG&E award, for example - which for several reasons seems to have adopted a more convincing approach to the issue of ‘state of necessity’ - attempted to balance deference for the activity of the sovereign state and protection of the investor’s rights in a rather unconvincing fashion. The LG&E tribunal, on the one hand, declared that ‘state of necessity’ is not self-judged but on the other hand shifted the burden of proof of its existence from the state to the investor. While the decision in LG&E seems to deserve support in that it leaves more leeway to states facing a severe crisis, the idea of protecting state sovereignty by altering the general rules on burden of proof is understandably regarded as problematic.¹¹
A further issue that has emerged in the two above-mentioned decisions is the issue of the consequences of a ‘state of necessity’. Both the CMS and LG&E tribunals found that a ‘state of necessity’ only justifies temporary suspension of a state’s obligations under international law. They differed, however, on the issue of which party should bear the financial consequences of the ‘state of necessity’. The CMS tribunal, again, privileged the application of the ILC Articles. It made recourse more precisely to Article 27(b) according to which it is for the host state to bear the consequences of ‘necessity’ and therefore to compensate the investor.
The tribunal in LG&E rejected the application of the ILC Articles and consequently required the investor to bear the losses that had been suffered during the ‘state of necessity’. The LG&E tribunal seems to have reached this conclusion by observing that article XI of the BIT, unlike ILC Article 27, does not deal with the issue of payment of compensation. Therefore, the state could not be ordered to bear the financial consequences of the ‘state of necessity’. Even though such solution has been criticised,¹² it seems that the position taken by the LG&E tribunal is at least tenable. If indeed, as both the CMS and LG&E tribunals seem to have agreed, a ‘state of necessity’ has the effect of suspending the obligations undertaken by the state in a BIT, then a state cannot be held liable for a breach of an obligation that, because of that suspension, was not operative. Obviously the consequences of a suspension should work differently depending on the nature of the suspended obligation. A ‘state of necessity’ could not, for example, remove an obligation to pay a sum. It seems, however, that by applying the view advocated in the LG&E award, the payment of interest, as well as any further consequences for the delayed performance of an obligation to pay, such as penalties, would not accrue during the period of ‘state of necessity.’ Similarly, if the obligations under the BIT are suspended, a state should not be held liable if during the ‘state of necessity’ it fails to provide the protection guaranteed under the same BIT such as, for example, the obligation to provide full protection and security or fair and equitable treatment.

* * *

The diverging decisions issued with regard to ‘state of necessity’ in the above-mentioned investment arbitrations do not constitute an absolute novelty in this area of practice. Inconsistency of decisions in investment arbitration has been the subject of debate and interesting reform proposals. Perhaps it is not appropriate to expect consistency from a system of dispute resolution that was engineered to provide ad-hoc decisions. However, even though full consistency cannot be expected from the investment dispute resolution system as it is structured now, it is also true that some kind of uniformity, at least in the methodology of analysis, should be encouraged as it is perhaps rather undesirable that inconsistency should be detected even in the identification of the legal norms to be used as well as their priority of application.

Notes

1 A D McNair (ed), International Law Opinions (Cambridge University Press, 1956), vol II, p232.
2 The Torrey Canyon, Cmnd 3246 (London, 1967).
3 Draft Articles on Responsibility of States for Internationally Wrongful Acts. Text adopted by the ILC at its 53rd session, in 2001, Yearbook of the International Law Commission (2001), vol II, part two.
4 Rainbow Warrior (New Zealand/France), UNRIAA, vol XX (1990), p254.
5 Gabcíkovo-Nagymaros Project (Hungary/Slovakia), judgment of 25 September 1997 at p36, available at www.icj-cij.org.
6 Draft Articles on Responsibility of States for Internationally Wrongful Acts with commentaries by Rapporteur Prof. J. Crawford, 2001. Available at http://untreaty.un.org/ilc/ilcintro.htm. The actual scope of the defence has been expressly defined in the commentaries attached to the ILC Articles, where it is explained that state of necessity is different from the other defences available under customary international law. In the formulation codified in the ILC Articles, state of necessity, unlike consent, self-defence or countermeasures, is not dependent on the prior conduct of the injured party, while, unlike force majeure, it does not involve conduct that is involuntary or coerced. State of necessity should also be distinguished from distress, as necessity consists not in danger to the lives of individuals in the charge of a state official but in a grave danger either to the essential interests of the state or of the international community as a whole. It arises where there is an irreconcilable conflict between an essential interest on the one hand and an obligation of the State invoking necessity on the other.
7 CMS Gas Transmission Company v Argentine Republic, ICSID case no. ARB/01/8, Award of 12 May 2005; 44 ILM 1205 (2005).
8 LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentine Republic, ICSID case no. ARB/02/1, Decision on Liability of 3 October 2006, available at www.worldbank.org/icsid/index.html.
9 Sempra Energy International v Argentina, ICSID case no ARB/02/16. Available at www.investmentclaims.com. It is worth noting that two of the arbitrators sitting in the Sempra tribunal had also participated as arbitrators in CMS v Argentina.
10 CMS Gas v Argentine Republic, Decision of ad hoc Committee on Application for Annulment of the Argentine Republic, 25 September 2007, available at www.worldbank.org/icsid/index.html.
11 See S Schill, ‘International Investment Law and the Host State’s Power to Handle Economic Crises: Comment on the ICSID Decision in LG&E v Argentina’, Journal of International Arbitration, 24.3 (2007), pp265–286.
12 S Schill, p286.

Chiomenti Studio Legale

Next Chapter: Arbitration of Intellectual Property Disputes