Ila Committee on International Commercial Arbitration, Final Report on ‘Lis Pendens and Arbitration’(Toronto, 2006)

__ , Recommendation 3(40) (Toronto, 2006)

Leboulanger, P ‘Multicontract Arbitration’, 13(4) Int'l Arb 43 (1996)

__ , ‘Res Judicata and the Rule of Law in International Arbitration’, 8 African J Int'l & Comp L 38 (1996)

Lowe, V, ‘Overlapping Jurisdiction in International Tribunals’, 20 Australian YB of Int'l L 191 (1999)

Mantilla Serrano, F, ‘La Nouvelle L?gislation colombienne sur l'arbitrage’, 1 Revue de l'Arbitrage 41 (1992)

Mayer, P, ‘Litispendance, connexit? et Chose jug?e dans l'arbitrage international’, in J-D. Bredin, P Lalive JF Poudret, and F Terre (eds), Liber Amicorum Claude Reymond (Paris, Litec, 2004)

Obadia, E, ‘ICSID, Investment Treaties and Arbitration: Current and Emerging Issues’, in Investment Treaties and Arbitration, ASA Swiss Arbitration Association Conference (25 January 2002)

Platte, M, ‘When Should an Arbitrator Join Cases?’, 18(1) Arb Int'l 67 (2002)

Reichert, D, ‘Problems with Parallel and Duplicate Proceedings: The Litispendence Principle and International Arbitration’, 8(3) Arb Int'l 237 (1992)

Reinisch, A, ‘The Use and Limits of Res Judicata and Lis Pendens as Procedural Tools to Avoid Conflicting Dispute Settlement Outcomes’, 3(1) The Law and Practice of Int'l Courts and Tribunals (2004)

__ , The ICSID Convention: A Commentary (Cambridge, Cambridge University Press, 2001)

__ , ‘Travelling the BIT Route: Of Waiting Periods, Umbrella Clauses and Forks in the Road’, 5(2) JWI 231 (2004)

Schreuer, C, ‘Calvo's Grandchildren: The Return of Local Remedies in Investment Arbitration’, 4 The Law and Practice of Int'l Courts and Tribunals (2005)

Shany, Y, The Competent Jurisdictions of International Courts and Tribunals (Oxford, Oxford University Press, 2003)

Sinclair, AC, ‘The Origins of the Umbrella Clause in the International Law of Investment Protection’, 20(4) Arb Int'l (2004)

Soderlund, C, ‘Multiple Judicial Proceedings and the Energy Charter Treaty’, in C Ribeiro (ed), Investment Arbitration and the Energy Charter Treaty (Huntington, NY, JurisNet, LLC, 2006)

Turner. P, ‘The “Fork in the Road” Revisited’, in F Ortino, A Sheppard, and H Warner (eds), Investment Treaty Law, Current Issues, Vol 1 (London, BIICL, 2006)

W?lde, T, ‘The Umbrella Clause in Investment Arbitration—A Comment on Original Intentions and Recent Cases’, 6 (2) JWIT (2005)

Yannaca-Small, K, ‘Consolidation of Claims: A Promising Avenue for Investment Arbitration?’, in OECD, International Investment Perspectives (Paris, OECD, 2006)

__ , ‘Interpretation of the Umbrella Clause in Investment Agreements’, OECD Working Paper (2006) Footnotes ?I would like to offer my sincere appreciation and thanks to my colleagues in the ILA Committee on Foreign Investment, Thomas W?lde and Jackie van Haersolte-van Hof, for their very valuable comments on this chapter. 1 Mark Friedman in his presentation on parallel proceedings at the 2006 ICCA Conference in Montreal stated that he was able to locate and review awards in 78 ICSID cases and found evidence of parallel proceedings in 41% of them: see M Friedman, ‘Related Dispute Resolution Regimes: Parallel Proceedings in BIT Arbitration’, Draft, 31 May 2006 (on file with the author). 2 The risk of parallel proceedings could also arise in the case of multi-party consortia which may include not only project sponsors but also financing institutions and possibly other participants. 3 Mr Lauder had made an investment in the Czech Republic through a Dutch intermediary holding company that he controlled, CME Czech Republic BV. It was alleged that the investment had been expropriated. Two arbitrations were commenced: one under the US-Czech Republic BIT by Mr Lauder; the other under the Netherlands-Czech Republic BIT by CME CR BV. For the purposes of the expropriation claims, the facts were the same in each arbitration. The Lauder tribunal issued its Final Award on 3 September 2001 finding the Czech Republic not liable. The CME tribunal issued a Partial Award on Liability on 13 September 2001 finding against the Czech Republic and holding it liable to pay damages. In the second (quantum) phase of the CME arbitration, the Czech Republic argued that the Lauder Final Award had a res judicata effect on questions of liability. 4 See Lauder v The Czech Republic , Final Award (3 September 2001) at para 172. 5 The CME tribunal likewise rejected the Czech Republic's argument that Mr Lauder had impermissibly ‘treaty shopped’: ‘The argument of abusive treaty shopping is not convincing. A party may seek its legal protection under any scheme provided by the laws of the host country. The Treaty, as well as the US Treaty, is part of the laws of the Czech Republic and neither of the treaties supersedes the other. Any overlapping of the results of parallel process must be dealt with on the level of loss and quantum but not on the level of breach of treaty’. CME Czech Republic BV v The Czech Republic , Final Award (14 March 2003) at para 419. 6 .CMS Gas Transmission Company v The Republic of Argentina ICSID case No. ARB/01/8, Decision on Objections to Jurisdiction, in 42 ILM 788 (2003), available at <http://www.asil.org/ilib/cmsargentina.pdf>. The CMS Gas Transition Company (‘CMS’) purchased shares of an Argentine company, Transportadora de Gas del Norte (‘TGN’), pursuant to Argentina's privatization programme in 1995. Argentina argued that CMS lacked standing to file its claim because it was merely a minority non-controlling shareholder and thus did not have standing to claim damages suffered by TGN. The tribunal ruled that the Convention did not require control over a locally incorporated company in order to qualify under the Convention. It also ruled that the Convention does not bar a claim brought by a minority non-controlling shareholder such as CMS, observing that previous ICSID tribunals in also finding jurisdiction had ‘not been concerned with the question of majority [ownership] or control but rather whether shareholders can claim independently from the corporate entity’. The tribunal answered this question in the affirmative. CMS , ibid at para. 55. 7 Other tribunals examining cases brought against Argentina came to the same conclusion. See Lanco Int'l Inc v Argentina Republic , Preliminary decision on jurisdiction, 40 ILM 457, 463 (2001). 8 This may occur when in addition to investor-state arbitration, inter-state dispute settlement is possible before the ICJ, for instance, after an espousal of the investor's claim by the home state. An attempt at such an espousal in addition to the investor-state arbitration was rejected by the tribunal in the case Banro American Resources Inc v Democratic Republic of Congo . In this case, the Canadian parent company—having the nationality of a non-contracting state and consequently no standing before an ICSID tribunal after the dispute had arisen—transferred the investment to the US subsidiary having such standing, a measure which made it possible for the Banro Group to benefit from both diplomatic protection and ICSID arbitration. The tribunal found that there were questions of international public policy involved and declined jurisdiction to hear the request of the US subsidiary. It stated that it ‘cannot allow the requirements of nationality imposed by the Washington Convention to be neutralised by investors who are seeking to avail themselves, depending on their own interests at a given point in time, simultaneously or successively, of both diplomatic protection and ICSID arbitration, by playing on the fact that one of the companies in the group does not have the nationality of a Contracting State party to the Convention, and can therefore benefit from diplomatic protection by its home State, while another subsidiary of the group possesses the nationality of a Contracting State to the Convention and therefore has standing before an ICSID tribunal’. Banro v Congo, ICSID Case No. ARB/98/7, Award on jurisdiction, 1 September 2000, at para 25. 9 On this topic see the detailed analysis of Jacomijn J van Haersolte-van Hof and Anne K Hoffmann, ‘The Relationship between International Tribunals and Domestic Courts’, ch 24 above. 10 B Hanotiau noted that ‘ … it is true that even if the ICC Rules of Arbitration do not contain any provisions concerning res judicata, it would be difficult to imagine the ICC International Court of Arbitration approving a second arbitral award between the same parties on the same subject matter that contradicts a prior award already approved by the Court’. B Hanotiau, ‘The Res Judicata Effect of Arbitral Awards’, in Complex Arbitrations: Perspectives on their Procedural Implications (Special Supplement—ICC Int'l Court of Arb Bull, 2003) at 49. 11 See A Reinisch, ‘The Use and Limits of Res Judicata and Lis Pendens as Procedural Tools to Avoid Conflicting Dispute Settlement Outcomes’, 3(1) The Law and Practice of International Courts and Tribunals (2004). 12 In his dissenting opinion in the Chorzow Factory case before the PCIJ, Judge Anzilotti referred to res judicata as one of the ‘general principles of law recognised by civilised nations’, in the sense of Art 38 of the PCIJ Statute. See Interpretations of Judgements Nos 7 & 8 concerning the case of the Factory at Chorzow, 1927 PCIJ Ser A No. 11, at 27, 16 December (Dissenting Opinion of Judge Anzilotti). B Cheng, General Principles of Law as Applied by International Courts and Tribunals (Cambridge, Cambridge University Press, 1953) also noted that: ‘There seems little, if indeed any question as to res judicata being a general principle of law as to its application in international judicial proceedings’. See also Vaughan Lowe, ‘Res Judicata and the Rule of Law in International Arbitration’, 8 African J Int'l Comp L 38 (1996); Yuval Shany, The Competent Jurisdictions of International Courts and Tribunals (Oxford, Oxford University Press, 2003); Hanotiau, above n 10; Reinisch, above n 11. 13 W Dodge, ‘National Courts and International Arbitration: Exhaustion of Remedies and Res Judicata under Chapter Eleven of NAFTA’, 23 Hastings Int'l & Comp L Rev (1999–2000) at 357. 14Ne bis in idem—‘not twice for the same’—no legal action can be instituted twice for the same cause of action. It is a general concept originating in Roman Civil Law—but also found in common law jurisdictions. It is usually used in criminal law. 15ILA Committee on International Commercial Arbitration, Interim Report on ‘Res Judicata and Arbitration’ (Berlin, 2004). 16 See G Sacerdoti's Expert Opinion in CME Czech Republic BV v The Czech Republic, published in 2(5) TDM (2005) at 109: ‘Under public international law, res judicata has a limited recognition, as equivalent to “finality” of a decision, but that it does not prevent a different adjudicatory body, absent a specific treaty provision to the contrary, from hearing either in parallel or subsequently a dispute being substantially the same than another one, previously examined by another body, if this body has competence in accordance with its own jurisdictional basis’. 17 See cases cited in the ILA Interim Report on Res Judicata, above n 15: Effect of Awards of Compensation Made by the United Nations Administrative Tribunal (1954) ICJ Report 47 at 53; Case Concerning the Arbitral Award Made by the King of Spain on 23 December 1906 ( Honduras v Nicaragua ) 1960 ICJ Rep 192; South West Africa Case (1966) ICJ Rep 4 at 240; Request for Interpretation of the Judgment of 11 June 1998 in the Case Concerning the Land and Maritime Boundary between Cameroon and Nigeria ( Cameroon v Nigeria ), (1999) ICJ Rep 31 at 39; Maritime Delimitation and Territorial Questions between Qatar and Bahrain, (2001) ICJ Rep para 303. 18 See cases cited in the ILA Interim Report on Res Judicata, above n 15: Case 14/64, Mrs Emilia Gualco v High Authority of the European Coal and Steel Community [1965] ECR 51; Cases 172, 226/83, Hoogovens Groep v Commission [1985] ECR 2831; Cases 358/85, 51/86, France v Parliament [1988] ECR 4846 at 4849–50. See also KPE Lasok, The European Court of Justice: Practice and Procedure (London, Butterworths, 2nd edn, 1994) at 219. 19Pious Fund of the Californias ( US v Mexico ), Hague Ct Rep (Scott) 1 at 5 (1902); Trail Smelter ( US v Canada ), 3 RIAA 1905 at 1950 (1935). 20Southern Pacific Properties (Middle East) Ltd v Egypt, 3 ICSID Rep 112 (1985) and 3 ICSID Rep 131 (1988). 21Amco Asia Corp v Indonesia (Resubmission: Jurisdiction), ICSID, 89 ILR 552 at 560. The Tribunal in this case was confronted with the situation where an ICSID award rendered in an earlier stage of the same proceedings accorded limited res judicata effect to the decisions of the Annulment Committee (recognizing the nullification of the first award but refusing to follow the reasoning adopted by the Committee during re-litigation). 22Waste Management Inc v United Mexican States (II) (Mexico's Preliminary Objection) ICSID Case No. ARB(AF)/00/3, 26 June 2002 41 ILM 1315. 23Waste Management Inc v United Mexican States (I) , Case No. ARB(AF)/98/2 (26 May 2000). 24Plama Consortium Ltd v Republic of Bulgaria , ICSID Case No. ARB/03/24, Decision on Jurisdiction, 8 February 2005, at paras 180–1: ‘any decision rendered by the Tribunal will be binding upon PCL as well as on Bulgaria and will be res judicata between the parties. If Dolsamex is found to be the owner of the shares of PCL, it may have a claim against the company or against Mr. Vautrin; but the Tribunal does not see how PCL, with new shareholders, could successfully make the same claims against the Respondent as in the present arbitration or reopen the Tribunal's decision’. 25 See eg I Brownlie, Principles of Public International Law (Oxford, Clarendon Press, 6th edn, 2003) at 50, where he states that ‘There is no effect of res judicata from the decision of a municipal court so far as an international jurisdiction is concerned …’. 26ILA Committee on Commercial Arbitration, Recommendation 3(40) (Toronto, 2006). 27 See Reinisch, above n 11. 28Certain German Interests, 1925 PCIJ (Ser A) No. 6 at 20; Chorz?w Factory, PCIJ (Ser A) No. 17, at 27; The Mox Plant ( Ireland v UK ) Order of 3 December 2001, 41 ILM (2002) 405. 29The Mox Plant Case ( Ireland v UK ), Request for Provisional Measures, ITLOS, Case No. 10, 3 December 2001, at para 51. 30CME , Final Award, above n 5 at 433. 31 See Reinisch, above n 11. 32 Ibid . 33Amco v Indonesia , Decision on Jurisdiction, 25 September 1983, 1 ICSID Reports at 389. 34Klockner v Cameroon , Award, 21 October 1983, 2 ICSID Reports at 9. 35 C Schreuer, The ICSID Convention: A Commentary (Cambridge, Cambridge University Press, 2001) Art 25, para 216. 36 G Sacerdoti in his Expert Opinion in CME Czech Republic BV v The Czech Republic noted that the cases Amco v Indonesia and Klockner v Cameroon involved indeed a local subsidiary incorporated in the host state. However, ‘the protection granted to the foreign investor for investments made through a local company in the host state was not dependent on an “economic approach” drawn from the group of companies theory. These decisions constitute an application, in conformity with the ICSID Convention object and purpose, of an accepted notion of general international law’. Sacerdoti, 2(5) TDM (2005) at 110–27. 37 In this case, the arbitral tribunal found a lack of identity of parties between one case brought by a company against the Republic of Congo, and a second case brought by the Government of Congo against the shareholder of the company SARL Benvenuti & Bonfant v People's Republic of Congo (Case No. ARB/77/2) 15 August 1980, VIII YB Com Arb 145 (1983) at para 1.14. 38 Sacerdoti, ‘Expert Opinion’, above n 16 at 110–27. 39CME , Final Award, above n 5 at para 432. 40 Shany, above n 12. 41 D Reichert, ‘Problems with Parallel and Duplicate Proceedings: The litispendence Principle and International Arbitration’, 8(3) Arb Int'l 237 (1992) concluding that ‘Decisions of arbitral tribunals on lis pendens are generally too sparse and contradictory to constitute in any way representative statements of an accepted practice amounting to a procedural rule for international arbitration’. 42 See Reinisch, above n 11; Shany, above n 12; V Lowe, ‘Overlapping Jurisdiction in International Tribunals’, 20 Australian YB Int'l L 191 (1999); P Mayer, ‘Litispendance, connexit? et chose Jug?e dans l'arbitrage international’, in J-D Bredin, P Lative, JF Poudret, and F Terre (eds), Liber Amicorum Claude Reymond (Paris, Litec, 2004). 43 Agreed Minutes between the Czech Republic and the Netherlands at 3, as referred to in Sacerdoti, above n 16. 44ILA Committee on International Commercial Arbitration, ‘Final Report on Lis Pendens and Arbitration’ (Toronto, 2006). 45 In international commercial arbitration, some consider the very concept of lis pendens a logical impossibility. This view has been increasingly challenged. In the landmark decision in Fomento de Construcciones y c Contratas S.S. v Colon Container Terminal SS , the Swiss Federal tribunal unhesitatingly cast aside doubts on the applicability and advisability of lis alibi pendens in international arbitration and stated that the rules regarding competing jurisdiction of national courts are applicable, by analogy, to competing jurisdiction between a national court and an arbitral tribunal. See E Geisinger and L L?vy, ‘Lis Alibi Pendens in International Commercial Arbitration’, in Complex Arbitrations—Special Supplement, ICC Int'l Court of Arb Bull (2003) at 53. 46 Sacerdoti above n 16. 47Southern Pacific Properties (Middle East) Ltd. v Egypt , above n 20. 48 Shany notes that, although this reasoning might seem consistent with previous case-law since the case involved competition between an international tribunal and a domestic court, ‘there was nothing in the dicta of the tribunal to indicate that the tribunal limited its view to that particular category of jurisdictional competition only, and it could be cited in support of the view that international law does not recognise the lis alibi pendens rule at all’. See Shany, above n 12. 49 The decision to suspend proceedings rather than to decline jurisdiction is based on an earlier observation by the tribunal according to which one of the problems of parallel proceedings might be that both competing fora would end up declining jurisdictions, leaving the applicant without an effective remedy—‘negative conflict of jurisdictions’. 50Decision on Jurisdiction I, 27 November 1985, 1 ICSID Reports 112 at 129, para 84. 51Benvenuti & Bonfant Ltd v Congo , 1 ICSID Rep 330, 340 (1980). 52 Friedman in his presentation at the ICCA 2006 Conference in Montreal found the tribunal's reference to Benvenuti & Bonfant‘intriguing because it was not a fork in the road clause or even an investment treaty clause. … the award contains no real exploration of the lis pendens rule. The Azurix tribunal applied these lis pendens principles to interpret a treaty based on the fork in the road provision’. Friedman, above n 1. 53 The tribunal cited as authority: Certain German Interests in Polish Upper Silesia, Jurisdiction (1925) PCIJ, Series A, No. 6, at 20 (PCIJ jurisdiction not barred by the existence of separate proceeding); American Bottle Company ( US v Mexico ), (1929) 4 RIAA 435 at 437 (submission to another tribunal of an identical dispute between the same parties has no effect on tribunal's jurisdiction) and SSP (ME) Ltd v Egypt , First Decision on Jurisdiction, 27 November 1985, 106 ILR 502 at 529. 54 C Schreuer, ‘Calvo's Grandchildren: The return of Local Remedies in Investment Arbitration’, 4 The Law and Practice of Int'l Courts and Tribunals (January 2005) and C Schreuer, ‘Travelling the BIT Route: Of Waiting Periods, Umbrella Clauses and Forks in the Road’, 5 (2) JWIT 231 (2004). 55 Guatemala has notified ICSID that it requires the exhaustion of local administrative remedies as a condition of its consent to arbitration under the Convention (2003). 56 See Romania-Sri Lanka BIT, Argentina-Spain BIT Art X(3)(a), and Emilio Augustin Maffezini v The Kingdom of Spain , Decision on Jurisdiction, 25 January 2000, 16 ICSID Rev—FILJ 203 (2001). 57 See France-Argentina BIT (Art 8.2): ‘If such dispute could not be solved within six months from the time it was stated by any of the parties concerned, it shall be submitted at the request of the investor: either to the national jurisdictions of the Contracting Party involved in the dispute; or to investment arbitration … . Once an investor has submitted the dispute either to the jurisdictions of the Contracting Party involved or to international arbitration, the choice of one of the other of these procedures shall be final’. 58 See eg the following cases: Olguin v Paraguay ; Vivendi v Argentina ; Genin v Estonia ; Lauder v the Czech Republic ; Middle East Cement v Egypt ; CMS v Argentina ; Azurix v Argentina ; and Enron v Argentina . For a more detailed and complete description of these cases and analysis of this provision, see Schreuer, above n 54. 59 The question one could ask is whether the fork in the road corresponds somehow to the treaty formulation of the lis pendens principle and should therefore be interpreted accordingly. 60 P Turner, ‘The “Fork in the Road” Revisited’ in F Ortino, A Sheppard, and H Warner (eds), Investment Treaty Law, Current Issues, Vol 1 (London, BIICL, 2006) at 177–82. 61 C Soderlund, ‘Multiple Judicial Proceedings and the Energy Charter Treaty’, in C Ribeiro (ed), Investment Arbitration and the Energy Charter Treaty (Huntington, NY, JurisNet, LLC, 2006). 62 Provisional measures: ‘Except as the parties otherwise agree, the Tribunal may, if it considers that the circumstances so require, recommend any provisional measures which should be taken to preserve the respective rights of either party’. 63 Interim measures of protection: ‘A Tribunal may order an interim measure of protection to preserve the rights of a disputing party, or to ensure that the Tribunal's jurisdiction is made fully effective, including an order to preserve evidence in the possession or control of a disputing party or to protect the Tribunal's jurisdiction. A Tribunal may not order attachment or enjoin the application of the measure alleged to constitute a breach referred to in Article 1116 and 1117. For purposes of this paragraph, an order includes a recommendation’. 64‘An investor may seek interim relief, not involving the payment of damages, from the judicial or administrative tribunals of a Contracting Party, for the preservation of its rights and interests pending resolution of the dispute, without being deemed, thereby, to have submitted the dispute for resolution for purposes of subparagraph 4(b)’. DAFFE/MAI/EG1(96)12 ‘Settlement of disputes between an investor and a contracting party’. 65Art 26(2)(3). 66Art 10.18. 67Art 10.17. 68Art 10.17. 69 C Brower and J Sharp, ‘Multiple and Conflicting International Arbitral Awards’, 4 J WIn 211 (2003). 70 In his comments on a draft of this chapter, Thomas W?lde noted that an investor can only waive ‘his/her right’ and not those of somebody else. For instance, if a minority shareholder, even one with a controlling influence, waives his/her right, this may create a potential conflict of interest situation if the other shareholders do not agree with that waiver. 71Robert Azinian et al v Mexico , Award, 1 November 1999 (Case No. ARB(AF)/97/2) ) at para 86. 72 However, due to the fact that NAFTA Article 1135 limits a Chapter 11 tribunal to awarding damages, restitution, and costs, there is a possibility for a certain ‘claim splitting’ to be allowed: a foreign investor may bring a NAFTA claim for damages and at the same time or afterwards seek injunctive relief in domestic courts, since it cannot be granted by NAFTA tribunals. See Dodge, above n 13. 73Waste Management Inc v Mexico (I) , above n 23. 74 Ibid at para 27, 235–6. 75Waste Management Inc v Mexico (II) , above n 22, at para 29. 76 For a comprehensive discussion and analysis on umbrella clauses, see AC Sinclair, ‘The Origins of the Umbrella Clause in the International Law of Investment Protection’, 20(4) Arb Int'l (2004); Schreuer, above n 54; T W?lde, ‘The Umbrella Clause in Investment Arbitration—A Comment on Original Intentions and Recent Cases’, 6(2) J W In T (2005); W Ben Hamida, ‘La clause relative au respect des engagements dans les trait?s d'investissement’ (May 2005) at 21 (draft on file with author); E Gaillard, ‘Centre international pour le reglement des differends relatifs aux investissements (CIRDI): chronique des sentences arbitrales’, Journal du droit int'l 219 (2006) at 326–50; and Jacomijn J van Haersolte-van Hof and Anne K Hoffmann, Ch 24 above. 77 As Professor W?lde notes: ‘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’. W?lde, above n 76. 78 The first ICSID case that addressed, but did not analyse, the umbrella clause arose in 1998: Fedax NV v Republic of Venezuela (Case No. ARB/96/3) based on the BIT between the Netherlands and the Republic of Venezuela. In this case, the tribunal was unaware that there was an umbrella clause, and did not carry out any in-depth examination of the clause or its application. It simply applied the ‘plain meaning’ of the provision, that commitments should be observed under the BIT, to the promissory note contractual document. It found that Venezuela was under an obligation to ‘honour precisely the terms and conditions governing such investment, laid down mainly in Article 3 of the Agreement, as well as to honour the specific payments established in the promissory notes issued’. Fedax v Venezuela , Award, 9 March 1998, at para 29. The merits of the case were partially settled by the parties. 79 K Yannaca-Small, ‘Interpretation of the Umbrella Clause in Investment Agreements’, OECD Working Paper (October 2006). 80SGS Soci?t? G?n?rale de Surveillance, SA v the Republic of the Philippines , ICSID Case No. ARB/02/6, Decision on Jurisdiction, 29 January 2004; Waste Management Inc v United Mexican States (II) , ICSID Case No. ARB (AF)/00/3, Award, 30 April 2004; Consorzio Groupement LESI-DIPENTA c. R?publique alg?rienne d?mocratique et populaire , ICSID Case No. ARB/03/08, Award, 10 January 2005; Sempra Energy International v Republic of Argentina , ICSID case No ARB/02/16, Decision on Objections to Jurisdiction, 11 May 2005; Eureko BV v Poland , Partial Award, 19 August 2005; Noble Ventures, Inc. v Romania , ICSID Case No. ARB/01/11, Award, 12 October 2005; and LG&E Energy Corp., LG&E Capital Corp, LG&E International Inc v the Argentine Republic , ICSID Case No. ARB/02/1, Decision on Liability, 3 October 2006. However, the tribunals in CMS v Republic of Argentina , ICSID Case No ARB/01/8, Award, 12 May 2005; El Paso International Energy Company v the Argentine Republic , ICSID Case No. ARB/03/150, Decision on Jurisdiction, 27 April 2006; and Pan American Energy LLC and BP Argentina Exploration Company v Argentine Republic , Decision on Preliminary Objections, 27 July 2006, are the only ones so far faced with a ‘proper’ umbrella clause, ie ‘any obligation’, which adopted a narrow interpretation. 81SGS Soci?t? G?n?rale de Surveillance, SA v Pakistan , ICSID Case No. ARB/01/13, Decision on Jurisdiction, 6 August 2003; Joy Mining Machinery Limited v The Arabic Republic of Egypt , ICSID Case No. ARB/03/11, Award on Jurisdiction, 6 August 2004; Salini Construttori SpA and Italstrade SpA v The Hashemite Kingdom of Jordan, ICSID Case No. ARB/02/13, Decision on Jurisdiction, 29 November 2004. 82 See K Yannaca-Small, ‘Consolidation of Claims: A Promising Avenue for Investment Arbitration ’ in OECD, International Investment Perspectives (Paris, OECD, September 2006). 83 See the Australian International Commercial Arbitration Act 1989. 84 The word ‘may’ shows that a court faced with a connexity defence has discretion to accept or refuse the action. It is different when the court is faced with a lis pendens defence. 85 An example of examination of this definition is found in the Ship Tatry case where the ECJ held that ‘the term “irreconcilable … judgment” must be interpreted by reference to … the objective of … Art. 22 of the Convention’ which is ‘to improve coordination of the exercise of judicial functions within the Community and to avoid conflicting and contradictory decisions even when the separate enforcement of each of them is not precluded’ Case C-406/92, The Tatry [1994] ECJ I-5439 at para 54. 86Article 1046, Netherlands Code of Civil Procedure 1986. 87Section 6B, Hong Kong Arbitration Ordinance 1997. 88Colombia Decree 2279 of 10 July 1989. 89 See F Mantilla Serrano, ‘La Nouvelle L?gislation colombienne sur l'arbitrage’, Revue de l'arbitrage 41 (1992) at 54. 90 The CME v Czech Republic tribunal, in its Final Award ordering the Czech Republic to pay damages, reiterated the respondent's repeated rejection of CME's offer to structure the two cases so as to avoid potentially conflicting arbitral awards: ‘At the hearing the Respondent declined anew to accept any of the Claimant's alternative proposals … i) to have the two arbitrations consolidated into a single proceeding, ii) to have the same three arbitrators appointed for both proceedings, iii) to accept the Claimant's nomination in this proceeding of the same arbitrator that Mr. Lauder nominated in the London proceeding, iv) to agree that the parties to this arbitration are bound by the London Tribunal's determination as to whether there has been a Treaty breach, v) that after the submission of the parties' respective reply memorials and witness statements in this arbitration, the hearing be postponed until after the issuance of an award in the London Arbitration’. 91 When reviewing possible improvements to the 1985 UNCITRAL Model Law, the Secretariat proposed work on a provision on consolidation of claims. The Working Party, however, could not reach agreement on the importance to attach to this issue nor whether this issue was capable of resolution at that time. Ultimately, the UNCITRAL Model Law 2002 contained no reference to consolidation of claims. 92 Crivellaro suggests that ‘Article 26 of the ICSID Convention which stipulates that “consent of the parties to arbitration under this Convention … be deemed consent to arbitration to the exclusion of any other remedy … ” is an important reference point as a policy of consolidation since it excludes the parallel referral of the dispute to domestic courts and serves to avoid duplication of proceedings’. A Crivellaro, ‘Consolidation of Arbitration and Court Proceedings in Investment Disputes’, presentation at the ICC Institute of World Business Law, 24th Annual Meeting (Paris, 15 November 2004) (draft on file with the author). Professor Schreuer also suggests that the function of Art 26 is to create a ‘rule of priority vis-?-vis other systems of adjudication in order to avoid contradictory decisions and to preserve the principle of “ne bis in idem”?’. Schreuer, above n 35 at 359. 93 DAFFE/MAI(98)7REV1, available at <http://www1.oecd.org/daf/mai/pdf/ng/ng987r1e.pdf>. 94Art 9e of the Draft MAI: ‘An investor may withdraw the dispute from arbitration under this paragraph 9 and such dispute may not be resubmitted to arbitration under paragraph 2c. If it does so no later than 15 days after receipt of notice of consolidation, its earlier submission of the dispute to that arbitration shall be without prejudice to the investor's recourse to dispute settlement other than under paragraph 2c’. 95 Available at <http://www.dfait-maeci.gc.ca/nafta-alena/chap11-en.asp?#article_1125>. 96US-Chile Free Trade Agreement (Art 10.24) signed on 1 March 2004. 97US-Morocco Free Trade Agreement (Art 10.24) signed on 15 June 2004. 98US-CAFTA-DR (Art 10.25) signed on 5 August 2004. 99 Available at <http://www.dfait-maeci.gc.ca/tna-nac/cda-chile/chap-g26-en.asp#II>. 100 Available at <http://www.sice.oas.org/ctyindex/MEX/MEXagreements_e.asp>. 101Art 83 of the Japan-Mexico Free Trade Agreement, available at <http://www.mofa.go.jp/region/latin/mexico/agreement/agreement.pdf>. 102Art 33, available at <http://www.state.gov/documents/organization/38710.pdf>. 103Article 32, available at <http://www.dfait-maeci.gc.ca/tna-nac/documents/2004-FIPA-model-en.pdf>. 104 Usually, the same fact is a state measure which is allegedly in breach of the state's obligation. ‘This concept is a more precise criterion for consolidation than the “same dispute” requirement under the traditional lis pendens/res judicata theories’. See Crivellaro, above n 92. 105NAFTA Art 1126(2)(b), Art 33(6)(b) of the US Model BIT, and Art 32(2)(b) of the Model FIPA. 106 Partial consolidation further raises the question whether, and if so, to what extent, the individual claim tribunals should adjourn the proceedings before them, pending resolution by the consolidation tribunal. The consolidation tribunal in the softwood lumber cases (see below) has raised but not examined the question. See Canfor Corp v United States of America , Terminal Forest Products Ltd v United States of America and Tembec Inc et al v United States of America , Order of the Consolidation Tribunal, 7 September 2007, para 158, available at <http://naftaclaims.com/Disputes/USA/Softwood/Softwood-ConOrder.pdf>. 107Corn Products International, Inc v United Mexican States and Archer Daniels Midland Company and Tate & Lyle Ingredients Americas, Inc v United Mexican States , Order of the Consolidation Tribunal, 20 May 2005. 108 Request on consolidation by the USA, 7 March 2005: available at <http://www.state.gov/documents/organization/43492.pdf>. 109 See transcript of the consolidation hearing available at <http://www.state.gov/documents/organization/48508.pdf>. 110Canfor Corp v United States of America , Terminal Forest Products Ltd v United States of America and Tembec Inc et al v United States of America , Order of the Consolidation Tribunal, 7 September 2007, available at <http://naftaclaims.com/Disputes/USA/Softwood/Softwood-ConOrder.pdf>. 111 E Obadia, ‘ICSID, Investment Treaties and Arbitration: Current and Emerging Issues’, in Investment Treaties and Arbitration, ASA Swiss Arbitration Association Conference (25 January 2002) at 67–77. 112Salini Construttori SpA and Italstrade SpA v Kingdom of Morocco , ICSID Case No. ARB/00/4 in 42 ILM 609 (2003). 113Consortium RFCC v Kingdom of Morocco , ICSID Case No. ARB/00/6, available at <http://www.worldbank.org/icsid/cases/rfcc-decision.pdf>. 114 See Crivellaro, above n 92. 115Sempra Energy International v Argentina , ICSID Case No. ARB/02/16, Decision on Objection to Jurisdiction, 11 May, 2005. 116Camuzzi International SA v Argentina , ICSID Case No. ARB/03/2, Decision on Objection to Jurisdiction, 11 May 2005. Camuzzi has also raised a second claim in relation to its electricity distribution and transportation enterprise, Camuzzi International SA v Argentina, ICSID Case No. ARB/03/7; this claim is being heard by a different tribunal, separately and independently of its other claim. 117 One arbitrator was appointed jointly by Sempra and Camuzzi, Argentina appointed the second arbitrator, and the president of the tribunal was appointed by the Secretary-General of ICSID. Other procedural matters also appear to have been agreed by the parties, including the timetable for submissions and, where appropriate, submission of consolidated pleadings. See Decision, at paras 9–14. 118Electricidad Argentina, SA, & EDF International SA v Argentina , ICSID Case No. ARB/03/22. 119EDF International SA, SAUR International SA & L?on Participations Argentinas SA v Argentina , ICSID Case No. ARB/03/23. No procedural history of these two cases is available, thus the degree of integration and the procedures adopted, are, at present, unclear. 120Aguas Provinciales de Santa Fe, SA, Suez, Sociedad General de Aguas de Barcelona, SA and Interagua Servicios Integrales de Agua, SA v Argentine Republic , ICSID Case No. ARB/03/17. 121Aguas Cordobesas, SA, Suez, and Sociedad General de Aguas de Barcelona, SA v Argentine Republic , ICSID Case No. ARB/03/18. 122Aguas Argentinas, SA, Suez, Sociedad General de Aguas de Barcelona, SA and Vivendi Universal, SA v Argentine Republic , ICSID Case No. ARB/03/19. 123 For a detailed discussion of the advantages and disadvantages of consolidation in the context of commercial arbitration, see J Chiu, ‘Consolidation of Arbitral Proceeding and International Arbitration’, 7 (2) J Int'l Arb 53 (1990). 124 On the other hand, it is also possible that a party's costs may actually increase through consolidation of claims; this may arise, for example, where the increased complexity of the case results in a longer procedure than would have occurred had a party been required to be present only at a single unconsolidated arbitration. See E Gaillard, ‘Consolidation of Arbitral Proceedings and Court Proceedings’, in Complex Arbitrations: Perspectives on their Procedural Implications, Special Supplement—ICC International Court of Arbitration Bulletin (December 2003) at 35–42; G Born, International Commercial Arbitration (The Hague, Kluwer, 2nd edn, 2001) at 674; Chiu, above n 123. 125 M Platte, ‘When Should an Arbitrator Join Cases ’, 18(1) Arb Int'l 67 (2002). 126 Chiu, above n 123. 127 An issue for reflection is whether where the treaty contains an umbrella clause, consolidation should extend to claims and counter-claims under covered contracts, eg, for additional costs, delay, or unpaid invoices. Would it be possible or desirable to permit the consolidation of proceedings under a BIT (in which an investor is pursuing contractual claims through an umbrella clause) with proceedings under a contract (where a state is pursuing contractual claims or counter-claims against the same investor), provided of course that all claims arise out of the same contractual relationship? 128 See Crivellaro, above n 92. 129 Order of the consolidation tribunal in the Softwood Lumber cases (Canfor Corp v United States of America , Terminal Forest Products Ltd v United States of America and Tembec Inc et al v United States of America ), 7 September 2005, at para 125. 130 E Gaillard as President of the Canfor Tribunal said that ‘ … a consolidation tribunal established pursuant to NAFTA Article 1126 could dispose of these issues for the sake of consistency and for the sake of fair and efficient resolution of the claims … .consolidating similar claims is a very important issue for the integrity of NAFTA, for the integrity of the process, for the sake of consistency, and the way the whole treaty works.’ Canfor Corporation v United States of America , Hearing on Jurisdiction, Hrg Tr (‘Canfor Hrg Tr’) Vol 1 at 15: 20-1 (7 December 2004). 131 See eg UNCITRAL Model Law on International Commercial Arbitration 1985 (UNCITRAL Model Law) Arts 34 and 36; Art V New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards 1958; Netherlands Code of Civil Procedure 1986 Arts 1062–8; French Code of Civil Procedure 1981 Arts 1502 and 1504 ; English Arbitration Act 1996 ss 68, 69 and 103. 132Order of the Consolidation Tribunal in the Softwood Lumber cases, above n 129 at para 131. 133 Platte, above n 125; see position of Tembec in Order of the Consolidation Tribunal, ibid at para 42. 134 P Leboulanger, ‘Multicontract Arbitration’, 13(4) Int'l Arb 43 (1996) at 62. 135 Another argument advanced against the consolidation of claims in particular in the context of commercial arbitration is the difficulty in ensuring enforceability of the award. Doubts have been raised over the enforceability of an award rendered by a consolidated tribunal where consolidation was not agreed upon by the parties. The relevant literature suggests the following grounds upon which recognition and enforcement of an award rendered by a consolidated tribunal could be resisted under Art V of the New York Convention: (i) absence of an agreement in writing and (ii) irregular composition of the arbitral tribunal. For a detailed discussion on this issue, see Chiu, above n 123; AJ Van den Berg, ‘Consolidated Arbitration and the 1958 New York Arbitration Convention’, 2(4) Arb Int'l 367 (1986); S Jarvin, ‘Consolidated Arbitration and the 1958 New York Arbitration Convention—A Critique of Dr. Van den Berg’, 3(3) Arb Int'l 254 (1987); AJ Van den Berg, ‘Consolidated Arbitration and the 1958 New York Arbitration Convention—A Replique to Mr. Jarvin’, 3(3) Arb Int'l 257 (1987). 136 A number of authors, however, suggest that the absence of a consolidation clause is not of itself indicative of intent on the part of the contracting parties to exclude consolidation. Rather, they posit this omission may merely reflect that such a possibility had not been considered during their negotiations. Further, drafting a consolidation clause before a dispute arises may prove difficult or impossible where the issues and parties are not yet known. Consequently, these authors argue the absence of an arbitration agreement does not necessarily suggest intent to preclude consolidation of claims. As regards judicial modification of the arbitration agreement, Chiu argues this is not precluded by application of the doctrine of sanctity of contracts. See Chiu, above n 123. 137 Hascher describes consolidation as reflecting the will of the courts rather than of the disputing parties. D Hascher, ‘Consolidation of Arbitration by American Courts: Fostering or Hampering International Commercial Arbitration?’, 1 J Int'l Arb (1984) at 133–4. 138 Order of the Consolidation Tribunal, above n 129 at para 78, citing Henri Alvarez, ‘Arbitration under the North American Free Trade Agreement’, 16 Arb Int'l 393 (2000) at 414. 139Corn Products International, Inc v United Mexican States and Archer Daniels Midland Company and Tate & Lyle Ingredients Americas, Inc v United Mexican States (HFCS cases), Order of the Consolidation Tribunal, 20 May 2005, at para 12. 140 See eg UNCITRAL Model Law Art 11(3)(a). 141 Gaillard, above n 124 and Platte, above n 125 stating that: ‘ … the right to nominate an arbitrator need not be treated as sacrosanct’. 142 Hascher, above n 137 at 135: ‘ … Matters are enormously complicated by the incorporation of separate disputes in a single arbitration proceeding. Each party assumes the additional burden of hearing claims, giving evidence and discussing testimonies with all the other parties involved. There is a higher probability of delays. Risks of omission and error are multiplied’. 143 See Chiu, above n 123. 144 See Platte, above n 125. 145 Order of the Consolidation Tribunal in HFSC cases, above n 139 at para 7. 146 In the context of commercial arbitration, notwithstanding these arguments, and the few legislative acts which provide for consolidation, many jurisdictions have found this consideration insurmountable. Concerns over confidentiality formed a determinative factor in the English legislature's rejection of a provision providing for court-ordered consolidation. 147 Order of the Consolidation Tribunal in the Softwood Lumber cases, above n 129 at para 147. 148 A Diamond, ‘Multi-party Arbitrations: A Plea for a Pragmatic Piecemeal Solution’, 7(4) Arb Int'l 403 (1991). 149 See eg Arts 40 and 41 Stockholm Chamber of Commerce Arbitration Rules 1999; Art 31(3) ICC Arbitration Rules 1998. 150 Leboulanger, above n 134 at 67. Authors: Thomas W W?lde, Borzu Sabahi Keywords: Remedies and costs – Compensation – Costs and expenses – Damages – Date of valuation – Discounted cash flow (DCF), anticipated future profits – Interest – Valuation This chapter begins by explaining the terminology, different functions, effect of damages awards on the acceptability of investment arbitration jurisprudence, and the main principles (mainly the Chorzow Factory dictum) governing awarding damages. Section 2 explains some of the elements that may have an impact on awarding damages, such as litigation dynamics, overcompensation, and moral hazard, and finally damnum emergens and lucrum cessans. Section 3 examines compensation for expropriation, including standard of compensation, valuation methods, compensation for creeping and regulatory expropriation, date and time of expropriation. Section 4 discusses compensation for breach of nonexpropriatory investment treaty disciplines such as national treatment, denial of justice, legitimate expectations. Section 5 examines the damages for contractual breaches as well as the breach of an umbrella clause. Section 6 touches upon the concepts of causation, as well as compensation-reducing and compensation-enhancing elements. Section 7 examines the role of equitable considerations in awarding damages. Section 8 examines different aspects of awarding interest. Section 9 discusses burden of proof and evidentiary issues. Section 10 provides an overview of other issues not discussed in this chapter.

0subscriber_article?script=yes&id=%2Fic%2FMonograph%2Flaw-iic-9780199231386&recno=62&searchType=browse Chapter 26 Compensation, Damages, and Valuation ?

(1)Fundamental Concepts relating to Damages, Indemnity, and Compensation1052

(a) Terminology 1052

(b) Different Functions of Compensation 1053

(c) Effect of Damages Awards on the Acceptability of the Investor-State Arbitration Regime: Commercial Arbitration versus Administrative Law Paradigm 1054

(d) Main Principles: Chorzow Factory Dictum 1056

(i) Chorzow Factory Dictum 1056

(ii) Rationale Underlying the ‘Practical’ Primacy of Compensation over Restitution: Is It Still Valid? 1058

(2)Valuation Methods and Techniques1062

(a) Litigation Dynamics and its Consequences for Valuation 1063

(b) Over-compensation, Moral Hazard, and Risk 1064

(c) Damnum Emergens and Lucrum Cessans (Direct Damage and Lost Profits) 1065

end p.1049

(3)Compensation for Expropriation and Action Tantamount to Expropriation1068

(a) Standard of Compensation for Expropriation 1068

(i) Historical Development 1068

(ii) Current Status of Standard of Compensation for Expropriation in Light of the Language of the BITs 1069

(b) Valuation of the Expropriated Property 1070

(i) Market-based Valuation Approaches 1070

(ii) Book Value and Related Methods Focusing on Historic Cost and Income Information 1072

(iii) Discounted Cash Flow (DCF) Method 1074

(iv) Net Book Value (NBV) v Discounted Cash Flow (DCF) 1075

(v) Appropriate Risk Factor/Discount Rate 1077

(vi) Special Case of Investment Projects before Start-up and without Extensive Record of Performance 1078

(c) Compensation for Creeping and Regulatory Expropriation 1079

(d) Relevant Valuation Time and Date 1081

(4) Compensation for Breach of Other Standards of Treatment 1082

(a) National Treatment (Non-discrimination) 1083

(b) Fair and Equitable Treatment 1086

(i) Compensation for Breach of Fair and Equitable Treatment 1087

(ii) Denial of Justice 1088

(iii) Legitimate Expectation 1088

(5)Compensation for Contractual Breaches—Umbrella Clause1090

(6)Causation, Compensation-reducing, and Compensation-enhancing Elements1093

(a) Causation 1093

(b) Compensation-reducing Elements 1095

(c) Compensation-enhancing Elements 1099

(d) Direct and Derivative Damage—Multiple Damages Claims 1101

(7)Role of Equitable Circumstances1103

(8)Interest1106

(a) Introduction 1106

(b) Simple v Compound Interest 1107

(c) Rate of Interest 1107

(d) Date from which the Interest Should Accrue 1108

(e) Pre-award and Post-award Interest 1109

(f) Interest and Islamic Law 1110

(9)Burden of Proof, Evidence, and Procedure1110

(10)Other Issues1112

Concluding Remarks1114

end p.1050

IN the dynamics of investment arbitration, the question of compensation and damages is often the poor cousin when the battle royal rages first about jurisdiction and then about the merits. For the lawyers, the question of win or lose is easily about these questions, with only an afterthought for ‘how much’. It is the same with academic study and professional comment. International lawyers and academics are also interested in creating new legal doctrines. Clients who have the key interest, by contrast, are not interested in building legal norms; they simply hope to get their grievance remedied, often by getting paid as much as possible. This is what this chapter is about: how investors obtain compensation and damages for their grievances once the tribunals have decided they are competent to make an award and once they have decided that at least on one count the investor has proven that the state is liable. 1

This chapter contains ten sections. Section 1 explains the terminology, different functions, effect of damages awards on the acceptability of investment arbitration jurisprudence, and the main principles (mainly the Chorzow Factory dictum) governing awarding damages. Section 2 explains some of the elements that may have an impact on awarding damages, such as litigation dynamics, overcompensation, and moral hazard, and finally damnum emergens and lucrum cessans. Section 3 examines compensation for expropriation, including standard of compensation, valuation methods, compensation for creeping and regulatory expropriation, date and time of expropriation. Section 4 discusses compensation for breach of non-expropriatory investment treaty disciplines such as national treatment, denial of justice, legitimate expectations. Section 5 examines the damages for contractual breaches as well as the breach of an umbrella clause. Section 6 touches upon the concepts of causation, as well as compensation-reducing and compensation-enhancing elements. Section 7 examines the role of equitable considerations in awarding damages. Section 8 examines different aspects of awarding interest. Section 9 discusses burden of proof and evidentiary issues. Section 10 provides an overview of other issues that have not been discussed in this chapter. Finally, few concluding remarks are presented.

end p.1051

(1) Fundamental Concepts relating to Damages, Indemnity, and Compensation

This section discusses fundamental concepts underlying the practice and theory of compensation. The more detailed subsequent discussion will then refer back to this general part.

(a) Terminology

‘Compensation’ and ‘damages’ are perhaps the two terms that more than any others are used in the international law terminology dealing with reparation, and naturally in this chapter. The term ‘compensation’ is sometimes understood to be distinguished from ‘damages’ (‘indemnity’) by being less linked to an unlawful act, as, for example, when compensation is due for an otherwise lawful expropriation. 2 Compensation can also be understood as the duty to make financial reparation for expropriation or quasi-expropriatory conduct. Since the several investment disciplines or obligations are closely related and could even be said to express the common theme of abuse of governmental power towards the investor, that could cover as well the non-expropriatory disciplines to the extent they produce a comparable effect on the investor. Finally, compensation may be closer to the concept that governments are responsible for the detrimental impact they cause through administrative conduct. 3

‘Damages’ has a closer connection to unlawful conduct in general and, more specifically, breaches of contract. It is generally seen as the legal duty to pay for the detrimental consequences that the victim of an unlawful act has suffered. 4 It is also linked more closely to the contractual approach to damages, focusing on

end p.1052

the consequence of breach of a contractual obligation, rather than to the expropriation approach, which is the predominant influence on international investment law. 5 Compensation, therefore, is more general in character. 6 For the purpose of this chapter, however, the two terms (plus ‘indemnity’) should be considered as largely synonymous. 7

(b) Different Functions of Compensation

Compensation not only indemnifies the victim for losses suffered; but also ensures that the perpetrator—the tortfeasor 8—does not profit from breaching the law by becoming unjustly enriched. 9 In this way, compensation also encourages compliance with the applicable legal rules by both providing a sanction for breaching a rule and removing an incentive (enrichment) for the same. The mere prospect of a plausible arbitral, contract— or treaty-based, claim will often help to settle or avoid disputes. It contributes thus to the good-governance signalling effect of investment arbitration, 10 by contributing to a definite resolution of the dispute in a juridified procedure—rather than by political or informal pressure or bribery.

end p.1053

(c) Effect of Damages Awards on the Acceptability of the Investor-State Arbitration Regime: Commercial Arbitration versus Administrative Law Paradigm 11

Very little proper economic analysis has been carried out on the question of compensation in investment disputes. 12 Nor have arbitral tribunals or the damages debate given much thought to it—partly because economic analysis of the larger impact of rules created by tribunals is at present still alien to both arbitrators and advocates, who mainly have a legal background. There are several issues that merit the attention of tribunals: an important one pertains to the perception of the interested and sometimes critical communities 13 of this novel regime of investor-state arbitration, which may have a direct impact on its acceptability and legitimacy. The perception of these communities is more likely to be formed by the damages awards, which are more tangible, by virtue of providing for the payment of a monetary amount and thus more easily intelligible, than by the jurisdictional and merits awards. There is nothing as likely to fuel a backlash as damages awards that are seen as excessive and

end p.1054

are not founded on satisfactory reasoning. 14 Damages awards are the most sensitive—but the least developed—part of investment arbitration awards. Any plausible criticism of ‘excessive’ awards is therefore likely to undermine the essential political legitimacy of investment arbitration. A trend more restrictive of investor rights has already started, 15 which could only worsen if arbitral tribunals render awards that are unsatisfactorily reasoned and can be legitimately viewed as excessive. 16 A few very visible awards tend to overshadow the fact that most investment claims fail either in general or to generate significant compensation awards, particularly when contrasted with the initial claims made. 17 One also needs to bear in mind that some precedents are of limited value—in particular from claims commissions, as these commissions have a limited budget to distribute.

Further, there is a divergence of the two main ‘paradigms’ with respect to damages/compensation. From a public and in particular administrative law perspective (both international and comparative), compensation is just one of several possible remedies. 18 The approach is typically more restrictive in light of the concern not to ‘chill’ and immobilize governmental action by the prospect of large damages awards; 19

end p.1055

it also prefers as a rule ‘restitution’—re-doing a procedurally flawed decision in a procedurally correct way or annulment—to payment of damages. 20 The commercial arbitration approach which has greatly influenced the materially quite different area of investment arbitration (in our eyes, a form of international administrative law), on the other hand, emphasizes the way damages are calculated in commercial arbitral awards for breach of contractual duties. This approach emphasizes the ‘benefits of the bargain’. 21 These issues are at the moment unresolved; it is not clear if investment arbitration will ultimately rather lean towards the precedents set by comparative administrative law or commercial arbitration. But the caution intuitively though inchoately applied by many investment treaty tribunals towards an extensive notion of damages may involve an understanding that judicial review of governmental action requires a very cautious approach to compensation, with restitution including annulment rather as the primary sanction followed, in appropriate cases, by financial compensation. Comparative administrative law also suggests that the amount of compensation should be related to the significance and intensity of breach (proportionality between injury and reparation); that principle is not yet well developed in investment law except in a few comments on the need for higher compensation in case of unlawful expropriation, breach of stabilization commitments, and egregious breaches. One needs to bear also in mind that well-established international courts may have more legitimacy for adjudication that intrudes into the national regulatory space; but then, very high compensation awards are likely to be often much more intrusive than restitutive orders to repeat an administrative procedure.

(d) Main Principles: Chorzow Factory Dictum

(i) Chorzow Factory Dictum

The classic starting point in a damages analysis is the 1928 PCIJ judgment in the Chorzow Factory case: 22

end p.1056

The essential principle contained in the actual notion of an illegal act—a principle which seems to be established by international practice and in particular by the decisions of arbitral tribunals—is that reparation must, as far as possible, wipe out all the consequences of the illegal act and re-establish the situation which would, in all probability, have existed if that act had not been committed. Restitution in kind, or, if this is not possible, payment of a sum corresponding to the value which a restitution in kind would bear; the award, if need be, of damages for loss sustained which would not be covered by restitution in kind or payment in place of it—such are the principles which should serve to determine the amount of compensation due for an act contrary to international law. (Emphasis added)

The Chorzow statement 23 expressed the prevailing opinion in 1928 on breaches of legal obligations between states. It regarded restitution as the first-ranked remedy 24 and monetary compensation as a secondary remedy in case direct restitution was not practical. The Chorzow dictum is also illustrative of a dilemma which will continue throughout this analysis: it requires the comparison between a real situation, on one hand, and a hypothetical situation, on the other; that is, how would reality have—in theory—evolved had the unlawful act not occurred. 25 That approach has been termed by Irmgard Marboe—in reliance on civil law damages concepts—the ‘differential method’ as it compares the real with a hypothetical course of events.

This standard relies on speculating how a hypothetical course of events would develop. Thus, in most cases, it will not provide significant certainty. It requires going back in time to the moment before the unlawful act occurred. From that moment on, however, the intellectual operation becomes difficult: should one omit only the unlawful act and forecast how things would have moved on; or should one conjecture how the government would have been able and likely to act in a lawful way? In this comparison, one should not assume that life would simply have stood

end p.1057

still, but should identify what it is most likely that the government would and could legitimately have done. That requires the construction of a hypothetical course of events with necessarily speculative elements. 26 Such a hypothetical course of events extends both into the past (how would the government have acted if it had acted lawfully?) and into the future, to calculate damages due to the interference in a long-term, usually contractual, relationship (how would life have moved on, or how is it likely to move on if the ‘lawful-conduct scenario’ were to apply?) 27 The longer into the future the extension is, the more speculative the claimed losses. 28 Given the reasonable aversion of investment arbitral jurisprudence against ‘speculative’ losses, one should raise the question whether ‘compensation’ under the ‘expropriation’ approach, that is, the focus on the going concern value, should not in most cases provide a comparator and perhaps also a cap for breach of non-expropriation treaty disciplines, except in cases of egregious breaches meriting a more extensive sanction.

(ii) Rationale Underlying the ‘Practical’ Primacy of Compensation over Restitution: Is It Still Valid?

A key element highlighted in the Chorzow award (‘if this is possible’—‘as far as possible’)—is the practicality of the remedy. The last 80 years, however, have been influenced by the idea that while restitution—restoring the status quo ante—may be preferable in theory, 29 it is not practical. 30 Because, first, the respondent state may be unwilling or unable to undo what was done (due to political or internal legal and

end p.1058

constitutional constraints); secondly, compelling a state by an arbitration award to annul the measures that it has taken may be seen as an undue interference with its sovereignty; thirdly, the situation may have evolved in such a way that restoration becomes impractical or impossible. 31 Finally, the discretion of an arbitral tribunal is limited by the rule non ultra petita; meaning that if the parties have not asked for a particular remedy, then the tribunal should not award it. 32

Compensation is the one remedy that is traditionally seen as the least intrusive into sensitive areas of government sovereignty because the international tribunal does not order the government to follow a particular course of conduct but rather requires it to pay in the fungible currency of money for the harm done. 33 The idea behind this traditional notion is that for individuals and companies, money is the best and most fungible expression of their interests, while for the state pride, prestige, and domestic political sensitivities against imposition of commands from abroad is more important than the much more anonymous role of money.

It is for these reasons that jurisprudence, theory, and now Article 35 of the International Law Commission's Articles on Responsibility of States for

end p.1059

Internationally Wrongful Acts, still start with the restitution obligation, but qualify it readily by considerations of impossibility and proportionality. 34 But these limitations—de facto excluding restitution and focusing exclusively on monetary compensation—may no longer be appropriate to the new disciplines in investment treaties, which have been designed in reaction to the emergence of the regulatory state.

The WTO dispute settlement regime, for instance, which is the most influential and effective system of international adjudication relating to external disciplines on government conduct, does not use financial compensation. 35 Similarly, in comparative administrative law and adjudication, financial compensation takes a second place behind orders against the pertinent state authority to rescind its action or to act in specific ways (‘specific performance’). 36 Similarly, in EU law, the normal remedy is striking down a government measure (regulation) or ordering a government to take measures (eg to recover unlawful state aids) rather than providing compensation. 37

In light of these developments, it is not certain if the traditional perspective is still valid as state authorities are now more used to interaction with international adjudicatory bodies; in addition, large damage awards have the same, or even more politically intrusive potential than direct orders requiring specific conduct. 38 The view that restitution—in particularly of severe breaches of procedural rules (‘denial of justice’), but also other forms of ‘fair and equitable’ and ‘national’ treatment—is so ultra-sensitive that in practice it has to yield to financial compensation may rather

end p.1060

reflect the attitudes of and about the nation state of the 19th century rather than the modern ‘market state’ embedded in an intricate international network of rules, institutions, procedures, and other forms of interaction. 39 In the public law of modern states (and in international human rights—perhaps the closest to investment arbitration), the primary consequence of unlawful conduct by the administration is injunctive relief 40—often the order to repeat the incriminated measure but this time in a lawful way—rather than compensation.

It is recognized that courts and tribunals have an inherent power to award a variety of remedies subject to the limitations available in the specific treaty and also in the wider framework of international law. 41 That inherent power provides considerable flexibility—more than is recognized at present. 42 The ILC Articles—with the broad concept of reparation, broken down into the subcategories of restitution, satisfaction, and compensation, including combinations of them—should therefore be read not only against the practice of international courts and tribunals, but also against the remedies available in national courts. 43 To the extent this inherent power to determine the proper remedy is not limited by treaty, 44 advocates and tribunals should therefore think not only of monetary damages as they usually do, but about the most suitable and practical forms of remedy permissible under the applicable investment treaty. 45 The Goetz v Burundi tribunal's both flexible and effective

end p.1061

procedure in this respect merits greater attention. The Goetz tribunal did not render a final award initially, but only a decision on liability, finding the government responsible for indirect expropriation. It postponed its decision on compensation, giving the parties time to agree on compensation within four months and also the government to consider reissuing a free zone certificate (ie the closest solution to full restitution). Ultimately, the government paid the compensation and also reissued the certificate. The tribunal then embodied the settlement agreement (consisting of both cash and a reissued certificate) in an award. 46

Even within the limitations imposed by treaties, tribunals still have flexibility in terms of using instruments such as recommendations, interim orders, and awards, and conditionalities attached to awards based upon specific powers granted by treaties or applicable procedural rules, provided that the ultimate, final, and enforceable award complies with the respective treaty.

(2) Valuation Methods and Techniques

Assessing damages requires valuation. A major challenge for the assessment of damages is therefore to find an objective valuation method which can be founded on rational reasoning, and on objective standards independent from an individual's subjective judgment or the psychological dynamics of litigation and internal tribunal decision-making. While earlier awards and judicial practice in some legal cultures accepted the ‘discretionary judgment’ of the adjudicators and did not establish too stringent expectations for the reasoning justifying a damages award, that is no longer the case. Damages awards are now expected to be explained carefully and in some detail. Various methods and techniques are used in order to assess damages. Some techniques rely on historic cost accounting and book value of the investment. 47 More recent techniques do not look backwards, but forwards: they attempt to estimate the market value of the investment based upon its ability to generate profits

end p.1062

in the future. ‘Market value’ looks at how the market—at present—does or would hypothetically value an asset. Market value may be determined in a variety of ways; for publicly traded companies in a competitive market, the easiest way would be to start with prices of publicly traded shares—though market volatility will in many cases require an averaging-out of share prices over a prolonged period. If there is no such price, which in investment arbitration is usually the case, then one may benchmark the investment against comparable transactions in that market or perhaps in other similar markets. A related method is to rely on an offer made in an arm's-length transaction to acquire the investment as a point of reference and make the necessary adjustments. 48 Application of the more recent methods is easier said than done, however. They raise serious problems which underlie most disputes and need more in-depth examination. Therefore, in the following subsections we will explain different concepts that have an impact on valuation such as litigation dynamics, over-compensation, moral hazard, damnum emergens, and lucrum cessans.

(a) Litigation Dynamics and its Consequences for Valuation

The damages issue cannot be understood without appreciation of the litigation dynamics and the political economy context. 49 Valuation complexities and lack of effective penalties for manifest over- or under-statement of the amount of damages are among such dynamics. They may lead tribunals to split the baby, 50 that is, award damages at a mid-point between claimant and respondent positions, 51 or engage in an ex-post facto rationalization of the amounts awarded, which in turn may

end p.1063

encourage the advocates (and their experts) to push their claims to very high or very low valuation estimates.

These dynamics call for both sanctions for overstated claims, for example through cost allocation, 52 and more commitment by tribunals to focus on the damages issue, 53 if need be by having a separate phase after the merits' award on damages. Tribunals may also consider appointing a neutral expert to assist them in understanding how the valuation models presented by the parties work. 54 Splitting the difference is an emergency brake that should in most cases not be necessary if tribunals are prepared to spend more time and resources on the issue of damages.

There are also internal tribunal dynamics at work. In a consensus-based award (which most chairpersons prefer) the compensation amount (including interest and the cost decision) may be used as an (implicit or explicit) bargaining variable to achieve a unanimous agreement. The choice of valuation methodology favouring a lower-end outcome or of compensation-reducing concepts (eg mitigation, contributory negligence) can be used to achieve a consensus. That incentive is absent in awards with a dissent where internal tribunal dynamics (both absence of incentives for a consensus and ‘chairman's revenge’ against the dissenter) might at times lead to higher compensation (and cost) awards.

(b) Over-compensation, Moral Hazard, and Risk

Compensation analysis needs to examine and understand properly the issue of risk. Misunderstanding the risk associated with a business activity when valuing the business may result in over-compensation. To avoid this situation, a damages award should ensure the investor is put into a risk/reward-wise equivalent position, for example, either equivalent to a situation where the business is continued with the same exposure to risk or to a risk-free investment (eg by cash compensation),

end p.1064

where the amount of compensation reflects freeing-up from the earlier risk exposure. Failure to factor in the impact of the risk would put the investor in a significantly better position. It works like substituting by compensation a high-yield, high-risk ‘junk bond’ for a high-yield, but now low-risk government bond. This latter scenario would create a situation well known in the insurance industry: moral hazard. 55 Moral hazard means that the investors no longer have to work at predicting, managing, and minimizing the political risk since by investment arbitration (functioning as a special form of political risk insurance) they can reallocate it to the respondent, with only the litigation risk (which can be substantial) remaining.

Compensation rules should create incentives for all relevant parties in an investment dispute to be mindful of the risk in the first place, to look for risk minimization strategies, and when the risk materializes, to adopt prudent risk mitigation measures. These efficiency incentives are already embedded in the principles of contributory negligence/damage mitigation. In addition, a proper reflection of the risk of the investment can be achieved primarily by applying a discount rate that reflects the risk.

(c) Damnum Emergens and Lucrum Cessans (Direct Damage and Lost Profits)

The concept of in particular ‘lucrum cessans’—lost profit—has been playing a very confusing role in compensation; what is more, it is a concept that can most easily lead (and has led) tribunals into awarding double recovery if the implications of this concept and other compensation and valuation methods are not fully thought through. 56 Reisman and Sloane, in their discussion of the valuation of a directly expropriated property, consider the concept as ‘economically anachronistic’ which

end p.1065

could lead to double-counting; they do, however, see value in the concept as a ‘deterrent’ to penalize egregious expropriations—‘essentially a fine’. 57

The damnum emergens/lucrum cessans combination, which was the traditional remedy for breach of contract, needs to be seen as arising at a time when valuation was backward/historic and based on the accounting value of individual items of property. That did not represent market value properly as the combination of all items—the package value including goodwill—and the ability to make profits was not taken into account. The lucrum cessans add-on was therefore meant, at least until payment of a judgment, to compensate for the shortcomings of a purely historic and cost-based focus on individual property items. 58 It was thus a reaction to the shortcomings of a purely historic-cost-based approach when applied to an operating business. Thus far, it may have made sense. It seems that (without further analysis of the history of the concept) it envisaged actions like damage to a workshop; compensation had then to enable the owner to repair and replace the workshop, but also to pay the profits he or she would have made until the workshop was operating again. 59

The problem with these concepts is that they coexist with modern valuation methods which are forward-looking and incorporate future net income streams; even more so, in the case of long-term contracts (where most risk of exorbitant double recovery exists and has materialized), 60lucrum cessans can tempt an unwary tribunal to first order payment of market value calculated in a historic, cost-based way and then again the value of future income expected long into the future. Carrying-over the lucrum cessans (‘lost profits’) concept into the world of modern, future-oriented valuation methodology thus very easily results in multiple recovery: the value based on historic costs (which should on average be equal to the going concern/market value), the going concern/market value based on future net income streams, and, if

end p.1066

the future income is not discounted by the risk factor, a third measure of the value. In addition, the investor can invest the prior investment expenditures—once recovered by compensation payment—in one new project without risk. That can provide the investor with twice the value (once from DCF-based net present value and once from future profits or from historic cost); it will also free him from the risk of the original investment. The slope towards multiple recoveries is therefore particularly slippery when long-term contracts are in dispute. 61

When dealing with ‘lucrum cessans’, one should consider that it is basically a concept for the awarding of damages for breach of contract, but not for expropriation where the commercial value of an investment (a ‘going concern’) 62 or ‘fair market value’ 63 of the project is primarily at issue. That limitation is often not appreciated. It only becomes relevant in expropriation valuation if a method was used that merely added the value of separate property items, without taking into account their package value as a business. 64 Where value is established either by market perception or comparable transactions or by an extensive cost-based (inflation-upgraded) approach or by the NPV/DCF method, lucrum cessans is incorporated in such value; any separate payment in recognition of lost profit would amount to double recovery. We therefore concur with the extensive study by Irmgard Marboe which concludes that the ‘damnum emergens’—‘lucrum cessans’ dichotomy is a concept related to the law of damages which does not fit compensation for expropriation. 65

The conclusion of this general discussion of valuation concepts, of relevance to the determination of compensation under the main ‘heads of claim’ discussed below, is that there is not one fool-proof method. There are a variety of valuation methods, each with its drawbacks. It is best—and that also seems to be the main current arbitral practice—to apply a combination of the various methods and then to try to narrow down the differences between the results of applying each by understanding, assessing, and judging the reasons for such differences.

end p.1067

(3) Compensation for Expropriation and Action Tantamount to Expropriation

(a) Standard of Compensation for Expropriation

(i) Historical Development

There has been extensive debate during the last 80 years over the appropriate standards of compensation for expropriation. The issue has been at the core of the confrontation between, on the one hand, capital-exporting (mainly Western developed) and capital-importing countries (here mainly the former Communist and then developing countries). 66 In essence, there have always been two main positions, each one with some modification and in a softer and a harder form.

The Western position 67 has always been that there is an international law obligation of full compensation; that is expressed in terms such as ‘prompt, adequate, and effective’ (the Hull formula), ‘market’ or ‘genuine value’. The term ‘appropriate’, in itself vacuous, has been understood as denoting both the Western position and the New International Economic Order (NIEO) position. It occurs in the almost unanimous and therefore more authoritative UN resolution on permanent sovereignty over natural resources of 1961. 68 That position drew strength from some statements of the PCIJ (Chorzow Factory case and others). 69

end p.1068

The NIEO position, best expressed in the majority-based resolution of the UN General Assembly in 1974 (against the vote of virtually all capital-exporting Western countries), contended that compensation issues are exclusively settled under national law, a principle that is related to the Calvo doctrine. 70 Therefore, if the expropriating state did not want to pay compensation, it was free to do so. The NIEO position also relied on the fact that in lump-sum settlements usually less than full compensation was paid. 71

(ii) Current Status of Standard of Compensation for Expropriation in Light of the Language of the BITs

Most if not almost all modern investment treaties include the principle of full compensation (market value, genuine value, adequate). 72 Whatever one makes of differences in language it seems that currently the overwhelming state practice in concluding investment treaties has outweighed the challenge to customary international law in the NIEO period. 73

end p.1069

This has two implications for the required standard of compensation for expropriation: first, nearly every expropriation dispute will be adjudicated not under customary international law, but under specific binding treaty language. Secondly, should there be cases where international law may still be applicable (eg by reference, Art 42 ICSID Convention, to international law) in lieu of a specific treaty standard, then the weight of state practice opinio iuris sive necessitatis is rather towards the ‘full’ than the ‘appropriate’ or state-regulated compensation standard. 74

(b) Valuation of the Expropriated Property 75

(i) Market-based Valuation Approaches

There is largely an agreement that the ‘going concern’ as ‘fair market value’ should be the principal objective of valuing expropriated assets. 76 However, this is easier said than done for several reasons. First, almost all of the recent and prospective takings involve unique assets for which there is no established competitive market. 77 Secondly, for the same reason, proxies for market value, such as transactions in other countries by other companies at other times involving comparable assets, are never likely to be fully comparable; adjustment for quality, time, location, strategic interest premium, and risk are inevitable. Thirdly, market instability makes the market value dependent to a large extent upon the timing of the expropriatory events, such as the valuation time, which may be difficult to determine in a precise manner. Benchmarking against comparable transactions may yield an objective value, but markets are not rational: 78 they can move within quite short periods between extreme highs and extreme lows, resulting inter alia from the divergent application of forward-looking methods in assessing the value of the respective assets. Internet

end p.1070

‘dotcom’ companies that were valued (based on NPV-DCF methodology) at 1000 in 2001, lost, often with the same business continuing, up to 99 per cent of their value by the end of 2002. 79 Small changes in the market's estimates over future growth rates and risk can have rapid and dramatic repercussions for valuation. Market attitude towards risk is particularly volatile.

Should one use market value, then, the choice of the comparable transaction and the time of the comparator transaction will make a crucial difference. For example, a valuation of an asset by Enron, endorsed by the world's leading business consulting firm in 2001, may easily have been 20 times more than a valuation of the same, continuously and stably producing asset, a year later. 80 The better approach is clearly to select not a point in time, but a larger range in time, which allows the averaging out of at least shorter-term swings in market moods. In particular, as expropriation risk evolves incrementally and as markets perceive and react to risk, the point of time—or the time range for averaging a market valuation—becomes crucial and inevitably a matter of subjective and biased selection. 81 Party-appointed experts will therefore have little difficulty in reading from market data either a very high value (ie before markets become concerned about political risk) or a very low value (ie after markets have become—with usually the same amount of information available to the public—worried about the political risk). The most persuasive indicators of ‘fair market value’ are valuations achieved by objective and competitive procedures, for example, public tender. But such a method is usually not available for investment projects except at the very outset when truly competitive tenders were used in the privatization process. If the investor, in such circumstances, and relatively recently, has received credible offers for its investment (all or in part), those may indicate the valuation by market participants, though again they may reflect both over- or under-valuations at a particular point in time. 82

end p.1071

(ii) Book Value and Related Methods Focusing on Historic Cost and Income Information

Traditional valuation techniques rested mainly on the available historical information, which consisted of expenditures incurred: book value. 83 These techniques are accounting methods and are used in many fields (such as tax, inventory, and so forth). Their advantage is that they are the most objective available; they are based on data generated for purposes other than litigation; 84 as such, they are less likely to be infected by potential bias that could cloud the judgment of a party-appointed expert. 85 It is often suggested that their weakness lies in their limited relationship to how markets—capital markets, investors in companies—value assets; markets tend to see less the money spent, but in the main what income an asset is likely to produce in the future. 86 Additionally, it has been suggested that while they may be suitable under certain circumstances for valuing an enterprise, they may not be so suitable for compensating an investor for loss of a contract. 87

On the other hand, a prudent—and successful—investment method tends to rely much more on historical data than speculative future projections, which tend to be very easily manipulated, 88 subject to various biases and rapidly changing sentiments in markets (eg on risk). 89 The Buffet/Graham method of valuation—arguably at present the most successful long-term method of prudent valuation, for example,

end p.1072

uses—inter alia—larger ranges of historical data, in particular earnings. 90 It is therefore not correct to contrast the ‘old-fashioned and obsolete’ method of relying very much on historical data against an allegedly ‘modern’ method of relying mainly or exclusively on future earnings predictions. The methodological framework for compensation should aim at being as objective—not prone to bias and easy manipulation—and as prudent as possible. 91

Objective historical data should in case of doubt be given rather greater weight than inherently speculative forecasts. 92 A certain preference for the ‘objective’ rather than the ‘subjective’ approach can therefore be advocated. But one cannot be dogmatic about either of the two main approaches to valuation. A pure cost-based method requires at a minimum an updating commensurate with inflation; it will not yield a proper result where the investment had to overcome significant and unusual risks. For example, in petroleum exploration, most, very costly, exploration wells are dry; they get compensated by the few successful results of a drilling campaign. 93 This means that the value of the successful exploration is—often by a multiple—much more than the expenditures incurred. In essence, expenses have either to be multiplied by the exploration risk (historic method) or in this situation (and other comparable situations where a particular high risk is overcome) one needs to look at comparable transactions and forecasts of future income. A combination of historic cost (adjusted by exploration risk), future income, and market-value- based valuations is here called for. The issue of whether country and project political risk should also lead to a discount for value achieved solely by historic cost (because the investment is now exposed to risk for its future duration) has to be left open here. 94

end p.1073

(iii) Discounted Cash Flow (DCF) Method 95

More modern—but not necessarily definitively better or conclusive—valuation techniques look forward rather than backward. The ‘net present value of discounted future cash flow method’ (NPV of DCF), which is currently dominant, estimates future cash flows; it then applies a ‘discount rate’ reflecting cost of capital and risk (eg in simplistic terms a rate that incorporates a risk-adjusted industry—, country—, and company-specific rate of return) to estimated future income streams to identify the present value. That method has been widely used over three decades. 96 It expresses, in essence, the evident truth that income at present is worth more than uncertain income in the distant future.

Debates among international lawyers are largely focused on the appropriateness of the traditional historic-cost method (which is still the most objective and reliable) and this ‘modern’, future-oriented, and thus speculative method—which is now present in probably all valuation debates before tribunals. 97

The difficulty with this method—as compared to the historic-cost method—is that while it may look objective and scientific when presented by experts using spreadsheet models, it does not provide objective and predictable outcomes. The DCF method is in essence a speculation about the future dressed up in the appearance of mathematical equations. The inherent subjectivity present in most of the assumptions of the financial model explains why rational bankers and business executives put highly different values on the same asset. 98 Small, and continuously occurring,

end p.1074

mood-influenced changes in, for example, future income growth rates or appetites towards risk have a disproportionate impact on the present value determination. As stock markets demonstrate on a daily basis, valuation can change dramatically if the present attitude to an essentially unknowable future event changes, even if only slightly. Party-appointed experts can therefore easily and plausibly add or subtract from their valuation by taking greater or lesser account of uncertain events in the future. Simply maintaining an initial high growth rate for the future can multiply present value, while assuming a frequent levelling off for such growth rates, or a decline through such possible and often plausible factors as competition from new entrants, higher taxes, or regulatory change will depress present value. There is no reliable way of predicting such events several years down the line as we are unable even to predict the stock market of tomorrow.

The NPV/DCF method is not the received wisdom on valuation. Successful investors use historical data extensively. 99 Book value or other ‘conservative’ methods are therefore less obsolete than may appear from a too modern, but na?ve perspective. 100 The late Tom Stauffer argued convincingly that inflation-updated book value, market price, and NPV of DCF should be largely similar in well-functioning, and efficient markets. 101 In applying the current valuation techniques, in particular scenario-based NPV of DCF, it is essential to identify the key assumptions underlying the spreadsheet-based financial models. Arbitral tribunals, therefore, should ensure they understand not so much the difference in result (which is inevitable between claimant and respondent experts), but rather the difference in the assumptions where they have an impact on the outcome (‘sensitivity analysis’). 102 This worthwhile goal is very often achieved by the use of tribunal-appointed experts. 103 Knowing the right questions to ask is crucial.

(iv) Net Book Value (NBV) v Discounted Cash Flow (DCF)

The main dispute here has been between historic and cost-based approaches on the one hand, and more subjective methods (like DCF) on the other. 104 Tribunals

end p.1075

have been led to consider NPV of DCF valuation, 105 but they have been much more cautious than more enthusiastic proponents. Almost every tribunal now repeats the mantra that ‘speculative profits’ or ‘speculative elements’ should be discounted in valuation. 106 This—arguably firmly established—principle is manifestly contradictory to the application of NPV of DCF methodology, which relies entirely on speculation. It was compared by the late Professor Seidl-Hohenveldern to the ‘obscure 16th century prophecies of Nostradamus’. 107 The widespread principle that NPV of DCF methodology should only be used for well-established projects with a solid record of commercial performance 108 again reflects a justified reluctance on the part of tribunals to get involved in what are essentially competing prophecies of often equal plausibility. The prudent way to value assets is to rely not on one specific method, but to allow the use of all methods—updated book value relying on historic-cost data, valuation based on historic earnings over an extended period (5–10 years), 109 replacement value if practical, 110 NPV of DCF with the proviso that projections must be based on a historic performance figure of 5–10 years and projected into the future only to a point of reasonable certainty. There needs to be a cut-off point where the projection is too much extended and where risk and probability considerations prevail over tangible evidence. Assumptions over the conclusion of new contracts and extension of existing contracts and licences should therefore in most

end p.1076

cases—except if highly probable, if not certain and of right—be excluded from the NPV of DCF calculation with the objective of minimizing speculative elements. 111

(v) Appropriate Risk Factor/Discount Rate

Future income is not certain; it should therefore not only be discounted (to reflect the time value of money), but also risk adjusted. Risk adjustment should reflect, first, the probability of the occurrence of future events that are considered a risk, and, secondly, the detrimental effect and aversion in markets over an extended period with respect to the probability that risks materialize—from a ‘prudent investor’ perspective. If compensation is paid using a low discount rate for a project with high risk (long exposure to high risk), the claimant is essentially allowed to swap a high-risk and presumably high-return investment 112 for a low-risk and high-return investment from the time the investment is paid. That involves an element of double recovery. Tribunals and reputed writers are as a rule reluctant to accept unusual profitability rates as a basis for the valuation; 113 unusually high profitability projections indicate an unusual risk and thus suggest a commensurate discount rate. This is particularly so in an economic crisis where profitability is likely to be squeezed under the impact of the crisis (collapsing demand, ability to pay). 114

The proper approach is to identify the risk implications and factor them into the compensation model. That can be done either by a discount factor that incorporates both a loss-of-value for time (inflation element) and a risk factor that includes both a reasonable, industry-wide return over inflation plus a rate that represents

end p.1077

the special risks over the industry average of the project, and the country risk. This approach would mirror the way risky projects are evaluated at the investment phase: they have to exceed not only a rate reflecting a safe investment rate (eg US government long-term bonds) plus a rate that reflects the investor's own cost of capital (say another 3–10%), but also an additional premium rate reflecting the investment in the particular industry, country, and projects (risk-reward function). The discount rate applied should reflect that risk premium.

The best way for tribunals to handle this sensitive and difficult issue, which transcends the professional competence of arbitrators, is to insist that the parties use a range of valuation methods, including an inflation-adjusted historic method, a comparison with comparable benchmarkable transactions, and a prudent application of NPV/DCF (and similar methods) with a strong emphasis on using longer-term historical project performance, a cut-off point when income is no longer certain or highly probable, and a discount rate that reflects the risk assumed by the investor in comparison to safe-investment benchmarks. 115 The ultimate result should be in a range that is not too wide; if it is too wide, the parties have to explain the assumptions (price, cost, risk) underlying their models. After due examination, the tribunal should—itself or with the help of its own expert—determine the correct assumptions (inevitably with some margin of appreciation) 116 and adjust the model. Splitting the difference only makes sense if the results of the various approaches are positioned on a relatively narrow range and once excess values have been taken off. 117 This method is not only the most rational one but is also most likely to reflect the intuitive approach of most prudent tribunals. 118 Splitting the difference between ‘excess values’ would only encourage the parties to come up with ever more exorbitant compensation claims.

(vi) Special Case of Investment Projects before Start-up and without Extensive Record of Performance