Jump to Content Jump to Main Navigation
Signed in as:

Part III Substantive Rights, 9 Compensation

From: International Investment Arbitration: Substantive Principles (2nd Edition)

Campbell McLachlan, Laurence Shore, Matthew Weiniger

From: Oxford Public International Law (http://opil.ouplaw.com). (c) Oxford University Press, 2022. All Rights Reserved. Subscriber: null; date: 09 December 2022

Compensation — Damages — Remedies and costs — Discounted cash flow (DCF), anticipated future profits — Interest — Moral damages — Valuation

(p. 413) Compensation

A.  Introduction

9.01  The question of the amount of compensation payable for a State’s breach of its international obligations has been one of the most contentious areas of international law. Not surprisingly, this question generates much heat in the field of investment dispute arbitration, although the issues in dispute have moved on from those debated in the classical sources of public international law. The different stances taken as between investors and States are highlighted by the adversarial nature of the arbitration process.

9.02  The compensation debate in investment arbitration has proceeded from the same starting-point as the debate in public international law. This is the obligation upon the State (p. 414) committing the international wrong to make reparation by way of restitution or, if this is not possible, to pay monetary compensation for the loss sustained.1

9.03  In customary international law, the most heated areas of debate relate to (a) the appropriate standard of compensation for expropriation and (b) compensation for lost profits, in particular the applicability of the discounted cash flow (DCF) method of valuation. As the analysis in this chapter will demonstrate, for investment dispute arbitrations the question of the appropriate standard has been largely settled by the wording adopted in investment treaties and national investment laws. Similarly, it is accepted in recent tribunal awards that the appropriate way to value an income-producing asset is by using the method most applied in commercial life—the DCF method. Yet, even allowing for acceptance of the ‘full’ standard of compensation and (in appropriate cases) the DCF method of calculating losses, there remain many areas of current debate which can have a significant effect on the amount of compensation actually payable.2

B.  International Law Standards of Compensation for Expropriation

9.04  As stated in the chapter introduction, the question of compensation for expropriation has long been one of the most contentious areas of international law.3 There has been little agreement between those advocating standards which would favour capital-exporting States, and those pressing on behalf of capital-importing States. Traditionally, the view favourable to investors has been the standard expressed by the former US Secretary of State, Cordel Hull. He declared in correspondence with the Government of Mexico in 1938 that ‘under every rule of law and equity, no government is entitled to expropriate private property, for whatever purpose, without provision for prompt, adequate, and effective payment therefor’.4

9.05  The restrictive approach is expressed by concepts such as the Calvo doctrine, formulated by the Argentinian diplomat and jurist Carlos Calvo.5 His doctrine, which still remains part of many Latin American constitutions, provides that foreign States and foreign nationals should settle claims by submitting to the jurisdiction of local courts. Diplomatic or military pressure is not to be used.

9.06  Another limiting formula is that expressed in the United Nations General Assembly Resolution on Permanent Sovereignty over Natural Resources6 and its Charter of Economic Rights and Duties of States.7 These provide that the owner of nationalized or expropriated (p. 415) property shall be paid ‘appropriate’ compensation. ‘Appropriate’ compensation is a more subjective standard than a requirement that compensation be ‘adequate and effective’.

9.07  The International Law Commission’s ‘Draft Articles on the Responsibility of States for Internationally Wrongful Acts’ (‘Draft Articles on State Responsibility’) approach the question of compensation on an objective basis, although their wording differs from that of the Hull formula. They provide that a responsible State ‘is under an obligation to make full reparation for the injury caused by the internationally wrongful act’.8 Full reparation includes ‘an obligation to compensate for the damage caused thereby, insofar as such damage is not made good by restitution’.9

9.08  Another influential attempt at codifying the relevant principles in this area are the World Bank Guidelines on the Treatment of Foreign Direct Investment.10 These begin by talking of ‘appropriate compensation’11 but continue to state that compensation will be deemed appropriate ‘if it is adequate, effective and prompt’.12

9.09  Given the difference between the various standards, it is not possible to say that there is a consensus on the position in international law.13 However, this lack of consensus does not generally pose a difficulty in modern investment disputes. Almost all bilateral investment treaties (BITs) and multilateral investment treaties contain specific provisions on the standard of compensation. Not surprisingly, given their stated aims of promoting investment, the vast majority of these treaties follow the Hull formula and require compensation to be ‘prompt, adequate and effective’.

9.10  The almost universal adoption of this formula does not, however, eliminate all the uncertainties. A wide range of compensation approaches can be characterised as ‘adequate and effective’. For this reason, many BITs14 also provide further clarification by referring to ‘genuine value’, ‘market value’ or ‘fair market value’. A similar approach has been adopted in the major multilateral investment treaties. For example, art 13(1)(d) of the Energy Charter Treaty (ECT) provides that expropriation should be ‘accompanied by the payment of prompt, adequate and effective compensation’ and that ‘[s]uch compensation shall amount to the fair market value of the Investment expropriated …’15

(p. 416) 9.11  Given the uniformity of treaty provisions, in almost all treaty disputes the problem of identifying a standard for compensation does not arise as a practical issue given that the treaty itself will contain provisions stating the appropriate standard. Nonetheless, the mere fact that the treaty contains a definition of the standard of compensation does not make the task of determining the amount of compensation any easier.

9.12  BITs and multilateral investment treaties often state that ‘adequate and effective’ compensation is to be assessed on the basis of the ‘fair market’ value or ‘genuine’ value of the asset. In the case of an asset which is regularly traded, or has recently been traded, a tribunal may determine a ‘fair market’ value by looking at the price at which it was recently traded. However, more often a tribunal must determine a ‘fair’ market value in respect of a unique asset (such as a corporation holding a valuable concession) which the seller did not want to sell and for which no willing buyer is likely to appear following an expropriation. In his Commentary on the ILC’s Draft Articles on State Responsibility, Crawford states that tribunals valuing nationalised businesses should:

… examine the assets of the business, making allowance for goodwill and profitability as appropriate. This method has the advantage of grounding compensation as much as possible in some objective assessment of value linked to the tangible asset backing of the business. The value of goodwill and other indicators of profitability may be uncertain, unless derived from information provided by recent sale or acceptable arms-length offer. Yet, for profitable business entities where the whole is greater than the sum of the parts, compensation would be incomplete without paying due regard to such factors.16

9.13  Many of the difficulties in this area arise when tribunals assess the likely future profitability of expropriated enterprises. Tribunals can be reluctant to award compensation for lost profits as this can oblige an often impecunious State to pay a large lump sum to a wealthy foreign investor. The tensions between awarding the full measure of compensation and not wishing to award a sum that is so large that it looks (even if it is not) punitive rather than compensatory lie behind many of the more contentious awards analysed in the following sections.17

9.14  The best practice is for tribunals to calculate compensation taking into account all taxes to be paid to the host State or authorities. As a consequence, compensation should ultimately be paid net of host State tax.18

9.15  However, it would be wrong for tribunals to take account of the possibility of taxes being imposed in other jurisdictions and a claim to gross up the award on this basis was rejected as ‘speculative and uncertain’ in Venezuela Holdings v Venezuela.19

(p. 417) C.  Determining the Amount of Compensation

9.16  The exercise of quantifying the quantum20 is primarily an economic exercise. The tribunal is assessing the value of the taken property and the value in this context is its economic value. The investor will usually be seeking to recover compensation for the income the asset would have generated when used or consumed.21

Traditional methods of valuation

Liquidation value

9.17  Liquidation value means the amounts at which individual assets comprised in the enterprise or the entire assets of the enterprise could be sold under conditions of liquidation to a willing buyer, less any liabilities which the enterprise has to meet.

9.18  The World Bank Guidelines suggest that this method is most appropriate for an enterprise which, not being a proven going concern, demonstrates a lack of profitability.22 This test reflects the common-sense proposition that an owner of commercial assets which cannot be put to profitable use would be acting more rationally by selling those assets.

Replacement value

9.19  Replacement value means the cash amount required to replace the individual assets of the enterprise in their actual state as of the date of the taking.

9.20  The replacement value method is rarely applicable in the case of governmental takings. It can only be employed when the investor can purchase replacement assets identical to the ones taken. It is rarely the case that individual discrete assets are taken by the State. In addition, the investor’s assets usually possess individual unique features rendering an assessment of replacement value impossible.

Book value

9.21  Book value is the difference between the enterprise’s assets and liabilities as recorded on its financial statements or the amount at which the taken tangible assets appear on the balance sheet of the enterprise, representing their cost after deducting accumulative depreciation in accordance with generally accepted accounting principles.

9.22  Book value is sometimes seen as a means of returning the investment to the investor without allowing any sums for lost profits. It is thus sometimes justified as a way of keeping compensation awards within reasonable (from the State’s point of view) bounds.23

9.23  However, the use of book value in quantifying compensation has been cogently criticized by various commentators.24 The principle objections are:

  1. (1)  The value of income-producing assets depends on the cash that they are expected to generate in the future. It would thus constitute economic nonsense to distinguish between (p. 418) the value of the assets and the profits or revenues that the assets would have generated in the future. Book value merely measures value as recorded on a balance sheet. It has nothing to do with the value that a businessman would ascribe to an economic venture, which is related to its ability to generate profit.

  2. (2)  The book value of an enterprise cannot measure with certainty the value of that enterprise’s contractual rights, know-how, goodwill, and management skills. Tribunals should not be deterred from evaluating these matters just because valuing them will be difficult.

9.24  A treaty arbitration where circumstances made it appropriate for the Tribunal to adopt the book value method was AAPL v Sri Lanka.25 The State’s failure in this case constituted a breach of its duty to provide full protection and security to the investment. Armed insurgents destroyed a shrimp farm which was the only asset of a company. The investor had a 48 per cent shareholding in the company. The Tribunal refused to compensate the investor for the lost profits claimed on the basis that: ‘A reasonable prospective purchaser would … be at least doubtful about the ability of the Company’s balance sheet to cease being in the red, in the sense that the future earnings become effectively sufficient to off-set the past losses as well as to service the loans which exceed in their magnitude the Company’s capital assets.’26

9.25  Having stated its belief that there was no possibility of profits for which the investor could be compensated, the Tribunal drew up a ‘comprehensive balance sheet’27 reflecting the company’s assets and liabilities. This balance sheet was drawn up on the basis of the company’s ‘tangible assets’.28 Although some commentators have suggested that the Tribunal was using the replacement value approach by looking at the company’s balance sheet which would record items at the price at which they were purchased,29 it is more likely, from its reliance on an accountant’s report,30 that the Tribunal was measuring the company’s book value in accordance with standard accounting principles.

Discounted cash flow value

9.26  Discounted cash flow (DCF) value means:

… the cash receipts realistically expected from the enterprise in each future year of its economic life as reasonably projected minus that year’s expected cash expenditure, after discounting this net cash flow for each year by a factor which reflects the time value of money, expected inflation, and the risk associated with such cash flow under realistic circumstances. Such discount rate may be measured by examining the rate of return available in the same market on alternative investments of comparable risk on the basis of their present value.31

(p. 419) 9.27  This valuation method generates the most controversy, generally because the sums involved can be so enormous. For example, in Occidental v Ecuador32 the Tribunal awarded compensation of US$1,769,625,000 following Ecuador’s cancellation of Occidental’s participation rights in a hydrocarbon block. The Tribunal looked to the DCF method to calculate the money the claimant would have reasonably been expected to earn but for the cancellation. It defined the methodology in the following terms:

The Tribunal is of the view that, in this case, the standard economic approach to measuring the fair market value today of a stream of net revenues (i.e., gross revenues minus attendant costs) that can be earned from the operation of a multi-year project such as OEPC’s development of Block 15 is the calculation of the present value, as of 16 May 2006, of the net benefits, or; ‘discounted cash flows’. These net cash flows are appropriately determined by calculating the flow of benefits (‘cash flows’) that the Claimants would have reasonably been expected to earn in the ‘but for’ state of the world in which the termination of the Participation Contract hypothetically did not occur relative to the actual cash flow that the Claimants will derive subsequent to the termination. The difference between these two cash flow streams (the ‘but for’ state of the world with no termination less the actual state of the world with contract termination), discounted to the date of the actual contract termination, is the economically appropriate and reliable measure of the cumulative economic harm suffered by the Claimants as a consequence of the contract termination.33

9.28  The World Bank Guidelines identify the DCF method as a reasonable methodology for determining the market value of a going concern with a proven record of profitability.34

9.29  The Report to the Development Committee which accompanies the World Bank Guidelines emphasises some of the problems with using the DCF valuation method:

However, particular caution should be observed in applying this method as experience shows that investors tend to greatly exaggerate their claims of compensation for lost future profits. Compensation under this method is not appropriate for speculative or indeterminate damage, or for alleged profits which cannot legitimately accrue under the laws and regulations of the host country.35

9.30  Notwithstanding its difficulties, the DCF method is almost universally used and accepted by both the business and academic community in valuing income-producing assets.36 The value of an income-producing capital asset can only be ascertained by valuing the cash the asset is expected to generate in the future. The DCF method is thus appropriate because it is designed to calculate the value on one specified date of cash flows that are to be received at different times.37

(p. 420) Other valuation methods

9.31  In addition to the four methods set out in the World Bank Guidelines, other methods have also been proposed. For example, the Tribunal in CMS Gas Transmission Co v Argentina38 also mentioned the ‘comparable transaction’ approach, which reviews comparable transactions in similar circumstances, and the ‘option’ approach. The ‘option’ approach studies the alternative uses which could be made of the assets in question, and their costs and benefits.39 The Tribunal in CME Czech Republic BV v Czech Republic40 relied mainly on the ‘comparable transaction’ approach. Evidence of other contemporaneous valuations of a taken asset, especially those made between willing buyers and willing sellers, will almost always be persuasive. However, such valuations are rarely available.

9.32  The Tribunal in Azurix v Argentina41 concluded that it was appropriate to use the monetary value of the actual investments made in ascertaining fair market value to compensate for a breach of the fair and equitable treatment standard. This was perceived to be appropriate because the investment had been made by public tender and was thus ‘recent and highly ascertainable’.42 Nonetheless, the Tribunal calculated the investment value with a degree of flexibility. For example, it discounted the up-front payment made by the investor in order to win the tender from US$439 million to US$60 million because of a belief that the investor had overpaid: ‘no more than a fraction of the [payment] could realistically have been recuperated under the existing Concession Agreement’.43

9.33  The above description does not purport to be a comprehensive description of the possible bases of valuation. In investment treaty arbitration parties generally retain accounting and economist experts to assist with presenting the compensation part of the case and designing the appropriate models. The lawyer’s role is to ensure that the models presented fall within the accepted legal guidelines, which are presented in this chapter. Legal works which address the accounting issues in greater detail include Calculation of Compensation and Damages in International Investment Law44 and Valuation for Arbitration.45

9.34  The next section addresses the basic legal principles emerging from judicial and arbitral decisions on awarding lost profits by way of compensation for international wrongs committed by States.

9.35  By way of introduction, it is important to understand that the role of a tribunal in calculating compensation is a matter for its informed estimation in the light of all the evidence available to it rather than something that can be resolved simply on the basis of the burden of proof. The tribunal must be satisfied that the claimant has suffered some damage as a result of the respondent’s breach but the amount of that damage is for the tribunal to determine rather than for the claimant to prove. The quantification of damages is an inherently uncertain (p. 421) exercise but this uncertainty is not a reason for a tribunal to decline to award damages at all. It is usual for a tribunal to be assisted by the parties’ respective expert reports and tribunals usually base their assessments upon a combination of various factors including expert reports, different bases of valuation and contemporaneous evidence of valuations undertaken prior to the commencement of arbitration. A tribunal is also able to consider events subsequent to the date of the breach if these provide useful clarification of the value that would have applied at the date of injury. In making its assessment, a tribunal is not bound to a binary choice between two rival expert reports.46

Tribunals awarding compensation on the basis of DCF value

9.36  One of the earliest decisions regularly cited in support of tribunals awarding sums to compensate for lost profits is the judgment of the Permanent Court of International Justice (PCIJ) in the Factory at Chorzów case.47 This was a State-to-State claim brought by Germany against the Polish Republic for reparation under the 1922 Convention Concerning Upper Silesia to compensate for the Polish Government’s taking of a factory to which German companies had rights.

9.37  The Court’s analysis commenced with the broad observation that ‘it is a principle of international law, and even a general conception of law, that any breach of an engagement involves an obligation to make reparation’.48

9.38  Such reparation was stated to be equivalent to: ‘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.’49

9.39  The Court then focused its attention on determining the value of the rights lost: ‘The whole damage suffered by the one or the other Company as the result of dispossession, in so far as concerns the cessation of the working and the loss of profit which would have accrued, is determined by the value of the undertaking as such …’50

9.40  In its order, the Court formulated certain questions to be addressed by a valuation expert. It instructed the expert to assess compensation for future profits by explaining that the purpose of its first question was, ‘to determine the monetary value, both of the object which should have been restored in kind and of the additional damage, on the basis of the estimated value of the undertaking including stocks at the moment of taking possession by the Polish Government, together with any probable profit that would have accrued to the undertaking between the date of taking possession and that of the expert opinion’.51

(p. 422) 9.41  The Chorzów Factory judgment was analysed in many of the decisions of the Iran–US Claims Tribunal which dealt with compensation. The earliest Iran–US Claims Tribunal decision awarding compensation on the basis of a DCF analysis was Starrett Housing International, Inc v Iran.52 This claim arose out of the Iranian Government’s expropriation of Starrett’s rights in a project for the construction of apartments near Tehran. The Tribunal expressly cited Chorzów Factory in appointing experts to ‘ascertain “the estimated value of the undertaking …” ’.53

9.42  The Tribunal specifically commended the DCF approach as a ‘logical and appropriate’54 method for determining fair market value. Similarly, in Phillips Petroleum Co Iran v Iran,55 a case arising out of the taking of the claimant’s rights under a joint structure agreement to explore and exploit petroleum reserves in the Persian Gulf, the Iran–US Claims Tribunal held that a fair market valuation would be calculated on the basis of the DCF method. The Tribunal characterised a DCF analysis not simply as ‘a request to be awarded lost future profits, but rather as a relevant factor to be considered in the determination of the fair market value of its property interest at the date of taking’.56

9.43  However, a notable example of the Iran–US Claims Tribunal refusing to adopt the DCF approach is the judgment in Amoco International Finance Corp v Iran.57 This case arose out of the expropriation of the claimants’ interest in a joint venture company, called Khemco, for the production and marketing of natural gas. Khemco had been active over many years and at the time of the expropriation was a going concern. The claimant requested compensation to be assessed on a DCF basis. This was rejected by the Tribunal which, nonetheless, said that it was basing its approach on the principles set out by the PCIJ in Chorzów Factory. The Tribunal construed Chorzów Factory as separating the value of the undertaking on the date of the taking (damnum emergens) from the lost profits (lucrum cessans).

9.44  The Tribunal’s approach to Chorzów Factory cannot be justified as a correct reading of that judgment. It is also based on flawed economic reasoning. The expected profitability of a going concern is a fundamental part of its value and the two aspects cannot be separated.58 This was understood by the PCIJ in Chorzów Factory. Its instructions to the expert had asked him to estimate ‘future prospects’ in determining the value of the undertaking. Further, in explaining the questions addressed to the expert, the Court expressly stated that the expert should determine ‘future prospects’ as part of the value of the undertaking including the lands, buildings, equipment, stocks and processes at its disposal, supply and delivery contracts, goodwill, and future prospects.59

(p. 423) 9.45  Any Tribunal rejecting the DCF methodology in an appropriate case would be moving away from the best practice of business and economics. This was understood by the Iran–US Claims Tribunal in Phillips Petroleum. In finding in favour of a DCF analysis it stated that ‘a prospective buyer of the asset would almost certainly undertake such DCF analysis to help it determine the price it would be willing to pay and that DCF calculations are, therefore, evidence the tribunal is justified in considering in reaching its decision on value’.60

9.46  The DCF method was adopted on a number of occasions by ICSID tribunals hearing contract disputes. An example is Amco Asia Corp v Indonesia.61 The claim arose out of a thirty-year lease of a hotel in Indonesia which was expropriated after eleven years. The Tribunal found that Amco was entitled to compensation for loss of the right to manage the hotel. It applied both Indonesian and international law to determine the amount of compensation.

9.47  In making its compensation award, the Tribunal recommended the DCF method as being ‘one that is logically indicated by the finding that the purpose of compensation is to put Amco in the position of having received the benefits of the contract’.62

9.48  It further stated that the DCF method was not speculative in its range of application, rather the tribunal found merit in its flexibility: ‘The method itself relies on the application of assumptions which are necessarily judgmental. The DCF method is at once a flexible tool, that allows for an application of factors and elements judged as relevant. At the same time, it allows for the application of these judgmental elements to be articulated.’63

9.49  Substantial sums by way of compensation for lost profits were also awarded by ICSID tribunals in SOABI v Senegal64 and LETCO v Liberia.65 SOABI v Senegal is unusual because the Tribunal, albeit applying the law of Senegal rather than international law, awarded compensation for lost profits although the claimant had not progressed to the stage of being able to demonstrate a proven record of profitability.

9.50  The award in CMS Gas Transmission Co v Argentina66 is an example of a tribunal using the DCF methodology in a treaty arbitration. This was not an expropriation case, but a case of a State’s failure to afford an investment fair and equitable treatment. Nonetheless, the Tribunal decided that it would be appropriate to use the expropriation standard of compensation due to the claimant having suffered ‘important long-term losses’.67 The Tribunal considered a number of different valuation approaches and concluded in favour of the DCF method: (p. 424) ‘the Tribunal … has no hesitation in endorsing [the DCF method] as the one which is the most appropriate in this case. [The company in which the claimant had invested] was and is a going concern; DCF techniques have been universally adopted, including by numerous arbitral tribunals, as an appropriate method for valuing business assets … there is adequate data to make a rational DCF valuation …’.68

9.51  CME Czech Republic BV v Czech Republic69 is a further example of a tribunal considering the DCF methodology in a treaty arbitration. In this award, the Tribunal was assessing the compensation due to a Dutch corporation following an earlier finding that the Czech Republic had violated investment protection provisions of the Czech–Netherlands BIT.70 The asset affected was a television broadcasting licence.

9.52  Article 5 of the BIT provided that:

Neither Contracting Party shall take any measures depriving, directly or indirectly, investors of the other Contracting Party of their investments unless the following conditions are complied with:

  1. c)  the measures are accompanied by provision for the payment of just compensation. Such compensation shall represent the genuine value of the investments affected …

9.53  The Tribunal read the inferences to ‘just compensation’ and ‘genuine value’ in the BIT as an invocation of the Hull formula. It recited the history of the controversy over the appropriate standard for compensation in public international law but concluded that ‘in the end, the international community put aside this controversy, surmounting it by the conclusion of more than 2200 bilateral (and a few multilateral) investment treaties. Today these treaties are truly universal in their reach and essential provisions. They concordantly provide for payment of “just compensation”, representing the “genuine” or “fair market” value of the property taken.’71

9.54  The Tribunal calculated the amount of compensation by reference to a previous transaction entered into between the claimant and a third-party purchaser on the basis of arm’s length negotiations,72 as well as by considering the DCF valuations of the parties’ experts. Ultimately, the Tribunal did not decide between the opposing DCF valuations, as they could best serve as confirmation of its findings based upon the third-party transaction. Similarly, in Enron v Argentina the Tribunal made reference to stock exchange value and contemporaneous share sales to cross-check the reasonableness of the expert’s DCF (p. 425) calculations.73 Tribunals using the DCF method have to grapple with complex models and choose between different valuation factors, many of which are necessarily speculative. Yet the inherent difficulties do not make the method unsuitable and tribunals should be ready to proceed despite the difficulties involved in projecting. For example, in Cargill v Mexico, an award notable for the depth of its compensation analysis, the Tribunal said the following:

The Tribunal acknowledges Respondent’s concerns about the difficulties in projecting the overall market for HFCS, Claimant’s market share, and the appropriate price of and demand for HFCS in light of Claimant’s four-year absence from a competitive market. The Tribunal does not find these projections, however, to be so unusual or difficult that employment of the method is inappropriate in this proceeding.74

9.55  In Lemire v Ukraine, the Tribunal adopted a similar approach. Once it became convinced that the claimant had suffered a loss, it did not require certainty in the calculation of damages:

While the existence of damage is certain, calculating the precise amount of the compensation is fraught with much more difficulty, inherent in the very nature of the ‘but for’ hypothesis. Valuation is not an exact science. The Tribunal has no crystal ball and cannot claim to know what would have happened under a hypothesis of no breach; the best any tribunal can do is to make an informed and conscientious evaluation, taking into account all the relevant circumstances of the case, not unlike that made by anyone who assesses the value of a business on the basis of its likely future earnings.75

The issues surrounding the measure of compensation for breaches of the FET standard are—as the Tribunal in LG & E said—‘particularly thorny’.76 To estimate the damages, the Tribunal will inevitably have to accept certain assumptions. These assumptions can and must be checked, applying tests of reasonableness. But in the end, there is no denying that the calculation of damages in a case like this inevitably requires a certain amount of conjecture as to how things would have evolved ‘but for’ the actual behaviour of the parties. This difficulty in calculation cannot, however, deprive an investor, who has suffered injury, from his fundamental right to see his losses redressed.77

Tribunals not adopting a DCF approach

9.56  While many tribunals have applied the DCF method, other tribunals have refused to apply it. In his Commentary to the ILC’s Draft Articles on State Responsibility, Crawford discusses some of this jurisprudence stating that:

The discounted cash flow (DCF) method has gained some favour, especially in the context of calculations involving income over a limited duration, as in the case of wasting assets. Although developed as a tool for assessing commercial value, it can also be useful in the context of calculating value for compensation purposes. But difficulties can arise in the application of (p. 426) the DCF method to establish capital value in the compensation context. The method analyses a wide range of inherently speculative elements, some of which have a significant impact upon the outcome (e.g. discount rates, currency fluctuations, inflation figures, commodity prices, interest rates and other commercial risks). This has led tribunals to adopt a cautious approach to the use of the method.78

9.57  A number of tribunals have refused to award DCF-based compensation to enterprises lacking a proven record of profitability.79 These awards do not represent any dissatisfaction with DCF methodology as a whole and are consistent with the World Bank Guidelines, which limit DCF awards to going concerns with a proven record of profitability.

Enterprises lacking a proven record of profitability

9.58  An example of such an award is the ICSID Additional Facility award in Tecnicas Medioambientales Tecmed SA v Mexico.80 This arbitration arose out of the expropriation by Mexico of a landfill site operated by a company controlled by Spaniards. The claim was brought under the BIT between Spain and Mexico. The Tribunal considered the DCF compensation claim submitted by the claimant but rejected the DCF approach. It noted the difference between the claimant’s investment, US$4 million, and the US$52 million compensation sought. In rejecting the DCF method, it stated that:

The non-relevance of the brief history of operation of the Landfill by Cytrar—a little more than two years—and the difficulties in obtaining objective data allowing for application of the discounted cash flow method on the basis of estimates for a protracted future, not less than 15 years, together with the fact that such future cash flow also depends upon investments to be made—building of seven additional cells—in the long term, lead the Arbitral Tribunal to disregard such methodology to determine the relief to be awarded to the Claimant.81

9.59  In the second Vivendi v Argentina award, the Tribunal concluded that Argentina had expropriated the claimant’s water concession. The claimants sought compensation based upon lost profits and pointed to the fact that their provision of water and sewage services was a going concern by the time of the arbitration. Yet the Tribunal noted that the claimant’s concession never made a profit and faced significant challenges. As a result it refused to apply a DCF analysis:

The Tribunal accepts, in principle, that fair market value may be determined with reference to future lost profits in an appropriate case. Indeed, theoretically, it may even be the preferred method of calculating damages in cases involving the appropriation of or fundamental impairment of going concerns. However, the net present value provided by a DCF analysis is not always appropriate and becomes less so as the assumptions and projections become increasingly speculative. And, as Respondent points out, many international tribunals have stated that an award based on future profits is not appropriate unless the relevant enterprise is profitable and has operated for a sufficient period to establish its performance record. The Tribunal notes that even in the authorities relied on by Claimants, compensation for lost (p. 427) profits is generally awarded only where future profitability can be established (the fact of profitability as opposed to the amount) with some level of certainty.82

9.60  A similar approach was taken by the NAFTA Tribunal in Metalclad Corporation v Mexico,83 the ICSID Tribunal in Wena Hotels v Egypt84 (a BIT case), and Biloune v Ghana Investments Centre85 (an award arising out of a concession agreement).86

9.61  An exception to this principle is the award in Rumeli Telekom v Kazakhstan.87 Kazakhstan was adjudged to have expropriated the claimant’s shares in a mobile telephone venture. Kazakhstan’s quantum expert argued that the DCF approach would not be appropriate because the expropriated venture was not a going concern.88 The Tribunal looked to calculate the fair market value of the claimant’s shares as of the date of the taking. It accepted that the venture was insolvent as of that date, but it noted that the only asset of real value was its mobile telecommunication licence. This would have a value far in excess of its book value and ‘[s]ince the value of that asset was directly linked to its potential to produce future income, there is no realistic alternative to using the DCF method to ascribe a value to it.’89 Similarly, in Al-Bahloul v Tajikistan,90 the Tribunal would have been willing to consider a DCF calculation to assess the value of a hydrocarbons permit:

The determination of the future cash flow from the exploitation of hydrocarbon reserves need not depend on a past record of profitability. There are numerous hydrocarbon reserves around the world, and sufficient data allowing for future cash flow projections should be available to allow a DCF-calculation.91

(p. 428) 9.62  In LG&E v Argentina92 the claimant argued that Argentina’s abrogation of the basic guarantees of its gas tariff regime constituted a continuous breach which would exist for as long as it held shares in its Argentinian subsidiary. However, the Tribunal was not willing to award the claimant losses arising in the future from the continuing nature of Argentina’s breach. These were seen as too uncertain and speculative.93

Awarding a reasonable equitable indemnification

9.63  In Libyan American Oil Co (LIAMCO) v Libya94 the sole arbitrator considered whether compensation should be paid for lost profits. The arbitration arose out of Libya’s expropriation of various petroleum concession agreements in 1973–74. In reviewing the practice of various tribunals, the sole arbitrator noted the many different approaches and quoted the ICJ, saying that ‘… the practice has been so much influenced by considerations of political expediency in the various cases, that it is not possible to discern in all this any constant and uniform usage, accepted as law, with regard to the alleged rule …’.95

9.64  The sole arbitrator awarded a substantial sum (US$66 million) as compensation for lost profits on the basis that the claimant should receive ‘a reasonable equitable indemnification’.96 However, the sole arbitrator did not provide any explanation of the rationale leading him to award this sum.

9.65  In the BIT-era, tribunals generally do not adopt figures without any justifying rationale. However, in certain cases tribunals have made significant approximations or made broad conclusions in arriving at figures. This is an understandable approach when tribunals are balancing a number of factors. The Tribunal in Siag v Egypt97 faced a difficult valuation exercise. It had to value the claimant’s losses following an expropriation of a tourist resort project. The resort was expropriated at an early stage in its construction and a DCF-based lost profits calculation was seen as inappropriate. The Tribunal adopted a Comparable Sales Valuation methodology even though there were few comparable properties. The expert said that his valuation could be wrong by 10 per cent in either party’s favour. The Tribunal used this concession to conclude ‘in all the circumstances’ the expert’s valuation should be discounted by 20 per cent in the respondent’s favour.98 Similarly, in ADC v Hungary the Tribunal said that it had to ‘stand back and look at the work product and arrive at a figure with which it is comfortable in all the circumstances of the case.’99

Compensation based upon legitimate expectations

9.66  Another award in which the Tribunal rejected the DCF approach put forward by the claimants and substituted other criteria is Kuwait v The American Independent Oil Co (AMINOIL).100 This arbitration arose out of Kuwait’s termination of an agreement between (p. 429) itself and AMINOIL for exploration and exploitation of petroleum and natural gas. The concession had been in place since 1948 and was due to run for many years.

9.67  As with LIAMCO, it is difficult to ascertain a precise methodology towards calculation from the award. The claimant asked for compensation based on two separate methods, each of which would have produced a capital sum based on total anticipated profits discounted to find present value. The Tribunal agreed that both methods were ‘acceptable’ but introduced other criteria into its discussion of the appropriate principles. One of these was that the compensation should reflect the parties ‘legitimate expectations’ on the basis that:

… with reference to every long-term contract, especially such as involve an important investment, there must necessarily be economic calculations, and the weighing-up of rights and obligations, of chances and risks, constituting the contractual equilibrium. This equilibrium cannot be neglected—neither when it is a question of proceeding to necessary adaptations during the course of the contract, nor when it is a question of awarding compensation. It is in this fundamental equilibrium that the very essence of the contract consists.101

9.68  The legitimate expectation of the parties was expressed as being equivalent to a ‘reasonable rate of return’. However, the Tribunal, like the Amoco Tribunal some years later, also divided up the value of the going concern into, ‘on the one hand of the undertaking itself, as a source of profit, and on the other of the totality of the assets, and adding together the results obtained’.102

9.69  In assessing its final sum the Tribunal stated that ‘having regard’ to all of the reasons set out in its discussion of compensation, but without giving any breakdown of its valuation, it would award AMINOIL US$179,750,764.

9.70  Introducing a search for the parties’ ‘legitimate expectations’ at the time of contracting into a compensation analysis does not solve any problems. It merely moves the debate between the investor and the host State to a different area. If ‘legitimate expectations’ are the criteria, an investor could argue that its legitimate expectations were that it would be able to realise its anticipated profit over the life of the concession. In contrast, the State would argue that it did not enter into a concession or a treaty to encourage and protect investments only to find that its policy-making freedom is so constrained that it must pay enormous sums by way of compensation having re-taken control over a particular area of its economy.

9.71  The issue of legitimate expectations was discussed by Brownlie in his Separate Opinion in CME Czech Republic BV v Czech Republic.103 Although Brownlie did not dissent from the Award on the basis that the sums awarded were ‘on [their] own terms, moderate’,104 his Separate Opinion contains a detailed analysis of the reasons why large sums should not be awarded to compensate for breaches of BITs. He based his analysis upon an assessment of the nature of a BIT stating that:

The [BIT] is not simply a vehicle for an arbitration clause. It is an Agreement on encouragement and reciprocal protection of investments. It is not a treaty for the protection of foreign property within the territory of the Czech Republic … In this context, it is simply unacceptable (p. 430) to insist that the subject-matter is exclusively ‘commercial’ in character or that the interests in issue are, more or less, only those of the investor … [nor] [i]s it reasonable to suppose that, when a State like the Czech Republic, with a deregulated sector of its economy, accepts foreign investment, it is accepting the risk of national economic disaster … The Czech Republic should have the benefit of civilised modern standards in the treatment of States. Even States which have been held responsible for wars of aggression and crimes against humanity are not subjected to economic ruin …105

Award limited by abuse of rights doctrine

9.72  A novel approach to the question of DCF valuation was taken by the Tribunal in Himpurna California Energy Ltd v PT (Persero) Perusahaan Listruik Negara (PLN).106 Himpurna brought a claim against the Indonesian State Electricity Corporation following PLN’s repudiation of a thirty-year Energy Sales Contract. Himpurna’s DCF-based calculation of its damages amounted to US$2.3 billion.

9.73  The Tribunal refused to award such a high sum, notwithstanding that it largely arose out of a straightforward application of the contract’s terms. The Tribunal would not turn the contract into ‘an astonishing bargain in circumstances when performance … would be ruinous’.107 It took refuge in the concept of abuse of rights, stating that the doctrine must be applied ‘to prevent the Claimant’s undoubtedly legitimate right from being extended beyond tolerable norms’.

9.74  One tribunal member dissented from the majority’s novel use of the doctrine of abuse of rights, refusing to apply it in a situation where malicious intent or lack of good faith could not be demonstrated. It is possible that the Tribunal could have achieved the same result by resort to the flexibility inherent within the DCF approach, without resorting to abuse of rights.108

Other reasons not to adopt a DCF approach

9.75  As indicated above, investment treaty tribunals are willing to use the DCF approach in suitable cases. In fact, tribunals do not presently tend to analyse in depth why they are using the DCF approach even where, as in CMS v Argentina,109 the Tribunal records the respondent’s arguments for the use of other valuation methods. It is not possible to ascertain from recent awards whether respondents have been attempting to resurrect other theories of international law to persuade tribunals to move away from valuation-based approaches in order to reduce compensation awards. Some of these theories, which continue to have academic support, are considered in the following paragraphs. Given the lack of a formal role of judicial precedent in investment arbitrations, there is no reason why respondents should not try to re-open the compensation debate on the basis of these doctrines, or the ones analysed above.

(p. 431) The distinction between lawful and unlawful takings

9.76  In most cases the amount of compensation will be the same irrespective of whether the taking is lawful or unlawful. It is notable that Crawford, in his Commentary to the ILC Draft Articles on State Responsibility,110 makes no distinction between lawful and unlawful takings in his analysis of the compensation payable for an internationally wrongful act. By contrast, Bowett has argued for the distinction, concluding ‘that the fundamental distinction between the lawful and unlawful taking has its most important consequence. For the correct principle is believed to be that loss of future profits, whilst a legitimate head of general damages for an unlawful act, is not an appropriate head of compensation for a lawful taking’.111 In the first edition of this book it was suggested that ‘ … in the BIT generation, tribunals no longer seem to consider whether a taking is lawful or unlawful’.112 However, this conclusion cannot be supported in the light of many of the valuation decisions that have come out since the publication of the first edition. The issue arose most starkly in ADC v Hungary, a case arising out of the unlawful expropriation of an airport concession. The relevant treaty was the Hungary–Cyprus BIT, art 4, which provided that:

  1. 1.  Neither Contracting Party shall take any measures depriving, directly or indirectly, investors of the other Contracting Party of their investments unless the following conditions are complied with:

    1. (a)  The measures are taken in the public interest and under due process of law;

    2. (b)  The measures are not discriminatory;

    3. (c)  The measures are accompanied by provision for the payment of just compensation.

  2. 2.  The amount of compensation must correspond to the market value of the expropriated investment at the moment of the expropriation.

  3. 3.  The amount of this compensation may be estimated according to the laws and regulations of the country where the expropriation is made.

  4. 4.  The compensation must be paid without undue delay upon completion of the legal expropriation procedure, but not later than three months upon completion of this procedure and shall be transferred in the currency in which the investment is made. In the event of delays beyond the three-months’ period, the Contracting Party concern shall be liable to the payment of interest based on prevailing rates.

9.77  This standard was arguably relatively limited, because it referred to ‘just compensation’ (this could allow for arbitrary considerations of what was just) and referred to Hungarian law. Although the respondent asserted that Hungarian law provided for expropriation to be accompanied by ‘full, unconditional and prompt compensation’,113 the Tribunal concluded that the BIT standard applied only to lawful expropriation. Accordingly, in the event of any unlawful expropriation, the Tribunal would be required to apply the default standard contained in customary international law. This was held to be the Chorzów Factory standard and the Tribunal considered this in detail.114

(p. 432) 9.78  The issue was particularly relevant because the value of the expropriated property had risen considerably between the date of the expropriation and the date of the award. This is not usually the case. The PCIJ in Chorzów Factory had stated that damages were not necessarily limited to the value of the undertaking at the moment of dispossession, and accordingly, by applying this standard, the Tribunal could award compensation on the basis of the higher present day value.

9.79  The Tribunal in Vivendi II adopted a similar approach. The treaty standard called for:

[T]he payment of a prompt and adequate compensation the amount of which, calculated in accordance with the real value of the investments in question, shall be assessed on the basis of the normal economic situation prior to any threat of dispossession.115

9.80  The Vivendi II Tribunal concluded that this treaty standard only applied to lawful expropriations and did not purport to set a standard for unlawful expropriation. Accordingly, the Tribunal felt free to cite the Chorzów Factory standard and conclude that:

… it is generally accepted today that, regardless of the type of investment, and regardless of the nature of the illegitimate measure, the level of damages awarded in international investment arbitration is supposed to be sufficient to compensate the affected party fully and to eliminate the consequences of the state’s action.116

9.81  On this basis, the Tribunal felt able to award compensation based upon a fair market value analysis established by a lost profit analysis.117

9.82  Another case where the distinction between the legal and illegal standard made a difference was Siemens v Argentina. The Tribunal again made a comparison between the treaty compensation formula and that laid out in the Chorzów Factory case. It concluded that ‘[u]nder customary international law, Siemens is entitled not just to the value of its enterprise as of May 18, 2001, the date of expropriation, but also to any greater value that enterprise has gained up to the date of this Award, plus any consequential damages.’118 In practice, this meant Siemens could recover its damages in US dollars rather than Argentinian pesos on the basis that only a dollar award would serve to wipe out the consequences of the illegal act.119 Further, Siemens could also recover certain post-expropriation costs incurred in continuing a skeleton operation. Again, the Tribunal believed that this claim was justified in order to wipe out the consequences of the illegal expropriation.120

9.83  The principles applied in the cases where the compensation standard differed because the Tribunal had ruled the expropriation illegal were neatly summarised in Funnekotter v Zimbabwe.121 The Tribunal made reference to two awards of the Iran–United States Claims (p. 433) Tribunal, namely Amoco122 and Phillips Petroleum123 which dealt with the difference between lawful and unlawful takings. In the words of the Funnekotter Tribunal:

As the Iran-United States Claims Tribunal rightly observed in the Amoco case, ‘obviously, the value of an expropriated enterprise does not vary according to the lawfulness or the unlawfulness of the taking … the difference is that, if the taking is lawful the value of the undertaking at the time of the dispossession is the measure and the limit of the compensation, while if it is unlawful, this value is or may be, only a part of the reparation to be paid.’ In general, as the same tribunal stated in the Phillips Petroleum case, ‘the lawful/unlawful taking distinction … is relevant only to two possible issues: where the restitution of the property can be awarded and whether compensation can be awarded for increase of the value of the property between the date of the taking and the date of the judicial or arbitral decision awarding compensation.’124

9.84  It is also useful to compare the practice of investment tribunals with academic writings. For example, Reisman and Sloane contend that in assessing compensation for an indirect expropriation (where expropriatory consequences result despite the absence of an expropriatory decree)125 an award of lost profits would amount to a punitive award and thus one without a basis in customary international law.126 While Brownlie in his separate opinion on quantum in CME v Czech Republic did not seek to make any differentiation based upon legal and illegal expropriations, he does distinguish between lawful and unlawful expropriation in his textbook.127

9.85  The usually accepted criteria for a lawful taking are that it should be taken in the public interest and under a due process of law, non-discriminatory and accompanied by provision for the payment of just compensation. The takings in ADC v Hungary and Vivendi II clearly did not purport to be made under any sort of due process of law. However, most expropriation cases that come before investment treaty tribunals will arise out of circumstances where the takings have not been accompanied by compensation. As a result, the expropriation cannot be said to have fulfilled all the criteria of a legal expropriation. The Tribunal in Tidewater v Venezuela considered in detail the circumstances in which a taking would be unlawful. It pointed out that declaring expropriations to be unlawful merely because they were not accompanied by the prompt payment of compensation, or even the prompt making of a good faith settlement offer, would make almost all expropriations unlawful. This was unlikely to be the intention of treaty drafting parties. A further difficulty would be caused by the fact that the question as to whether there had indeed been an expropriation was often an issue in an arbitration and while the matter was being adjudicated it would be unreasonable to expect the state to offer compensation. Further, since the task of determining compensation was granted by States to (p. 434) arbitration tribunals, it would be unreasonable to determine expropriations to be unlawful for want of compensation before tribunals had completed that task. The conclusion was:

An expropriation only wanting fair compensation has to be considered as a provisionally lawful expropriation, precisely because the tribunal dealing with the case will determine and award such compensation.128

Claims settlement agreements

9.86  Another argument that would reduce compensation payable but is not featuring in BIT cases would be one based on claims settlement agreements. According to Bowett: ‘the many post-1945 agreements disclose settlement figures ranging between ten and ninety per cent of asset value, and they disclose no support for anything like going-concern value’.129

9.87  The reasons why such agreements should not be binding on tribunals are obvious—they are negotiated settlements and take account of many non-judicial considerations which vary widely from case to case. It is thus hard to assert that agreements based on non-legal principles should be considered by tribunals applying strict standards of law.130

9.88  Nonetheless, there is nothing wrong with discerning international law from State practice, which includes such agreements. Crawford refers to certain ‘without prejudice’ claim settlements as manifestations of the general international law principle of reparation in his Commentary on the ILC’s Draft Articles on State Responsibility.131 The absence in compensation awards of tribunals even referencing a respondent’s attempt to rely upon such agreements shows how tribunals are more influenced by awards of other tribunals than by any other source of international law.

Conclusion on the practical advantages of the DCF method

9.89  The DCF approach has become widely accepted because it is the best method for valuing lost profits. Critics who complain that it produces excessive levels of compensation are perhaps not paying sufficiently close attention to how the method is operated in practice. The DCF method, properly employed, reduces speculation because it forces claimants to explain and quantify each individual area of their claim. Differences between the parties can thus be highlighted and speculative valuations can be dismissed.

9.90  It would be wrong to dismiss the method because, wrongly applied previously ‘it has been used to justify valuations which reach beyond the “fanciful” to “wonderland proportions” ’.132 For example, an economist has critically analysed the US$2.5 billion AMINOIL133 claim to demonstrate, from material in the public record, how the maximum value should have been US$267million. With the appropriate use of experts, tribunals can be put in a position to critically examine DCF claims.

9.91  Indeed, tribunals are becoming much better at these analyses. A 2001 analysis of a number of tribunal findings arising out of DCF claims concluded that while:(p. 435)

‘Baby-splitting’ may be too harsh a metaphor for the valuation methods in these cases … one must wonder how much of the tribunals’ calculations in these cases was based on an analytical application of DCF or any other method of valuation. In some cases, the evidence may have been insufficient to permit anything but rough approximations. In others, the result may reflect compromises between differing views of the arbitrators, views that perhaps were based on different and irreconcilable analyses.134

9.92  In contrast, the tribunals’ damages awards in CME v Czech Republic and CMS v Argentina show the tribunals considering the claimants’ DCF valuations in a painstaking manner, and fully reasoning their decisions to adopt, or reject, various expert reports. Tribunals are assisted in this process by the adversarial nature of arbitration proceedings, which allows them to compare parties’ opposing cases and to ask questions of the experts in order to discern the reasons for the points of difference. As the CMS Tribunal states: ‘estimates need not be arbitrary or analogous to a shot in the dark; with the appropriate methodology and the use of reasonable alternative sets of hypotheses, it is possible to arrive at figures which represent a range of values which can be rationally justified’.135

9.93  It is not always the case that respondents offer their own competing valuation evidence. For example, in National Grid v Argentina,136 Argentina chose not to present its own model or to attempt to value the financial impact of the measures it had taken. Its expert instead focused only on what it characterised as errors in the claimant’s expert report. In the absence of assistance from an expert retained by the respondent, the Tribunal decided to appoint its own expert whose draft report was submitted to the parties for comment.

9.94  Tribunals have also appointed their own experts even though both parties have offered valuation evidence.137 In Siemens v Argentina, one arbitrator provided a Separate Opinion to express his dissatisfaction with the decision of the majority not to accede to Argentina’s request to appoint an independent expert:

… I find reasonable the request of its appointment and unjustified its refusal, as such a request never seemed impertinent or untimely to me, but rather reasonable, which acceptance would not have implied any inconveniences.138

D.  Compensation for Non-Expropriatory Breaches of International Law

9.95  Investment treaties do not generally address issues of the compensation due for breaches of their provisions, save for that which is due following a finding of expropriation. However, (p. 436) tribunals have not held that as a result no compensation should be payable for any breach other than an expropriation. Tribunals that have considered this issue to date have all proceeded on the basis that, as a general principle, ‘a wrong committed by one state against another gives rise to a right to compensation for the economic harm sustained’.139 Further, ‘By not identifying any particular methodology for the assessment of compensation in cases not involving expropriation, the Tribunal considers that the drafters of the NAFTA intended to leave it open to tribunals to determine a measure of compensation appropriate to the specific circumstances of the case, taking into account the principles of both international law and the provisions of the NAFTA … The Tribunal already has suggested that whatever precise approach is taken, it should reflect the general principle of international law that compensation should undo the material harm inflicted by a breach of an international obligation.’140

9.96  Tribunals have taken the view that the failure of investment treaties to specify compensation for breaches other than expropriation gives them considerable discretion in fashioning remedies for non-expropriatory breaches.141 This discretion extends to compensating for consequential losses.142

9.97  Tribunals have attempted to construct an approach to compensation for non-expropriatory breaches on an incremental basis, building on existing principles of public international law. Yet the task is more difficult than that of assessing compensation for expropriation. When considering an expropriation, investment treaty tribunals have been able to build upon the experience of other international law tribunals over the past hundred years, such as the PCIJ, tribunals considering the energy sector nationalizations in the 1970s, and the Iran–US Claims Tribunal. There is less precedent for compensation for breaches of other treaty standards.143

9.98  Tribunals to date have approached the question of compensation for breaches of these provisions on two bases:

  1. (1)  breaches which do not involve the total loss or deprivation of an asset; and

  2. (2)  breaches that amount to the total loss or deprivation of an asset.

Breaches not amounting to a total loss or deprivation of an asset

9.99  The clearest position of the applicable principles is contained in the second partial award rendered by the NAFTA tribunal in SD Myers Inc v Canada.144 This case concerned breaches of (p. 437) the NAFTA principles of fair and equitable treatment, non-discrimination, full protection and security, and the international minimum standard of treatment (arts 1102 and 1105 of NAFTA). The claimant was a US company whose business consisted of the transport of hazardous waste materials for remediation. It developed a business line consisting of transporting waste from Canada to be processed in the United States. This business was curtailed by the conduct of Canadian public authorities in issuing a measure which closed the US/Canada border to the export of the waste in question. The Tribunal determined that this measure was passed in breach of Canada’s obligations under NAFTA. The claimant sought compensation for the damage caused to its business during the period in which the border was closed.

9.100  The Tribunal rejected Canada’s contention that the claimant should merely be compensated by being reimbursed for its initial expenditures: ‘The Tribunal concludes that compensation should be awarded for the overall economic losses sustained by SDMI that are a proximate result of Canada’s measure, not only those that appear on the balance sheet of its investment.’145

9.101  In order to compensate the investor fully, the Tribunal wanted to award it damages for the opportunity that had been lost. It approached the matter by applying a common law tort standard: ‘The inquiry in this case is more akin to ascertaining damages for a tort or delict. The damages recoverable are those that will put the innocent party into the position it would have been in had the interim measure not been passed. The focus is on causation, not foreseeability in the sense used in the law of contract. In contract law, foreseeability may limit the range of recoverability. That is not the case in the law of tort or delict. Remoteness is the key.’146

9.102  By referring to the tort standard of recovery, the Tribunal pushed questions of causation to the forefront (‘To be recoverable, a loss must be linked causally to interference with an investment located in a host State’).147 Yet future tribunals may find it difficult to build upon this analysis as the doctrine of causation has yet to be fully developed in public international law.

Causation in international law

9.103  The International Law Commission Draft Articles on State Responsibility only address the principle of causation in general terms. In describing an ‘injury’ for which a State is obliged to make reparation, art 31(2) provides that injury ‘includes any damage, whether material or moral, caused by the internationally wrongful act of a State’.

9.104  Deciding when an action is the cause of a loss is a difficult question, whatever legal test is applied. The US–Yugoslavia International Claims Commission set out the generally accepted principle in the following terms in the Dorner Claim: ‘Generally, international and domestic arbitral tribunals in the determination of international claims allow compensation for indirect damages such as loss of use of property, loss of profits and the like, if such losses are reasonably certain and are ascertainable with a fair degree of accuracy. They do not allow compensation for indirect damages if they are conjectural or speculative or not reasonably certain or susceptible of accurate determination.’148

(p. 438) 9.105  Graefrath in his Hague lectures listed a number of synonyms that cover the same basic principle. He speaks of ‘proximate cause’, ‘adequate causality’, ‘ordinary cause of events’, ‘the cause must not be too remote or speculative’, there must ‘be a sufficiently direct causal relationship’ and also the term ‘foreseeability’.149

9.106  It can thus be seen that a test has been defined by international scholars and tribunals. The real difficulty will lie in applying whatever test is taken.

9.107  However the test is defined, it will always be difficult to apply in borderline cases. Graefrath correctly alludes to the difficulty of the task in quoting with approval an observation: ‘that indeed the arbitration courts declared every damage they regarded as not to be justified an indirect damage. In this way, as a result, it was ensured that direct damage only had to be compensated’.150

9.108  If further clarification of the applicable test is required, tribunals will have to look to definitions contained in private law. This approach would be consistent with the classic statement of the sources of public international law set out in art 38 of the Statue of the International Court of Justice. This directs the ICJ to have regard to ‘the general principles of law recognized by civilized nations’.151

9.109  Crawford refers to a comparative approach in his Commentary on the ILC’s Draft Articles on State Responsibility.152 Similarly, Brownlie recognises the practical need to have regard to private law solutions in stating that: ‘International tribunals face the same problems as other tribunals in dealing with [albeit on a contract not a tort basis] indirect damages and deal with the issues in much the same way.’153

9.110  An example of an arbitral tribunal following an approach to causation derived from private law is Sapphire International Petroleum Ltd v National Iranian Oil Co (an award based upon breach of a concession contract).154 Here the sole arbitrator undertook a comparative law analysis before concluding that:

… the object of damages is to place the party to whom they are awarded in the same pecuniary position that they would have been in if the contract had been performed in the manner provided for by the parties at the time of its conclusion. They should be the natural consequence of the breach … This rule is simply a direct deduction from the principle pacta sunt servanda, since its only effect is to substitute a pecuniary obligation for the obligation which was promised but not performed.155

9.111  In setting out this approach, the sole arbitrator was applying ‘general principles of law based upon the practice common to civilized countries’156 rather than public international law. However, these rules are part of public international law as the sole arbitrator recognised in stating that: ‘These rules are enshrined in Article 38 of the Statute of the International Court (p. 439) of Justice as a source of law, and numerous decisions of international tribunals have made use of them and clarified them.’157

9.112  The corresponding part of this analysis as set out in the first edition of this work was cited with approval by the Tribunal in Biwater v Tanzania.158 The same approach was taken in Kardassopoulos v Georgia.159 In El Paso v Argentina, the Tribunal stated that it ‘shares the view expressed by other investment treaty tribunals that the test of causation is whether there is a sufficient link between the damage and the treaty violation’.160

Practical examples of tribunals awarding damages

9.113  In applying its causation test, the Tribunal in SD Myers, Inc v Canada was prepared to award compensation for the profits lost during the time the Canadian border was closed. However, it refused to award the claimant a further sum based on the profits it would have made with the money to be earned from the business it could not carry out. Such a claim was described as ‘speculative and too remote’.161

9.114  A similar approach was adopted by the tribunal in Pope & Talbot, Inc v Canada.162 Here the Tribunal held that under art 1116 of NAFTA an investor owning a corporation, which was the applicable investment, could bring a claim for loss or damage to its interest in that investment. The investor would have to show ‘that loss or damage was caused to its interest, and that it was causally connected to the breach complained of’.163

9.115  On this basis, the Tribunal awarded damages for certain out-of-pocket expenses but denied a claim for the value of management time and lost revenues from a plant shutdown, where the evidence showed that the shutdown did not cause any loss.

9.116  In Maffezini v Spain,164 an ICSID arbitration arising out of a BIT, the Tribunal rejected a claim for damage caused by the allegedly bad information contained in a feasibility study carried out by a Spanish public body. This body had carried out an economic evaluation prior to the investor making his investment. The Tribunal held that nevertheless the investor was responsible for his losses on the basis that ‘… Bilateral Investment Treaties are not insurance policies against bad business judgments’.165

9.117  In the Tribunal’s view, the shortcomings that existed in the evaluation did not relieve the investor of the business risks inherent in any investment. However, in Maffezini, the State was held not to have breached any obligations to the investor as the evaluation was not intended for third parties to rely upon. Accordingly, in this instance the quoted dicta cannot (p. 440) be taken as the Tribunal using the phrase to describe a break in the chain of causation, as liability was not established in any event.

9.118  Biwater v Tanzania is a stark illustration of a causation-based analysis being used to reduce a damages award, in this case to zero. Tanzania was adjudged to have expropriated the claimant’s water concession but the Tribunal also considered the fact of the claimant’s poor tender preparations and numerous management and implementation difficulties. These caused the concession to have a fair market value of nil as at the date of the expropriation. Accordingly, no compensation was due for the expropriation:

Causation: Applying the principles elaborated earlier, the Arbitral Tribunal concludes in all the circumstances that the actual, proximate or direct causes of the loss and damage for which [the Claimant] now seeks compensation were acts and omissions that had already occurred by [the date of expropriation]. In other words, none of the Republic’s violations of the BIT between 13 May 2005 and 1 June 2005 in fact caused the loss and damage in question, or broke the chain of causation that was already in place.166

9.119  Interestingly, the concurring and dissenting opinion did not rule that the compensatory part of the claim failed on grounds of causation. Rather, it concluded that the compensation was zero because the injury suffered by the claimant had no monetary value associated with it. This is perhaps more of a classical public international law analysis—the wrong had caused damage but the damage had no quantifiable value. Nonetheless, this type of loss could be compensated in appropriate cases by an award of moral damages167 or costs or in some other appropriate manner.168

9.120  The question of the relationship between liability and damage was also addressed in Merrill & Ring v Canada. Here, the claimant alleged that measures relating to the regulatory framework of log exports breached various NAFTA provisions. Yet the Tribunal struggled to find any specific interest of the claimant that fell to be protected under NAFTA. This brought it to consider whether a wrongful act could be established in the absence of an injury:

In the commentaries to the International Law Commission Articles on State Responsibility, the issue of whether a wrongful act could exist in the absence of damage being caused was considered. The commentaries state that ‘whether such elements [damages] are required depends on the content of the primary obligation, and there is no general rule in this respect’. Valid as that conclusion may be as far as state responsibility is concerned, in the case of conduct that is said to constitute a breach of the standards applicable to investment protection, the primary obligation is quite clearly inseparable from the existence of damage. Indeed, a finding of liability without a finding of damage would be difficult to explain in the context of investment law arbitration and would indeed be contrary to some of its fundamental tenets.169

Modelling the ‘but-for’ scenario

9.121  In Suez v Argentina the Tribunal entered into a sophisticated analysis to assess the damages arising following a finding of a breach of a fair and equitable treatment provision. The facts giving rise to the breach were that the Argentinian government passed an emergency (p. 441) law ending the fixed exchange rate of 1 US dollar to 1 Argentine peso. This devalued the Argentine peso to one-third of its previous value. When the claimant’s subsidiary holding a water concession in Argentina tried to renegotiate its tariff to take account of this reduction in income, the Argentinian government refused. This deprived the concession of the revenue required to meet financial obligations to lenders, to make investments in the water systems for which it was responsible and to earn a reasonable return. The concession company defaulted on its debts and the claimants became liable as guarantors.170

9.122  The main actions giving rise to Argentina’s breaches took place in 2001–2002 as a result of its failure to agree tariff revisions. However, the life of the concession stretched for many years beyond these failed negotiations. The Tribunal wanted to assess damages by comparing what would have happened over the life of the concession had the parties successfully negotiated in 2001–2002 against what happened in practice. The Tribunal appointed its own expert to assist in constructing:

… a hypothetical, counterfactual scenario as to the situation of the Claimants’ investments had Argentina not treated those investments unfairly and inequitably. At its heart, that scenario must answer the following question: what specific actions should the Argentine government and its regulators have taken in respect of tariff revisions in 2001–2002 for its conduct to have resulted in fair and equitable treatment of … the Claimants’ investments?171

9.123  The Tribunal accepted that its exercise relied upon an examination of hypothetical events. However, the hypothetical nature of the exercise did not deter the tribunal from undertaking it:

Based on his expertise, and in light of the legal framework, [the expert] has developed a scenario hypothesizing what a fair and equitable regulator would do in the situation that Argentina faced at the time of the crisis and afterwards. While the parties have objected to various aspects of [the expert’s] scenario, neither of the parties has provided this Tribunal with another equally detailed and convincing scenario of how a reasonable regulator intent on according fair and equitable treatment to [the claimant] would have behaved … Although both parties have objected to certain elements of the [expert’s] scenario in their pleadings in these cases, the parties’ litigation positions are not reflective of what a reasonable regulator and a reasonable operator would have done in the circumstances. In the view of the Tribunal, a confrontational scenario in all probability would not have existed if Argentina had fulfilled its obligation of fair and equitable treatment. On the other hand, the fulfillment by Argentina of its obligation of fair and equitable treatment would in all probability have fostered cooperation between the parties to achieve a solution of the kind suggested by [the expert’s] Model.172

Contributory Negligence

9.124  Causation has also been used in certain cases where an investor’s damages have been reduced due to its own acts. The earliest such case was MTD Equity Sdn Bhd v Chile. In this case, a Malaysian investor provided capital to its Chilean subsidiary to develop a real estate project in Chile. While the investment was approved by Chile, it subsequently rejected the project on the grounds that it was against Chile’s urban policy. This action was held to be a breach of the obligation to afford investors fair and equitable treatment, as laid down by the Chile–Malaysia BIT. In examining the compensation payable, the Tribunal took care to distinguish (p. 442) between those losses caused by Chile’s breach, and those which were caused to the claimant by its own actions:

The BITs are not an insurance against business risk and the Tribunal considers that the Claimants should bear the consequences of their own actions as experienced businessmen. Their choice of partner, the acceptance of a land valuation based on future assumptions without protecting themselves contractually in case the assumptions would not materialize, including the issuance of the required development permits, are risks that the Claimants took irrespective of Chile’s actions.173

9.125  As a result of the claimants needing to bear part of the damages suffered, the Tribunal reduced the compensation payable by 50 per cent.174 In Bogdanov v Moldova the Tribunal also reduced the damages payable for a breach of the fair and equitable treatment obligation as a result of the claimant’s own shortcomings.175

9.126  The award in Occidental v Ecuador shows the extent of the discretion enjoyed by tribunals in reducing damages awards on the basis of a claimant’s own wrongs. This makes it impossible to predict in advance the extent of any reduction in a specific case. The claimant in Occidental had transferred its interest in an Ecuadorian oilfield without obtaining the relevant ministry’s prior consent. This was found to be negligent and unlawful conduct. As a result, their damages were reduced by 25 per cent:

Having considered and weighed all the arguments which the parties have presented to the Tribunal in respect of this issue, in particular the evidence and the authorities traversed in the present chapter, the Tribunal, in the exercise of its wide discretion, finds that, as a result of their material and significant wrongful act, the Claimants have contributed to the extent of 25% to the prejudice which they suffered when the Respondent issued the Caducidad Decree. The resulting apportionment of responsibility as between the Claimants and the Respondent, to wit 25% and 75%, is fair and reasonable in the circumstances of the present case.176

9.127  The dissenting opinion provided more reasoning in support of a view that the reduction should have been 50 per cent. While recognising that the matter was within the Tribunal’s discretion, the dissenting arbitrator pointed out that both the original MTD Tribunal and the Annulment Committee imposed a 50 per cent reduction in respect of behaviour that was merely imprudent not illegal. It would thus be consistent for behaviour that went beyond mere imprudence to attract at least the same financial consequence.177

9.128  In Yukos v Russia the Tribunal accepted that the claimant bore some responsibility for its losses. It reduced the compensation by 25 per cent ‘in the exercise of its wide discretion.’178

Concurrent causes and breaches in the chain of causation

9.129  Cases such as MTD and Occidental can also be categorised as concurrent cause cases, with the concurrent cause being the defaults of the investor itself, rather than those of a third (p. 443) party. Such cases commonly give rise to difficult questions in domestic tort systems in circumstances when responsibility for damage can be ascribed to more than one tortfeasor. Sometimes, notwithstanding the tortfeasor’s breach, he is relieved of responsibility because of subsequent events which are determined to have broken the chain of causation.

9.130  Examples of these types of cases can be found in public international law. For example, it is established in public international law that a State may be fully liable to compensate for losses in a situation where it is not the sole cause of the loss.179 In the Corfu Channel case, Albania was held liable to Great Britain for all damage caused to the latter’s warships as a result of mines laid by Yugoslavia.180 The mines could not have been laid without the knowledge of the Albanian Government. Yet Yugoslavia, in not giving the warning required by international law, was also at fault.

9.131  An example of a concurrent cause case in the field of investment treaty arbitrations is AAPL v Sri Lanka.181 This was an ICSID arbitration arising out of the UK–Sri Lanka BIT. The investor’s farms were destroyed by military action but neither claimant nor respondent was able to demonstrate whether the atrocities had been carried out by terrorist insurgents or by State forces. The majority of the Tribunal held that in the absence of evidence ‘about what effectively caused the destruction of the farm premises’, Sri Lanka would be responsible for a failure to provide adequate protection that could have stopped the destruction taking place totally or partially.182 The Tribunal categorised the case as one where proof of facts was posing ‘extreme difficulty’ and was thus satisfied with ‘less conclusive proof’.183 The dissenting opinion pointed out the ‘troublesome questions of causation’ raised by this decision.184 By holding Sri Lanka responsible for failure to take precautionary measures prior to launching a counterinsurgency action, Sri Lanka was being held liable even if the damage was inflicted by third parties taking advantage of the Government’s legitimate operation to commit unlawful acts. Rather than making the investor prove that Sri Lanka had caused its loss, the Tribunal was condemning Sri Lanka for its failure to prove otherwise. Accordingly, precedent exists for tribunals ascribing all of the damage to a State in breach of an international obligation, although it is not the only party in default.

9.132  In Impregilo v Argentina the Tribunal was unable to decide whether Argentina’s violations of the fair and equitable treatment standard had actually caused loss to the claimant. This is because the concession was proving unprofitable as a result of various business results for which the claimant had assumed the risk. As a result, the Tribunal decided to award compensation based upon the claimant’s capital contribution.185 The Tribunal’s conclusion could be described as deciding that the risks for which the claimant bore responsibility were a concurrent cause in the lost profit chain of causation. Further, this concurrent cause was one which absolved Argentina of financial liability for its own contribution to those lost profits.

(p. 444) 9.133  The parties pleaded as to who bore the burden of establishing responsibility in a concurrent cause situation in Yukos v Russia. The respondent submitted that once it can be established that the claimant’s own conduct was part of the cause of its own loss, the claimant should bear the burden of establishing that the loss was caused by the wrongful conduct of the respondent, rather than by its own behaviour, or that of third parties or conduct of the respondent that was not wrongful. The Tribunal rejected this approach, deciding that a concurrent action that is not a breach does not by itself breach the chain of causation:

Rather, it falls to the Respondent to establish that a particular consequence of its actions is severable in causal terms (due to the intervening actions of claimants or a third party) or too remote to give rise to Respondent’s duty to compensate.186

9.134  Similarly, in Stati v Kazakhstan, the Tribunal concluded that the evidence that respondent’s actions caused the destruction of the claimant’s business was so strong that the burden of proof shifted to the respondent to demonstrate that in spite of its conduct the claimant contributed to its own losses in a relevant way.187

9.135  The final award in Lauder v Czech Republic188 is an example of a tribunal finding that a subsequent action breaks the chain of causation. In this ad hoc arbitration brought under the US–Czech Republic BIT, the Tribunal found that the Czech Republic had taken a discriminatory and arbitrary measure against the claimant. This consisted of having initially accepted that the claimant could invest directly in a Czech television company but subsequently requiring the investment to be channelled through a third company. Yet the Tribunal refused to award compensation for the breach because the matters characterised by the Tribunal as the ‘last, direct act, the immediate cause’ were ‘so unexpected and so substantial as to have to be held to have superseded the initial cause and therefore become the main cause of the ultimate harm’.189 Later acts, characterised as a ‘typical commercial dispute’190 for which the respondent State was not responsible were found to be ‘the real cause for the damage which apparently has been inflicted to the Claimant’.191

9.136  As noted in Chapter 8 above, a different tribunal (CME Czech Republic v Czech Republic)192 examining the same facts reached a different conclusion on the merits. This difference of opinion extended to the two tribunals’ findings on causation. The CME Tribunal rejected the argument that the facts amounting to the ‘typical commercial dispute’ absolved the Czech Republic of responsibility for the fate of the investment. The Tribunal stated that such an argument would fail ‘under the accepted standards of international law’.193 The Tribunal cited the following rule from a comparative law treatise: ‘It is the very general rule that if a tortfeasor’s behaviour is held to be a cause of the victim’s harm, the tortfeasor is liable to pay for all of the harm so caused, notwithstanding that there was a concurrent cause of that harm (p. 445) and that another is responsible for that cause … In other words, the liability of a tortfeasor is not affected vis-à-vis the victim by the consideration that another is concurrently liable.’194

Valuing an arbitration claim or court proceedings

9.137  With the rise in claims arising out of underlying court or arbitration proceedings, tribunals have had to address the question of how to value those underlying proceedings. Saipem v Bangladesh was a case in which the Tribunal found that an ICC arbitration award had been expropriated by the actions of the Bangladeshi courts. The Tribunal applied the Chorzów Factory principles of payment of a sum corresponding to the value represented by restitution in kind. Accordingly, it ruled that:

[T]he expropriation of the right to arbitrate the dispute in Bangladesh under the ICC Arbitration Rules corresponds to the value of the award rendered without the undue intervention of the court of Bangladesh.195

9.138  In White Industries v India, the Tribunal took a similar approach. Its finding on the merits was that the Indian judicial system’s extensive delays in dealing with enforcement of an underlying arbitration award amounted to a breach of the India–Australia BIT.196 Having satisfied itself that the award was fully enforceable in India, the Tribunal awarded the claimant the amounts originally payable under the award.197

9.139  The analysis in Chevron v Ecuador was complicated by the fact that the underlying breach was that the Ecuadorian courts had failed to progress the claimant’s claims. Consequently, there was no underlying judgment or award. However, the parties agreed that the Tribunal should apply the Chorzów Factory analysis and award compensation if the claimants could prove that they would more likely than not have prevailed on the merits before the Ecuadorian courts:

In essence, the Claimants must prove the element of causation—i.e., that they would have received judgments in their favour as they allege ‘but for’ the breach by the Respondent.198

9.140  As a result, the Tribunal had to conduct a trial within a trial and ask itself how a ‘competent, fair, and impartial’ Ecuadorian court would have decided the underlying claims. While the Tribunal was prepared to take into account any judgment actually rendered, they owed that judgment no deference. Further, the Tribunal refused to assess damages on a ‘loss of a chance’ basis. Had the underlying claims proceeded to judgment in Ecuador, the claimants would have made a full recovery without any discount factor based on a loss of a chance. The Tribunal also noted that the loss of a chance principle did not have wide acceptance across legal systems which would allow it to be considered a general principle of law recognised by civilized nations pursuant to art 38 of the Statute of the International Court of Justice.199

(p. 446) Loss of a tender opportunity

9.141  In Lemire v Ukraine the Tribunal found that Ukraine’s breach of the USA–Ukraine BIT led to the claimant wrongfully failing in tenders to be awarded Ukrainian radio licences. The Tribunal considered whether the claimant would have won the tender competitions had they been held on a fair basis:

Given the characteristics of the Ukrainian process for the awarding of licences, it is impossible to establish, with total certainty, how specific tenders would have been awarded if the National Council had not violated the FET standards. The best that the Tribunal can expect Claimant to prove is that through a line of natural sequences it is probable—and not simply possible—that Gala would have been awarded the frequencies under tender. If it can be proven that in the normal cause of events a certain cause will produce a certain effect, it can be safely assumed that a (rebuttable) presumption of causality between both events exists, and that the first is the proximate cause of the other.200

9.142  In concluding that the claimant would have succeeded in the underlying tenders, the Tribunal considered the facts such as the claimants’ existing radio business successes, the quality of its broadcast programmes and its proven and successful track record as a music transmitter and news provider.201

Breaches amounting to a total loss or deprivation of an asset

9.143  This matter was addressed by the Tribunal in CMS v Argentina. The claim arose out of Argentina’s suspension of a tariff adjustment formula for gas transportation applicable to an enterprise in which the claimant was a minority shareholder. The Tribunal found that Argentina’s measures breached the fair and equitable treatment standard contained in the US–Argentina BIT.202 In considering the compensation to be paid by Argentina, the Tribunal, having explicitly dismissed an expropriation claim, nonetheless decided to use the expropriation compensation standard: ‘the Tribunal is persuaded that the cumulative nature of the breaches discussed here is best dealt with by resorting to the standard of fair market value. While this standard figures prominently in respect of expropriation, it is not excluded that it might also be appropriate for breaches different from expropriation if their effect results in important long-term losses’.203

9.144  The Tribunal in BG v Argentina adopted a different approach. It referred to the CMS v Argentina award, but described its reasoning as ‘scant’. It stated that it would be wrong to read a compensation standard into the fair and equitable treatment provision that the parties had deliberately confined to the expropriation standard. Nevertheless, as a matter of customary international law, it was prepared to use a ‘fair market value’ standard.204 The Tribunal’s different approach did not therefore yield a different result.

9.145  In Walter Bau v Thailand the Tribunal did not feel any need to justify the fact that it was awarding damages by reference to any legal standard. Having found that the respondent (p. 447) had committed acts which constituted breaches of the fair and equitable treatment requirement, it continued to say that ‘[i]t remains for the Tribunal to evaluate the damages owing to Claimant for such breaches.’205

9.146  Most tribunals are content to adopt the CMS reasoning. For example, the Tribunal in Vivendi II stated:

[the fair market value] standard has also generally been accepted as appropriate compensation for expropriation. However, as pointed out by the tribunal in CMS Gas Transmission Co v Argentine Republic, a ‘fair market value’ standard might also be appropriate for other breaches which result in long-term losses.206

E.  Moral Damages

9.147  In Desert Line v Yemen,207 the Tribunal made what is generally believed to be the first award of moral damages in an investment arbitration. The investor had undertaken a project to build roads in Yemen, which ended in a construction dispute over payment. The dispute resulted in an arbitration pursuant to which the investor was awarded damages. The investor sought to enforce the award through the Yemeni courts, and via settlement discussions. This process culminated in a settlement agreement which the investor claimed it was forced to sign in order to avoid physical and economic harm at the hands of the State. The Tribunal found that the settlement agreement had been imposed on the claimant under physical and financial duress and was not the result of a fair and sincere negotiation.208 The Tribunal ordered Yemen to pay compensation as set out in the original arbitration award. The claimant also sought moral damages including damages for loss of reputation.

9.148  There is no controversy as to whether moral damages can be obtained under classical principles of public international law. The best known examples are the Lusitania cases decided by the United States–Germany Mixed Claims Commission after the First World War. The claims arose out of the sinking of a British passenger vessel by a German submarine. Crawford’s Commentary on the ILC’s Draft Articles on State Responsibility describes the case as follows:

… non-material damage is financially assessable and may be the subject of a claim of compensation, as stressed in the ‘Lusitania’ case. The umpire considered that international law provides compensation for mental suffering, injury to feelings, humiliation, shame, degradation, loss of social position or injury to credit and reputation, such injuries being ‘very real, and the mere fact that they are difficult to measure or estimate by money standards makes them none the less real and affords no reason why the injured person should not be compensated …’209

(p. 448) 9.149  Even though moral damages can be granted under principles of international law, it does not automatically follow that they should be available in investment arbitrations. It is generally accepted that damages should be compensatory and not punitive, and a distinction could be drawn between moral damages awarded to natural persons and those awarded to corporate individuals. Nonetheless, in Desert Line, Yemen did not seek to contend that investment arbitrators could not award moral damages. The Tribunal awarded US$1 million to Desert Line on the basis that this sum was ‘more than symbolic yet modest in proportion’. It reasoned that:

It is generally accepted in most legal systems that moral damages may also be recovered besides pure economic damages. There are indeed no reasons to exclude them … It is also generally recognised that a legal person (as opposed to a natural one) may be awarded moral damages, including loss of reputation, in specific circumstances only … The Arbitral Tribunal finds that the violation of the BIT by the Respondent, in particular the physical duress exerted on the executives of the Claimant, was malicious and is therefore constitutive of a fault-based liability. Therefore, the Respondent shall be liable to reparation for the injury suffered by the Claimant, whether it be bodily, moral or material in nature. The Arbitral Tribunal agrees with the Claimant that its prejudice was substantial since it affected the physical health of the Claimant’s executives and the Claimant’s credit and reputation.210

9.150  Moral damages were also awarded in von Pezold v Zimbabwe.211 A number of the claimants recounted details of the mistreatment they had received from ‘war veterans’ in being dispossessed of their Zimbabwean farms. One difficulty faced by the claim for moral damages was that the owners of certain claimant entities were not present in Zimbabwe at the time of the acts about which complaint was being made. Nonetheless, there was no doubt that their staff members had suffered. The Tribunal was prepared to award moral damages to the claimants on the basis of poor treatment received by staff members. The Tribunal was motivated to reach this conclusion because of the unlikelihood of the staff members obtaining relief through domestic courts. An award of moral damages ‘… serves not only the function of repairing intangible harm, but also of condemning the actions of the offending State.’212

9.151  Moral damages were also considered in detail by the Tribunal in Lemire v Ukraine. The claimant alleged that Ukraine’s actions in rejecting all of its applications for new radio frequencies, together with oppressive business inspections, gave rise to a claim for moral damages in addition to a more conventional compensatory claim. The Tribunal denied the moral damages claim because the circumstances were not exceptional. It reviewed the relevant international law and investment arbitration authorities before concluding that:

… as a general rule, moral damages are not available to a party injured by the wrongful acts of a State, but … moral damages can be awarded in exceptional cases, provided that

  • - the State’s actions imply physical threat, illegal detention or other analogous situations in which the ill-treatment contravenes the norms according to which civilised nations are expected to act;

  • (p. 449) - the State’s action cause a deterioration of health, stress, anxiety, other mental suffering such as humiliation, shame and degradation, or loss of reputation, credit and social position; and

  • - both cause and effect are grave or substantial.213

9.152  Moral damages were also rejected in Funnekotter v Zimbabwe where the dispossessed Dutch farmers sought to bring a moral damages claim on top of their compensation claim. The Tribunal had ordered payment of €20,000 as a reparation for the disturbances involved in having to leave a home in Zimbabwe and start life again in another country. Moral damages were seen to overlap with this disturbance payment.214 They were also rejected in Siag v Egypt where the Tribunal noted that the BIT did not allow punitive damages. It referred to Desert Line, but stated that the jurisprudence of the Iran–US Claims Tribunal only allowed punitive or moral damages for extreme cases of egregious behaviour.215 In Rompetrol v Romania moral damages were rejected on the basis that they did not serve as a proxy claim in circumstances where claimant was unable to prove actual economic damage.216 They were rejected in Arif v Moldova because the Tribunal believed that the developing nature of Moldova’s governing institutions would raise the standard an investor would need to reach to demonstrate ‘exceptional circumstances’. The Tribunal accepted that the conduct of the Moldovan authorities provoked stress and anxiety in the claimant, but it did not reach the level of ‘gravity and intensity’ which would be required to trigger a moral damages claim.217

9.153  The dissenting arbitrator in Biwater v Tanzania was prepared to consider moral damages. This was a case where the Tribunal found that Tanzania had breached the BIT but the claimant failed to establish that it had suffered any financial loss. In the circumstances, the dissenting arbitrator took the view that:

In circumstances where a State deliberately conducts itself in a manner it knows at the time to be wrongful, disregarding the basic legal rights and protections of private parties, it is at best anomalous for a tribunal to grant no affirmative relief. It is ancient law that there is no right without a remedy (Ubi jus ibi remedium) and that adage applies here no less than elsewhere. Whether denominated as moral damages (as some tribunals have done, but which has not been specifically requested here), recognized by way of a costs award (as other tribunals have done), or otherwise, it better advances the objectives of bilateral investment treaties and the ICSID Convention to require a measure of tangible reparations for violation of internationally-protected rights.218

9.154  In the circumstances, the dissenting arbitrator would have made an award of costs in favour of the investor.219

(p. 450) 9.155  In two cases the respondent State sought to turn the tables on the investor and assert a claim for moral damages based on the number and allegedly abusive nature of legal and media claims brought by the investor in support of its dispute. These were both rejected; the Tribunal in ST-AD v Bulgaria pointed out that it is the right of any person to pursue any avenues it believes it possesses to assert its rights.220

F.  Mitigation

9.156  A respondent State will not be liable to pay damages in respect of losses which could have been mitigated by actions of the claimant. The principle of mitigation was stated by the International Court in a case concerning the Gabčíkovo-Nagymaros Project (Hungary v Slovakia)221 as being ‘a principle that an injured State which has failed to take the necessary measures to limit the damage sustained would not be entitled to claim compensation for that damage which could have been avoided’.222

9.157  In Middle East Cement v Egypt223 the Tribunal stated that a duty to mitigate loss is one of the general principles of law which are part of international law.

G.  Non-Pecuniary Remedies

9.158  To date, the remedy awarded by almost all tribunals has been the payment of monetary compensation. Yet this past practice should not obscure the fact that tribunals have the power to be much more flexible in their choice of remedy.224

9.159  An order to a State to carry out a particular act would be seen as a far greater infringement of State sovereignty than an award of compensation. However, as investment treaty arbitrations have sometimes been criticised for their potential to result in large monetary awards against relatively poor nations, it may be time to rethink this position. Tribunals should be willing to consider preliminary orders to lift discriminatory treatments or to seek other administrative remedies that can provide full satisfaction to investors before moving to award compensation. Such an approach has been suggested by Wälde and Sabahi.225 The difficulty faced by tribunals wishing to adopt such an approach is the slow pace at which investment arbitrations move. By the time the tribunal addresses questions of compensation, any possibility of ordering a remedy to restore the situation as it existed prior to the infringement has usually been irrevocably lost.

9.160  There is little tribunal practice to provide guidance in this area. In one BIT arbitration where the Tribunal considered the issue of injunctive relief in depth, the respondent State took the (p. 451) traditional view of State sovereignty. In Enron v Argentina226 Enron’s Argentinian subsidiary had been the subject of various tax assessments. Enron sought relief pursuant to the US–Argentina BIT before the taxes had to be paid. Argentina challenged the Tribunal’s jurisdiction to grant an injunction which would prevent it from collecting the taxes. It contended that, even if the tax assessments would constitute an expropriatory act, ‘an ICSID tribunal cannot impede an expropriation that falls exclusively within the ambit of State sovereignty; that tribunal could only establish whether there has been an expropriation, its legality or illegality and the corresponding compensation’.227

9.161  Thus, Argentina would have preferred to be left to proceed with the expropriation, and assume the risk of monetary compensation, than be told not to enforce the tax assessment.

9.162  The Tribunal rejected Argentina’s jurisdictional objection. It noted that the parties had both agreed that it had the power to issue declaratory relief but that in addition it had ‘the power to order measures involving performance or injunction of certain acts’.228

9.163  In arriving at this conclusion, the Tribunal considered the jurisprudence of other international courts and tribunals, including the International Court of Justice. Most notably it cited the award in the Rainbow Warrior Affair. Two French agents had arranged for an explosion on a vessel in a New Zealand harbour. In order to settle the dispute arising out of this act, France and New Zealand had agreed that the agents should be kept at a remote island. The arbitration arose out of New Zealand’s attempt to seek an order for the return of the two agents whom they contended France had repatriated prematurely. The Tribunal not only confirmed its power to grant injunctive relief but also found that such powers arose inherently from its jurisdiction:

The authority to issue an order for the cessation or discontinuance of a wrongful act or omission results from the inherent powers of a competent tribunal which is confronted with the continuous breach of an international obligation which is in force and continues to be in force. The delivery of such an order requires, therefore, two essential conditions intimately linked, namely that the wrongful act has a continuing character and that the violated rule is still in force at the time in which the order is issued.229

9.164  The question was also addressed in Micula v Romania. The claimant asked, as part of the relief sought, for the interim relief previously awarded by the Tribunal to be continued as permanent injunctive relief. The Tribunal concluded that it would have power to do this as part of its mandate to resolve disputes between the claimants and the respondent arising out of the claimants’ investments. Neither the ICSID Convention nor the BIT placed any limitations on the remedies that could be awarded by the tribunal. Nonetheless, the claimants’ request was rejected as there was no continuing right to the maintenance of any status quo following resolution of the claimants’ claims on the merits. It also would not be practicable for a tribunal to grant measures that may need to be reconsidered after the tribunal had become functus officio.230

(p. 452) 9.165  Another case where restitution was ordered is von Pezold v Zimbabwe.231 This was a claim brought by a number of farmers seeking restitution and monetary compensation for farms taken by groups that had been supported by the Zimbabwe government. The Tribunal recognised that restitution was not often awarded but noted that nothing in the BIT would prevent its award. It applied the ILC Draft Articles on State Responsibility and in particular considered whether any of the principal ‘defences’ to restitution appearing in art 35 applied. These are material impossibility and disproportionality. The Tribunal concluded that it was possible for Zimbabwe to make a change in the Land Registry and that such relief would be limited to the claimants and therefore not disproportionate. On that basis, it was prepared to order restitution. It also awarded monetary compensation in addition for losses caused to the claimants as a result of land damage and losses to productivity.

9.166  A difficulty with the enforcement of non-pecuniary remedies is that they fall outside the specific ICSID enforcement framework. Article 54(1) of the ICSID Convention provides that: ‘Each Contracting State shall recognize an award rendered pursuant to this Convention as binding and enforce the pecuniary obligations imposed by that award within its territories as if it were a final judgment of a court in that State.’

9.167  It is also notable that the US model BIT limits tribunals to the award of monetary damages and restitution of property.232

9.168  Such potential obstacles can be overcome by a tribunal ordering a two-stage remedy. It can order specific performance in the first place within a limited period and then provide for the payment of monetary compensation in the event that a State would not comply with the specific performance order. Such an approach, making compliance with specific performance ‘voluntary’, would also help to satisfy those who support maintaining the strict view of State sovereignty.233

9.169  A two-stage approach was adopted by the tribunal in Goetz v Burundi.234 The Belgian claimant owned a company incorporated in Burundi which enjoyed the benefit of a free zone certificate granting tax and customs exemptions. Burundi withdrew the certificate and Goetz brought a claim under the Belgium–Burundi BIT. The Tribunal gave Burundi a choice of restoring the previous situation or paying monetary compensation:

… it falls to the Republic of Burundi, in order to establish the conformity with international law of the disputed decision to withdraw the certificate, to give an adequate and effective indemnity to the claimants as envisaged in Article 4 of the Belgium–Burundi investment treaty, unless it prefers to return the benefit of the free zone regime to them. The choice lies within the sovereign discretion of the Burundian government. If one of these two measures is not taken within a reasonable period, the Republic of Burundi will have committed an act contrary to international law the consequences of which it would be left to the tribunal to ascertain.235

(p. 453) 9.170  In Arif v Moldova, the Tribunal found that the State’s conduct in closing the investor’s airport duty-free shop was a breach of the France–Moldova BIT. The respondent wanted the Tribunal to investigate the possibility of restitution as an alternative to a damages award. However, the claimant insisted on reparation, contending that Moldova would be unwilling or unable to make restitution, backed by adequate guarantees. The Tribunal’s preference was for restitution viewing it as ‘more consistent with the objectives of bilateral investment treaties, as it preserves both the investment and the relationship between the investor and the Host State.’236 However, recognising the fact that it would be unable to supervise any restitutionary remedy, it ordered restitution and compensation as alternatives, with the compensation suspended for a period of ninety days.

H.  Interest

9.171  Article 38 of the ILC Draft Articles on State Responsibility provides for the payment of interest when necessary in order to ensure full reparation. It also provides that the interest rate and mode of calculation shall be set so as to achieve that result.237

Pre- and post-award interest

9.172  As long as the purpose of interest is seen as compensating the claimant for the cost of being deprived of its money, there should be no difference between pre-and post-award interest. The Tribunal in Micula v Romania stated that:

… the Tribunal does not see why the cost of the deprivation of money (which interest compensates) should be different before and after the Award …238

9.173  However, some suggest that there is a difference between interest awarded prior to an award and interest awarded following an award. Whereas pre-award interest operates solely to compensate a claimant for being kept out of its money, interest following an award should be fixed at a higher level because the respondent is in default of payment. In the words of the Kardassopoulos v Georgia Annulment Committee, such interest has ‘a bearing on the efficacy of the award’.239

9.174  This idea was also adopted in Gold Reserve v Venezuela where the Tribunal stated:

As requested by Claimant, the Tribunal may also determine a different interest rate to apply to post-Award interest than that applied to pre-Award interest. This is because the purpose of post-Award interest is arguably different—damages become due as at the date of the Award, and from this time, Respondent is essentially in default of payment. As such, the Tribunal considers that continuing to apply a risk-free interest rate would be inappropriate.240

(p. 454) 9.175  It is however not the role of an investment tribunal to police its own awards or to apply sanctions for non-compliance. Accordingly, tribunals should be reluctant to provide for a higher rate of interest solely to encourage respondent States to pay promptly.241

Compound interest

9.176  The ILC Commentary sets out the traditional public international law position, derived from State-to-State practice, that interest should be simple, not compound. It quotes the Iran–US Claims Tribunal in RJ Reynolds Tobacco Co v Iran, stating that:

As noted by one authority, ‘(t)here are few rules within the scope of the subject of damages in international law that are better settled than the one that compound interest is not allowable’ … Even though the term ‘all sums’ could be construed to include interest and thereby to allow compound interest, the Tribunal, due to the ambiguity of the language, interprets the clause in the light of the international rule just stated, and thus excludes compound interest.242

9.177  Other commentators have noted that ‘arbitral tribunals (like national courts) have been nearly unanimous: compound interest (or interest on interest) is not allowed’.243

9.178  If the rule in State-to-State disputes is against compound interest, this is perhaps one area where investment treaty arbitration is developing its own practice. While there is no unanimity, many tribunals have been prepared to award compound interest.

9.179  In Santa Elena v Costa Rica244 the Tribunal distinguished much of the State-to-State jurisprudence on interest on the grounds that it had arisen ‘principally in relation to cases of injury or simple breach of contract. The same considerations do not apply to cases relating to the valuation of property or property rights. In cases such as the present, compound interest is not excluded where it is warranted by the circumstances of the case’.245

9.180  The Tribunal viewed entitlement to compound interest as part of the claimant’s entitlement to receive the full present value of the compensation that it should have received at the time of the taking:

… where an owner of property has at some earlier time lost the value of his asset but has not received the monetary equivalent that then became due to him, the amount of compensation should reflect, at least in part, the additional sum that his money would have earned, had it, and the income generated by it, been reinvested each year at generally prevailing rates of interest. It is not the purpose of compound interest to attribute blame to, or to punish, anybody for the delay in the payment made to the expropriated owner; it is a mechanism to ensure that the compensation awarded the Claimant is appropriate in the circumstances.246

(p. 455) 9.181  It would perhaps be possible to distinguish the Santa Elena award as having arisen out of extreme circumstances. The expropriation took place in 1978, almost twenty-two years before the date of the award. Nonetheless, subsequent BIT tribunals have cited Santa Elena as authority for the principle that compound interest is payable even in cases where no such extreme circumstances exist. An example would be Middle East Cement v Egypt where the Tribunal concluded that ‘to make the compensation “adequate and effective” pursuant to Art. 4.c) of the BIT, it is appropriate that the interest pursuant to the last sentence of Art. 4.c) of the BIT be awarded as compound interest’.247

9.182  The Tribunals in CMS v Argentina248 and Azurix v Argentina249 granted compound interest to the claimants. The awards contain little analysis on this point but the Azurix Tribunal stated that ‘compound interest reflects the reality of financial transactions, and best approximates the value lost by an investor’.250

9.183  While most recent investment treaty tribunals have awarded compound interest, a number have awarded simple interest. It thus can be argued that there is ample precedent on both sides of the debate. Yet on closer inspection it is possible to say that two of the simple interest awards251 fell into the category specifically identified by the Santa Elena Tribunal as being suitable for simple interest, being claims for ‘simple breach of contract’.252

9.184  This restrictive reading of the Santa Elena Tribunal was specifically cited in Autopista v Venezuela.253 Likewise Occidental v Ecuador254 was not an expropriation claim.

9.185  While CME v Czech Republic255 was an expropriation claim where the Tribunal analysed the issue of interest in some depth, the Tribunal specifically stated that it did not need to award compound interest to fully compensate the damage sustained because of the generous interest provision it was applying by reference to a Czech statute.256 Thus, it is submitted that a better reading of the Tribunal awards on this point will lead to the conclusion that in expropriation cases compound interest is to be preferred.257 This is supported by the conclusion of a study finding ‘a growing recognition that compound interest may be necessary to fully and adequately compensate an injured investor’.258

(p. 456) 9.186  The study conducted by Uchkunova and Temnikov of 42 ICSID awards up to 2014 found that following the Santa Elena Award only 10 out of 42 awards applied simple interest.259 Most of the decisions to apply simple interest could be explained by factors peculiar to the individual cases.

9.187  The Tribunal in Gemplus v Mexico put the case for compound interest very highly in saying that:

… there is now a form of ‘jurisprudence constante’ where the presumption has shifted from the position a decade or so ago with the result it would now be more appropriate to order compound interest, unless shown to be inappropriate.260

9.188  That probably goes too far but it is however reasonable to propose that an award of compound interest should be the rule rather than the exception.

9.189  The other important questions that arise in considering questions of interest are the date from which interest is payable and the rate of interest. Both of these questions are usually highly fact specific and it is difficult to offer guidance beyond that set out at art 38(2) of the ILC Draft Articles on State Responsibility: ‘Interest runs from the date when the principal sum should have been paid until the date the obligation to pay is fulfilled.’

I.  Costs

9.190  In commercial arbitrations it is common for the tribunal to order the losing party to pay the costs incurred by the winning party. These costs include sums paid to lawyers and the winning party’s share of the arbitration expenses. By contrast, the public international law tradition is for each party to bear its own legal costs and for the tribunal costs to be shared equally.

9.191  The prevalent position in investment arbitration was originally to follow the public international law tradition and avoid the ‘loser pays’ principle. In Metalclad v Mexico261 the costs were split evenly between the parties even though the claimant was successful. Where the claimant was the losing party, tribunals also adhered to the principle of equal allocation. For example, in Tradex v Albania262 and the NAFTA case ADF v United States,263 both State parties prevailed on the merits but in each case the losing investors were not ordered to pay the costs of the winning State. Instead, each party was ordered to bear its own legal costs and to share the costs of the arbitration.

9.192  Limited reasoning for this was provided in ADF, although the Tribunal considered ‘the circumstances of [the] case, including the nature and complexity of the questions raised by the disputing parties’.264 In Tradex, the Tribunal took into account that Tradex had prevailed on (p. 457) the jurisdiction challenge and that its claim could not ‘be considered as frivolous in view of the many difficult aspects of fact and law involved and dealt with …’.265

9.193  This traditional approach is changing, although arbitration costs awards continue to demonstrate how each tribunal is sovereign in respect of its own dispute. The most significant indicator of a change in tribunal practice was provided by the costs decision in a NAFTA case, Thunderbird v Mexico.266 Here the Tribunal criticised the reasoning of previous tribunals in cases where the ‘loser pays’ approach had been rejected. The majority of the Tribunal said that it ‘[failed] to grasp the rationale’ of cases in which the losing investor is not ordered to pay the costs of the Government, ‘except in the case of an investor with limited financial resources where considerations of access to justice may play a role’.267 The Tribunal argued that the same rules on costs should apply to international investment arbitration as in international commercial arbitrations.

9.194  The Thunderbird Tribunal considered four factors that had previously been applied (in Azinian v Mexico)268 to decide that the losing investor need not pay the costs of the State party. These were (a) the novelty of NAFTA as a dispute resolution mechanism; (b) whether the claimant presented its case in an efficient and professional manner; (c) whether the respondent may be said to some extent to have invited litigation and (d) whether the persons most accountable for the claimant’s wrongful behaviour would be the least likely to be affected by an award of costs.

9.195  The Tribunal decided that factor (a) was no longer applicable because NAFTA arbitration is now well established. Factors (c) and (d) were inapplicable to the case. Although factor (b) was satisfied, the Tribunal considered it insufficient to deviate from the general principle that the losing party should bear the costs. Accordingly, the Tribunal determined that Mexico could recover an appropriate proportion of the costs of its legal representation and of the arbitration. Given that Mexico did not prevail on the issues of jurisdiction and admissibility, the majority of the Tribunal exercised its discretion and allocated the costs between Thunderbird and Mexico on a 75:25 per cent basis. In a Separate Opinion, the third arbitrator recognised that the allocation of the bulk of the costs against the claimant was a ‘significant departure from established jurisprudence’.269

9.196  Another early case adopting the ‘loser pays’ approach was Methanex Corporation v USA.270 Here the Tribunal dismissed all of Methanex’s claims on the merits. The Tribunal decided that ‘there is no compelling reason not to apply the general approach required by the first sentence of Article 40(1) of the UNCITRAL Rules’271—that the unsuccessful party should bear the costs of the arbitration. Accordingly, Methanex was ordered to pay the costs of the arbitration. With respect to the parties’ legal costs, the Tribunal said that ‘as a general (p. 458) principle the successful party should be paid its reasonable legal costs by the unsuccessful party’.272 Given that the United States had been successful in terms of both jurisdiction and merits, the Tribunal ordered Methanex to pay the United States’ reasonable legal costs of approximately US$3 million.

9.197  The principle of equal allocation of costs was also not applied in CSOB v Slovakia,273 an ICSID case in which CSOB prevailed on the merits. The Tribunal ordered Slovakia to pay its own costs, as well as a US$10 million contribution (about 60 per cent) towards CSOB’s costs. The decision is also notable for the fact that the Tribunal allowed the successful party to recover costs incurred by party employees on the grounds that those employees had prepared specific contributions that were caused by the arbitration and made several statements filed with the Tribunal.274

9.198  Unusually, in EnCana v Ecuador275 the Tribunal ordered the winning party (Ecuador) to pay the arbitration costs (but not the legal costs) of the losing party. Although the Tribunal cited the general principle under art 40(1) of the UNCITRAL Rules that the prevailing party was in principle entitled to its costs, it decided that it would be just and equitable to order Ecuador to bear the costs in view of the events giving rise to the arbitration.

9.199  A half way principle is for the losing party to have to pay the prevailing party’s costs of the arbitration with both parties being required to bear their own legal costs. This was the approach adopted by the Tribunal in Azurix v Argentina.276

9.200  A 2014 study by Hodgson analysed 221 arbitrations where data was available up to 31 December 2012. Hodgson found that 56 per cent of tribunals had required each party to bear its own costs, irrespective of the outcome on the merits. Ten per cent of tribunals assigned the whole of the costs to one party and 34 per cent made an order adjusting some of the costs. Hodgson also found that the frequency of tribunals making costs-shifting orders was increasing.277


1  Factory at Chorzów (Germany v Poland) (Merits) (1928) PCIJ Rep Series A No 17, 47; ILC ‘Draft Articles on the Responsibility of States for Internationally Wrongful Acts’ [2001] 2(2) YB ILC 26, arts 34–6. See also Crawford, State Responsibility: The General Part (2013) especially chaps 15 (Reparation) and 16 (Remedies).

2  For a book length treatment of this topic see Ripinsky and Williams, Damages in International Investment Law (BIICL, 2008).

3  O Schachter, ‘Compensation for Expropriation’ (1984) 78 AJIL 121.

4  The correspondence is reprinted ‘Official Documents: Mexico-United States’ (1938) 32 AJIL Supp 181; (1942) 3 Hackworth Digest of International Law 655.

5  C Calvo, Derecho Internacional Teorica y Practico de Europa y America (1868).

6  UNGA ‘Permanent Sovereignty over Natural Resources’, UNGA Res 1803 (XVII) (14 Dec 1962).

7  UNGA ‘Charter of Economic Rights and Duties of States: resolution’ UNGA Res 29/3281 (12 Dec 1974) UN Doc A/Res/29/3281; UNGA Res 39/163 (17 Dec 1984) UN Doc A/Res/39/163.

8  ILC ‘Draft Articles on Responsibility of States for Internationally Wrongful Acts’ (Crawford, Special Rapporteur) [2001] 2(2) YB ILC 26, art 31(1) [emphasis added].

9  ibid art 36(1).

10  World Bank, ‘Report to the Development Committee and Guidelines on the Treatment of Foreign Direct Investment’ (1992) 31 ILM 1366, see <www.italaw.com/documents/WorldBank.pdf> (Appendix 13 below) (‘World Bank Guidelines’). These Guidelines have often been cited by investment arbitration tribunals. Examples include Fedax NV v Venezuela (Decision on Jurisdiction) ICSID Case No ARB/96/3, 5 ICSID Rep 183, IIC 101 (1997, Orrego Vicuña P, Heth & Owen) para 35, CME Czech Republic BV v Czech Republic (Final Award) 9 ICSID Rep 264, IIC 62 (UNCITRAL, 2003, Kühn P, Brownlie & Schwebel) paras 161, 498 and 501 and Rumeli Telekom AS v Kazakhstan (Award) ICSID Case No ARB/05/16, IIC 344 (2008, Hanotiau P, Boyd & Lalonde) paras 730, 801–4 and 811.

11  World Bank Guidelines Guideline IV(1).

12  ibid Guideline IV(2).

13  R Dolzer, ‘New Foundations of the Law of Expropriation of Alien Property’ (1981) 75 AJIL 553; ILC ‘Responsibility of States for Internationally Wrongful Acts: Draft Articles with Commentaries’ (Crawford, Special Rapporteur) [2001] 2(2) YB ILC 30, art 36 commentary.

14  eg Treaty Concerning the Encouragement and Reciprocal Protection of Investment (US–Ecuador) (signed 27 August 1993, entered into force 11 May 1997) Senate Treaty Doc 103-15.

15  See also North American Free Trade Agreement (adopted 17 December 1992, entered into force 1 January 1994) CTS 1994 No 2, art 1110 (Appendix 1 below) (NAFTA); ASEAN Comprehensive Investment Agreement (signed 26 February 2009, entered into force 29 March 2012) art 14(2) (b) (Appendix 3 below) (ACIA).

17  See the Tribunal’s question in Himpurna California Energy Ltd v PT (Persero) Perusahaan Listruik Negara (UNCITRAL, 1999, Paulsson P, de Fina & Setiawan) Mealey’s Int Arb Rep 14(12) (12/99) A-I, where it asks ‘whether the result, in that case and often elsewhere, is not to create an illusion of scientific analysis to mask the reality of subjective approximations’.

18  Siemens AG v Argentina (Award) ICSID Case No ARB/02/8, 14 ICSID Rep 513, IIC 227 (2007, Rigo Sureda P, Bello Janiero & Brower) para 403; Venezuela Holdings BV v Venezuela (Award) ICSID Case No ARB/07/27, IIC 656 (2014, Guillaume P, El-Kosheri & Kaufmann-Kohler) para 389.

19  Venezuela Holdings BV v Venezuela para 388.

20  TR Stauffer, ‘Valuation of Assets in International Takings’ (1996) 17 Energy LJ 459, 460.

21  PD Friedland and E Wong, ‘Measuring Damages for the Deprivation of Income-Producing Assets: ICSID Case Studies’ (1991) 6 ICSID Rev-FILJ 400, 404.

22  World Bank Guidelines art IV(5)(ii) (Appendix 13 below).

23  WM Reisman and RD Sloane, ‘Indirect Expropriation and its Valuation in the BIT Generation’ (2003) 74 BYIL 115, 138.

24  Friedland and Wong, ‘Measuring Damages for the Deprivation of Income-Producing Assets: ICSID Case Studies’ 404; WC Lieblich, ‘Determining the Economic Value of Expropriated Income-Producing Property in International Arbitrations’ (1991) 8 J Int’l Arb 59, 63–9.

25  Asian Agricultural Products Ltd (AAPL) v Sri Lanka (Award) ICSID Case No ARB/87/3, 4 ICSID Rep 245, IIC 18 (1990, El-Kosheri P, Goldman & Asante (dissenting)).

26  ibid para 103.

27  ibid para 98.

28  ibid para 97.

29  Friedland and Wong, ‘Measuring Damages for the Deprivation of Income-Producing Assets: ICSID Case Studies’ 421.

30  AAPL v Sri Lanka para 291.

31  World Bank Guidelines art IV(6)(iii) (Appendix 13 below).

32  Occidental Petroleum Corp v Ecuador (Award) ICSID Case No ARB/06/11, IIC 561 (2012, Fortier P, Stern & Williams).

33  ibid para 708.

34  World Bank Guidelines art IV(6)(i).

35  World Bank Guidelines Report (1992) 31 ILM 1366 para 42.

36  WC Lieblich, ‘Determinations by International Tribunals of the Economic Value of Expropriated Enterprises’ (1990) 7 J Int’l Arb 37, 38.

38  CMS Gas Transmission Co v Argentina (Award) ICSID Case No ARB/01/8, 14 ICSID Rep 158, IIC 65 (2005, Orrego Vicuña P, Lalonde & Rezek).

39  ibid para 403.

40  CME Czech Republic BV v Czech Republic (Final Award) 9 ICSID Rep 264, IIC 62 (UNCITRAL, 2003, Kühn C, Brownlie & Schwebel) (CME I).

41  Azurix Corp v Argentina (Award) ICSID Case No ARB/01/12, 14 ICSID Rep 374, IIC 24 (2006, Rigo Sureda P, Lalonde & Martins).

42  ibid para 425.

43  ibid para 429.

44  I Marboe, Calculation of Compensation and Damages in International Investment Law (2009).

45  M Kantor, Valuation for Arbitration (2008).

46  See the discussion of the appropriate approach to valuation in Rumeli Telekom AS v Kazakhstan (Decision on Annulment) ICSID Case No ARB/05/16, IIC 420 (Schwebel P, McLachlan & Silva Romero) paras 141–51.

47  Factory at Chorzów (Germany v Poland) (Merits) (1928) PCIJ Rep Series A No 17.

48  ibid 29.

49  ibid 47.

50  ibid 49.

51  ibid 52. The Chorzów Factory approach has been validated by numerous investment arbitration tribunals. A notably comprehensive analysis can be found at paras 479–500 of ADC Affiliate Ltd v Hungary (Award) ICSID Case No ARB/03/16, 15 ICSID Rep 534, IIC 1 (2006, Kaplan P, Brower & van den Berg).

52  Starrett Housing Corp v Iran (1987) 16 Iran-USCTR 112.

53  ibid 196.

54  ibid 201.

55  Phillips Petroleum Co Iran v National Iranian Oil Co (1989) 21 Iran-USCTR 79.

56  ibid 123.

57  Amoco International Finance Corp v Iran (1987) 15 Iran-USCTR 189.

59  See ibid 77, for a criticism of the Amoco Tribunal’s rejection of the DCF method. Lieblich’s article is a robust defence of the DCF method as the only basis on which the decisions of tribunals can ‘be anything more than arbitrary results reached under the guise of abstract legal standards’ (Lieblich, 60). However, while Lieblich’s analysis is correct as a matter of economics, it fails to recognise the political and other considerations behind many of the decisions.

60  Phillips Petroleum Co Iran para 123.

61  Amco Asia Corp v Indonesia (Resubmitted Case: Award) ICSID Case No ARB/81/1, 1 ICSID Rep 376 (1990, Higgins P, Lalonde & Magid). See also Benvenuti & Bonfant Srl v Democratic Republic of Congo (Award) ICSID Case No ARB/77/2, 1 ICSID Rep 330, 340 (1980, Trolle P, Bystricky & Razafindralambo); AGIP SpA v People’s Republic of the Congo (Award) ICSID Case No ARB/77/1, 1 ICSID Rep 306 (1979, Trolle P, Dupuy & Rouhani).

62  Amco Asia Corp v Indonesia (Resubmitted Case: Award) para 197.

63  ibid.

64  Société Ouest Africaine des Bétons Industriels [SOABI] v Senegal (Award) ICSID Case No ARB/82/1, 2 ICSID Rep 164 (1988, Broches P, Mbaye & Schultsz).

65  Liberian Eastern Timber Corp [LETCO] v Liberia (Award) ICSID Case No ARB/83/2, 2 ICSID Rep 343 (1986, Cremades P, Pereira & Redfern).

66  CMS Gas Transmission Co v Argentina (Award).

67  ibid para 410.

68  ibid para 416. Other tribunals adopting the DCF methodology include National Grid plc v Argentina (Award) IIC 361 (UNCITRAL, 2008, Rigo Sureda P, Kessler & Miguel Garro) paras 275–7; Enron Corp v Argentina (Award) ICSID Case No ARB/01/3, IIC 292 (2007, Orrego Vicuña P, van den Berg & Yves-Tschanz) para 385; Walter Bau Ag v Thailand (Award) IIC 249 (UNCITRAL, 2009, Barker P, Bunnag & Lalonde) para 14.26. In one particularly large DCF-based award, Occidental Petroleum Corp v Ecuador (Award) ICSID Case No ARB/06/11, IIC 561 (2012, Fortier P, Stern & Williams), both parties agreed that it was a suitable methodology (para 690), as they did in Guaracachi America Inc v Bolivia (Award) PCA Case No 2011-17, IIC 628 (2014, Miguel Júdice P, Conthe & Vinuesa) para 445.

69  CME I.

70  Agreement on Encouragement and Reciprocal Protection of Investments (Netherlands–Czech and Slovak Republic) (signed 29 April 1991, entered into force 1 October 1992) 2242 UNTS 206, Tractatenblad 1992, 142.

71  CME I para 497.

72  ibid paras 514 and 560.

73  Enron Corp v Argentina paras 424–39. See also National Grid plc v Argentina para 275 and Lemire v Ukraine (Award) ICSID Case No ARB/06/18, IIC 485 (2011, Fernández-Armesto P, Paulsson & Voss) paras 298–309.

74  Cargill Inc v Mexico (Award) ICSID Case No ARB(AF)/05/2, IIC 479 (NAFTA, 2009, Pryles P, Caron & McRae) para 445.

75  Lemire v Ukraine para 248.

76  ibid para 249; LG&E Energy Corp v Argentina (Award) ICSID Case No ARB/02/1, IIC 295 (2007, Maekelt P, Rezek & van den Berg) para 30.

77  Lemire v Ukraine para 249. See also Quasar de Valores SICA SA v Russia (Award) SCC Case No 24/2007, IIC 557 (2012, Paulsson C, Brower & Landau) paras 215–7.

78  ILC ‘Responsibility of States for Internationally Wrongful Acts: Draft Articles with Commentaries’ (Crawford, Special Rapporteur) [2001] 2(2) YB ILC 30, art 36 commentary para 26.

79  See 9.58 et seq below.

80  Tecnicas Medioambientales Tecmed SA v Mexico (Award) ICSID Case No ARB(AF)/00/2, 10 ICSID Rep 130, IIC 247 (2003, Grigera Naón P, Bernal Verea & Fernández Rozas).

81  ibid para 186.

82  Compañia de Aguas del Aconquija SA and Vivendi Universal v Argentina (Award) ICSID Case No ARB/97/3, IIC 307 (2007, Rowley P, Bernal Verea & Kaufmann-Kohler) (Vivendi II) para 8.3.3.

83  Metalclad Corp v Mexico (Award) ICSID Case No ARB(AF)/97/1, 5 ICSID Rep 209, IIC 161 (NAFTA, 2000, Lauterpacht P, Civiletti & Siqueiros).

84  Wena Hotels Ltd v Egypt (Award) ICSID Case No ARB/98/4, 6 ICSID Rep 67, IIC 273 (2000, Leigh P, Fadlallah & Wallace).

85  Biloune and Marine Drive Complex Ltd v Ghana Investments Centre (Awards) 95 ILR 183 (1989 and 1990, Schwebel P, Monroe Leigh & Wallace).

86  See also Walter Bau Ag v Thailand (Award) paras 14.14 to 14.19 where the Tribunal specifically addresses the cases of Siemens AG v Argentina (Award); Wena Hotels Ltd v Egypt; Metalclad v Mexico; PSEG Global Inc v Turkey (Award) ICSID Case No ARB/02/5, IIC 198 (2007, Orrego Vicuña P, Fortier & Kaufmann-Kohler); Tecmed v Mexico and MTD Equity Sdn Bhd v Chile (Award) ICSID Case No ARB/01/7, 12 ICSID Rep 6, IIC 174 (2004, Rigo Sureda P, Lalonde & Oreamuno Blanco) as instances where a DCF analysis was not appropriate due to an underlying lack of profitability. Further examples include Siag v Egypt (Award) ICSID Case No ARB/05/15, IIC 374 (2009, Williams P, Pryles & Orrego Vicuña); Impregilo SpA v Argentina (Award) ICSID Case No ARB/07/17, IIC 498 (2011, Danelius P, Brower & Stern (dissenting)); and Tenaris SA v Venezuela (Award) ICSID Case No ARB/11/26, IIC 764 (2016, Beechey P, Kessler & Landau).

87  Rumeli Telekom AS v Kazakhstan (Award) ICSID Case No ARB/05/16, IIC 344 (2008, Hanotiau P, Boyd & Lalonde).

88  ibid para 726.

89  ibid para 811.

90  Al-Bahloul v Tajikistan (Award) SCC Case No 064/2008, IIC 475 (2010, Hertzfeld C, Happ & Zykin). In this case the Respondent State did not appear. The Tribunal’s analysis is sound but, in common law systems, judgments following uncontested hearings are of less precedential value.

91  ibid para 75. See also Micula v Romania (Award) ICSID Case No ARB/05/20, IIC 621 (2013, Lévy P, Abi-Saab & Alexandrov) paras 1006–10. This conclusion was accepted in Abi-Saab’s Separate Opinion despite the fact that the lack of a track record of profitability rendered the claim ‘highly speculative’ (para 16). See also Quiborax SA v Bolivia (Award) ICSID Case No ARB/06/2, IIC 739 (2015, Kaufmann-Kohler C, Lalonde & Stern).

92  LG&E Energy Corp v Argentina (Award) ICSID Case No ARB/02/1, IIC 295 (2007, Maekelt P, Rezek & van den Berg).

93  ibid para 89–90.

94  Libyan American Oil Co [LIAMCO] v Libya (1982) 62 ILR 140 (Ad Hoc, 1977, Mahmassani).

95  ibid para 209, citing Asylum Case (Colombia v Peru) (Merits) [1950] ICJ Rep 266, 277.

96  ibid para 214.

97  Siag v Egypt (Award) ICSID Case No ARB/05/15, IIC 374 (2009, Williams P, Pryles & Orrego Vicuña).

98  ibid paras 575–6.

99  ADC Affiliate Ltd v Hungary (Award) ICSID Case No ARB/03/16, 15 ICSID Rep 534, IIC 1 (2006, Kaplan P, Brower & van den Berg) para 521.

100  Kuwait v American Independent Oil Co (AMINOIL) (Final Award) (1984) 66 ILR 518 (Ad Hoc, 1982, Reuter P, Hamed Sultan & Fitzmaurice).

101  ibid para 148.

102  ibid para 164.

103  CME I (Separate Opinion, Brownlie).

104  ibid para 121.

105  ibid paras 73–5 and 77.

106  Himpurna California Energy Ltd v PT (Persero) Perusahaan Listruik Negara (UNCITRAL, 1999, Paulsson P, de Fina & Setiawan), Mealey’s Int Arb Rep 14(12) (12/99) A-1.

107  ibid A-51.

108  H Weisburg and C Ryan, ‘Means to be Made Whole: Damages in the Context of International Investment Arbitration’ in R Kreindler and Y Derains (eds), Evaluation of Damages in International Arbitration (2006) 165, 178.

109  CMS Gas Transmission Co v Argentina paras 411–17.

110  ILC ‘Responsibility of States for Internationally Wrongful Acts: Draft Articles with Commentaries’ (Crawford, Special Rapporteur) [2001] 2(2) YB ILC 30.

111  DW Bowett, ‘State Contracts with Aliens: Contemporary Developments on Compensation for Termination or Breach’ (1988) 59 BYIL 49, 63.

112  1st edn, 2007 para 9.67.

113  ADC v Hungary para 482.

114  ibid paras 484–95.

115  Agreement on the Promotion and Reciprocal Protection of Investments (France–Argentina) (signed 3 July 1991, entered into force 3 March 1993) 1728 UNTS 282 art 5(2).

116  Vivendi II para 8.2.7.

117  See Siag v Egypt para 541.

118  Siemens AG v Argentina (Award) ICSID Case No ARB/02/8, 14 ICSID Rep 513, IIC 227 (2007, Rigo Sureda P, Bello Janiero & Brower) para 352.

119  ibid para 361.

120  ibid paras 386–7.

121  Funnekotter v Zimbabwe (Award) ICSID Case No ARB/05/6, IIC 370 (2009, Guillaume P, Cass & Wasi Zafar).

122  Amoco International Finance Corp v Iran (1987) 15 Iran-USCTR 189 para 197.

123  Phillips Petroleum Co Iran v National Ir