This article is an extract from The Investment Treaty Arbitration Review, 7th Edition. Click here for the full guide.


 

I Introduction

As the title of this chapter suggests, the twin principles of contributory fault (that a tribunal may take into account the extent to which a claimant's own conduct contributed to its loss) and mitigation (that a claimant has a duty to mitigate that loss) are two of the ways in which a respondent can mount a defence against a damages claim. It is generally recognised that any relevant contribution on the part of the claimant will typically take place prior to the breach that forms the substance of a claim (or prior to the claimant reasonably having knowledge of that breach), whereas any opportunities for mitigation will generally arise subsequently.

Other defences covered here include corruption (that the agreement under which the original investment was made may have been procured corruptly), on which there are several cases that deserve comment. Also covered is the (potentially) more quantitative issue of investment risk (and, in particular, country risk, which may feature directly in circumstances where losses are assessed using discounted cash flow (DCF) models). Investors take on risk in making investment decisions abroad, including, potentially, the risk of expropriation. The question of whether expropriation risk should be taken into account when assessing damages is topical.2

II Contributory fault

The possibility that a claimant may have contributed to his or her own loss is recognised within the International Law Commission's Articles of State Responsibility (ILC Articles), adopted in draft form by the Commission in 2001, and generally cited in the cases reviewed below. Under Article 31, an infringing state 'is under an obligation to make full reparation for the injury caused by the internationally wrongful act'; and Article 39 recognises: 'In the determination of reparation, account shall be taken of the contribution to the injury by wilful or negligent action or omission of the injured State or any person or entity in relation to whom reparation is sought.'3

Contributory fault has been taken into account, through a proportionate reduction in the losses identified, in a small number of cases, as discussed below. Principles established in these cases are that the contribution needs to be material, that the respondent bears the burden of proof, and that the tribunal has discretion in deciding on the scale of the proportionate reduction applied. Those reductions have varied from 25 per cent (Occidental, Yukos and STEAG) to 50 per cent (MTD and UAB).

The MTD case4 involved the development of a large-scale real estate project in Chile and required re-zoning permits to be secured from the Chilean Ministry of Housing and Urban Development, which were not granted. The tribunal found the Foreign Investment Commission (FIC) in Chile, which was the counterparty to the agreement with MTD) at fault in authorising the investment, but also found MTD to have acted unwisely in proceeding as far as it had without greater due diligence5 and therefore reduced damages by 50 per cent.6

The tribunal did not explain its reasons for choosing a 50 per cent reduction rather than any other number, but the International Centre for Settlement of Investment Disputes (ICSID) annulment committee that reviewed the award in 2007 supported the decision, noting the different nature of the parties' contributions to the claimant's loss, and the corresponding margin of discretion available to the tribunal.7

The finding against Chile was narrow (not relating to the failure on Chile's part to grant the permits, but rather to the FIC's initial approval of the project in circumstances in which the permits were unlikely to be granted), and the contribution of MTD's business decisions to its ultimate loss was found to be significant. The tribunal noted:

The BITs [bilateral investment treaties] 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.8

Similarly, the tribunal in the UAB case9 applied a 50 per cent deduction to the final damages in a December 2017 award on account of contributory fault. In annulment proceedings, Latvia claimed that the 50 per cent reduction was arbitrary. In the April 2020 decision on annulment, the committee recognised that 'more detailed reasoning could have been provided by the tribunal with respect to its decision that there had been contributory fault'10 but added, citing the MTD case: 'It is not uncommon for tribunals to use their discretion to estimate how damages should be apportioned once it is established that both parties contributed to the loss.'11

The Occidental case12 was lengthy and complex and defies a straightforward summary. The original breach in the underlying agreement was one by Occidental, but the subsequent termination of the agreement by Ecuador was found to be a disproportionate response, for which Occidental was entitled to damages. The tribunal allocated a 75 per cent responsibility for those losses to Ecuador (and 25 per cent to Occidental).13

The reasons for the split (as against any other numbers) were again not set out. The tribunal cited the decision of the ICSID annulment committee in MTD in explaining the discretionary nature of the allocation. Professor Brigitte Stern, however, also referred to the MTD case and stated, in the dissenting opinion, that a 50/50 split would have been more reasonable because the claimants have acted 'both very imprudently and illegally'.

In the Yukos suite14 of arbitrations, the tribunals15 also adopted a 75/25 split based on contributory fault. The Yukos awards are lengthy and wide-ranging. On the question of contributory fault, the tribunal noted that there needs to be a 'sufficient causal link between any wilful or negligent act or omission of the Claimants . . . and the loss Claimants ultimately suffered'.16

The tribunal also noted that the contribution 'must be material and significant'; and, following MTD: 'In this regard, the Tribunal has a wide margin of discretion in apportioning fault'. Helpfully, the tribunal also noted three points that emerged from their review of cases in which the issue of contributory fault had been at issue:

Firstly, the legal concept of contributory fault must not be confused with the investor's duty to mitigate its losses . . .Secondly, there are cases where the contributory fault of the investor, while it may have increased the loss which it sustained, was unrelated to the wrongdoing of the State . . .Finally, the Tribunal identified certain decisions where the tribunals found that the victim contributed to the State's wrongful conduct. The contributory fault of the investor in those cases provoked the wrongful conduct of the State.17

The MTD case was highlighted by the tribunal as being of the first type (unrelated to the wrongdoing), whereas the Occidental case was identified as being of the second type (provoking the wrongful conduct of the state). The circumstances of the Yukos case, in which the tribunal found that some elements of Yukos' tax minimisation strategies were abusive and unlawful, are likewise of this kind. The tribunal noted:18

While . . . Respondent's tax assessments and tax collection efforts against Yukos were not aimed primarily at the collection of taxes, but rather at bankrupting Yukos and facilitating the transfer of its assets to the State, it cannot ignore that Yukos' tax avoidance arrangements in some of the low-tax regions made it possible for Respondent to invoke and rely on that conduct as a justification of its actions against Mr. Khodorkovsky and Yukos.

There was, as a result, 'a sufficient causal link between Yukos' abuse of the system in some of the low-tax regions and its demise' to lead to a finding of contributory fault.

In the recent STEAG case,19 another dispute arising from Spain's change to the regulatory subsidy regime for concentrated solar power producers during the period 2012–2014, the tribunal found, by a majority, that Spain was responsible for violating the claimant's expectations on the price regime.

The majority, however, found that STEAG had contributed to its losses by making investments between the second half of 2012 and 2014, when Spain's new measures were already announced or implemented. Although STEAG argued that the investments were a part of the project's financing requirements, the tribunal decided on a 25 per cent reduction of damages reflecting STEAG's partial contribution in its losses.20

Other cases in which the issue of contributory fault have been considered include Bear Creek21 and Burlington Resources.22 None of these cases led to a finding of contributory fault. In Bear Creek, the tribunal noted the principles involved in such a finding:

the doctrine of contributory fault under international law requires that the party advocating for its application demonstrate – in addition to contribution to the investor's harm – that the investor's willful or negligent conduct or omission materially and significantly contributed to its harm, directly causing it. There must also be a sufficient causal link between the negligent or willful act or omission and the harm, in accordance with ILC Article 31. Respondent cannot meet its burden of proof for such a finding.23

In a partial dissenting opinion, Professor Philippe Sands disagreed on this point, suggesting that the claimant's actions in and around its mining operations in Peru had itself contributed to the social unrest in the region that, in turn, had led to the withdrawal of the mining concession. There was, in Professor Sands' view, enough evidence to establish a connection – at least a 'partially causal relationship' – between the claimant's actions and the loss that was suffered.24

In the Burlington Resources case, the argument put forward by Ecuador was that Burlington's failure to pay certain taxes was the triggering factor that led to its operations being taken over by Peru (and therefore constituted a contributory fault). A majority of the tribunal disagreed, finding that Burlington's failure to pay the taxes was neither a triggering nor a decisive factor behind the eventual expropriation.25 The tribunal appears to have relied in part on the commentary to Article 31 of the ILC Articles:

The Commentary to Article 31 clarifies that 'unless some part of the injury can be shown to be severable in causal terms from that attributed to the responsible State, the latter is held responsible for all the consequences, not being too remote, of its wrongful conduct'.26

The fact that there was a (firm) dissension on what is effectively a factual issue – whether the expropriation would have occurred, absent the failure on Burlington's part to pay the taxes – only emphasises the difficulties involved in a finding of contributory fault. That is notwithstanding the several expositions of, and some convergence in, tribunals' understanding of the underlying principles (as summarised above).

III Mitigation

The duty to mitigate arises after a breach and, therefore, is relevant in relation to quantum, rather than in relation to liability. It is not that the duty to mitigate imposes any kind of legal obligation but rather that a failure to mitigate will reduce the relevant component of losses (to the extent that it was reasonably capable of being mitigated).27

The circumstances in which a tribunal will identify a failure to mitigate (and the underlying principles of international law) were extensively discussed and usefully summarised in Clayton and Bilcon v. Canada,28 in an award delivered in January 2019. Any failure needs to consist of an 'unreasonable failure by the claimant to act subsequent to the breach of the treaty, where it could have reduced the damages arising' or the 'unreasonable incurring of expenses by the claimant subsequent to a treaty breach, which results in increasing the size of its claim'.29

Both points demonstrate that a failure to mitigate might equally be thought of as a break in the chain of causation. Expenses that were unreasonably incurred, for example, could not be said to have been caused by the breach itself. The test of reasonableness operates to constrain the circumstances in which a failure to mitigate will be recognised.

The nature of that constraint was discussed in another the Inicia case.30 Here, the issue concerned actions taken by Hungary in relation to agricultural land, with the respondents arguing that the claimants could have mitigated their losses by leasing alternative land. The tribunal noted that it was 'not prepared to speculate whether the Claimants should have exercised a better business judgment'.31

It is perhaps not surprising that tribunals will not generally wish to second guess the actions taken by a claimant in response to a breach. The principle that the reasonableness of any given action needs to be assessed against what was known at the time precludes any sort of hindsight-based assessment of the claimant's actions, and claimants must generally be entitled to a presumption that they have at least tried to act in their own self-interest.

The tribunal in the Cairn case32 in an award issued in December 2020 noted:

A mitigation defence is difficult to prove, given that it is in a claimant's own best interest to minimise its loss. As a rule, it will require sufficient evidence to show that a claimant's conduct (action or inaction) following the Respondent's breach was unreasonable, abusive or against its own economic interests. For this reason, tribunals are seldom persuaded by speculative options of mitigation that are proposed in hindsight.33

That may be why, although discussions of the duty to mitigate feature regularly in arbitral awards (see, for instance, Olympic v. Ukraine34 and Eco Oro v. Colombia35), actual findings of a failure to mitigate are rare.36 Rather, the primary effect of the existence of a recognised duty to mitigate may be to discourage claims for losses being framed in such a way that the losses would in principle have been straightforwardly mitigable. As the tribunal noted in Clayton and Bilcon: 'By its nature, the duty to mitigate is a restriction on compensatory damages.'37

Thus, for example, losses relating to a failure of supply of a given good might be straightforwardly mitigable if an alternative source of supply were easy to identify. In these circumstances, the claimable loss would be limited to the price differential between the two supply sources (plus any other incremental costs incurred). The limitation on losses might in principle be imposed by a tribunal, following a finding of a failure on the claimant's part to mitigate, but would arise more normally as a consequence of the claimant, or the claimant's quantum expert, limiting the amount of losses identified or damages sought.

Similarly, losses claimed in relation to a lost opportunity to invest will be mitigable, in principle, if alternative opportunities were available. Such losses will be claimable, in principle, only to the extent that the opportunity that was lost would have provided returns in excess of those otherwise available. Tribunals are typically reluctant to interpret a duty to mitigate as a duty to seek, or accept, alternative investment opportunities (as in AIG v. Kazakhstan, for example);38 but these types of opportunities are built into the logic of any DCF calculation.

That is because the discount rate used in a DCF calculation – typically described as a cost of capital – is better understood as an opportunity cost of capital (reflecting the returns otherwise available on an investment of equivalent scale and risk).39 Any DCF calculation (for a lost opportunity) therefore already carries within it an implicit reflection of the investor's duty to mitigate. Put another way, the results of such a calculation are in reality the excess profits (those in excess of the profits otherwise available on investments of equivalent size and risk) that are said to have been available in relation to the investment opportunity said to have been lost. We return to this topic further in Section V, below.

IV Corruption

Corruption and bribery have been invoked as a defence by host states on many occasions; on the basis that an investment obtained through corruption is not protected by the treaty under which the action is brought.40

The very nature of corruption and bribery, namely disguise and opacity, makes their identification difficult, however. Several recent awards show the lengths to which tribunals have been obliged to go to reach a decision on the facts before them. One such is the Niko Resources case, in which a 500-page decision, dealing only with the corruption claim, was issued in February 2019.41

The Niko Resources case concerned a joint venture agreement and a gas sale and purchase agreement in Bangladesh. It had been established in the prior decision on jurisdiction that there had been acts of corruption, but not that those had been instrumental in securing the agreements under which the investments were made: 'there is no link of causation between the established acts of corruption and the conclusion of the agreements, and it is not alleged that there is such a link'.42

In its decision in relation to the corruption claim, the tribunal set out an extensive review of each of the allegations made by the respondent, before concluding that they were unfounded.

A second case in which corruption issues were addressed is Lao Holdings, for which the award was issued in early August 2019.43 The case concerns the fallout of the Dutch Lao Holdings and its Laotian business partner, ST Holdings. The companies originally planned to build casinos and slot machine clubs in Laos. Although the tribunal dismissed the case on the grounds that Lao Holdings failed to establish the facts necessary to establish liability, it nevertheless reviewed the corruption claim asserted by the government of Laos.

The tribunal laid out its view as to the appropriate standard of proof required to reach a finding of corruption:

In the Tribunal's view there need not be “clear and convincing evidence” of every element of every allegation of corruption, but such “clear and convincing evidence” as exists must point clearly to corruption. An assessment must therefore be made of which elements of the alleged act of corruption have been established by clear and convincing evidence, and which elements are left to reasonable inference, and on the whole whether the alleged act of corruption is established to a standard higher than the balance of probabilities but less than the criminal standard of beyond reasonable doubt, although of course proof beyond a reasonable doubt would be conclusive. This approach reflects the general proposition that the graver the charge, the more confidence there must be in the evidence relied on.44

The appropriate standard of proof was a central issue when the Hague Court of Appeal overturned an International Chamber of Commerce (ICC) award, in late 2019, against a subsidiary of Venezuela's state-owned oil and gas company, Petróleos de Venezuela, SA, on the ground that the underlying contract was procured through corruption. In the domestic court's view, the ICC tribunal had applied overly strict requirements by demanding 'clear and convincing evidence' although the allegations 'undoubtedly raise questions as to the (legitimate) nature of [the company's] conduct'.45

In the Infinito Gold v. Costa Rica case46 concluded in June 2021, although Costa Rica withdrew its objection that the claimant's procurement of a mining concession involved corruption, the tribunal reviewed the corruption-related evidence, noting that it had a duty to do so since corruption allegations 'raise an issue of international public policy'.47 The tribunal's review led to no findings of corruption even under a circumstantial evidence (or 'red flags') approach:

When assessing the record and reaching the findings just set out, the Tribunal has taken into consideration that it is notoriously difficult to prove corruption and that, as a result, tribunals tend to focus on circumstantial evidence, relying on indicia or red flags. Even adopting such less demanding standard of proof, it cannot conclude that the 2008 Concession was procured by corruption.48

In the Penwell v. The Kyrgyz Republic decision of October 2021,49 however, the tribunal found that Penwell committed two separate acts of corruption and, therefore, dismissed Penwell's expropriation claim. On the appropriate standard of proof, the tribunal adopted the 'red flags' approach noting:

It is undeniable that the red flags methodology is increasingly used by arbitral tribunals to consider the circumstances before them with an intellectually honest and pragmatic eye, reading between the lines where necessary, and/or 'connecting the dots', in order to grasp the true picture and expose the fraudulent activities involved.50

Recent decisions, therefore, show some consistency in applying a flexible approach to rules of evidence in arriving at a decision when corruption has been alleged. The two standards – that evidence should be either 'clear and compelling' or that it should establish corruption 'on a balance of probabilities' – have been seen as extremes within which tribunals should form their own views as to whether they are satisfied that the existence of corruption has been established.

V Investor risk

The MTD award noted above, in relation to contributory fault, made the point that bilateral investment treaties (BITs) 'are not an insurance against business risk'. The question that has arisen – particularly in the context of valuing an investor's investment – is the extent to which investors are protected from the risk that their investment might be expropriated.

The question arises particularly in relation to the discount rate to be applied in a valuation based on DCFs: the discount rate is often adjusted to reflect country risk,51 but if there is a component of that country risk that relates to the possibility of expropriation, should that component be included? Including a country risk premium serves to increase the discount rate applied, and thereby to reduce the resulting valuation. The award in the ConocoPhilips case included some discussion on this point.52

The tribunal noted first that quantifying a country risk premium using government bond yields would, in the case of Venezuela, be inappropriate: 'to apply to the Projects a country risk premium that reflects the likelihood that Venezuela will default on billions of dollars of sovereign debt is entirely unjustified'.53 The rationale follows in the next paragraph: 'The inclusion of unlawful state action would result in higher discount rates and thus allow Venezuela to benefit from its own unlawful acts.'54

As the award demonstrates, however, it is easier to elucidate a principle than it is to apply it in practice. There is no settled view among the valuation community as to how to calculate a country risk premium, let alone how to identify a component of the result that relates to the risk of lawful or unlawful expropriation. There was, perhaps for the same reason, no convergence of views between the quantum experts involved in this case.55

In concluding, the tribunal clarified that although the BIT provides protection against illegal acts, the same level of protection is not provided in respect of 'legal intrusions from the state affecting the economics of an investment'56 and, therefore, modified the position put forward by the claimants:57

one of the basic assumptions of the Claimants' experts, i.e. that “in these proceedings the discount rate must be free of expropriation risk and of the risk of wrongful taxation” must be tempered. The Projects were not free of the risk of expropriation, provided it was lawful within the framework of Article 6 of the BIT.

This same issue – the boundary between lawful acts that affect the economics of an investment, and unlawful measures – has been explored in a number of recent awards on solar energy cases, which concern the legality or otherwise of measures taken by governments of European Union Member States to reduce the level of regulatory support provided to investors in renewable electricity projects.

In 2007, Spain introduced a feed-in tariff scheme under which investors in solar, wind and other renewable assets were to receive a fixed price for every kWh of electricity produced.58 For photovoltaic solar plants, the tariff was provided for the lifetime of the plant (albeit with a 20 per cent reduction in the rate after 25 years). From 2010 to 2013, Spain enacted a series of measures, including a cap on the number of operating hours that would be eligible for the tariff, and in 2014, replaced the original tariff scheme with a different, new scheme for existing solar plants.

The various tribunals dealing with the numerous arbitrations that resulted generally agreed that 'the fair and equitable treatment standard does not give a right to regulatory stability per se. The state has a right to regulate, and investors must expect that the legislation will change, absent a stabilization clause or other specific assurance giving rise to a legitimate expectation of stability'.59

They also agreed, however, that 'the right to regulate must be subject to limitations if investor protections are not to be rendered meaningless'.60 Those limitations were expressed by the Charanne tribunal: 'as long as the changes are not capricious or unnecessary and do not amount to suddenly and unpredictably eliminate the essential characteristics of the existing regulatory framework'.61

In all the cases covered here, the tribunals found that the several piecemeal changes made by Spain prior to the introduction of the new regulatory scheme in June 2014 did not violate the treaty. Their opinions diverged, however, and quite significantly, with respect to the new scheme.

In a majority of cases, the tribunals found that the introduction of the new regulatory scheme was unlawful. The change was described as 'total and unreasonable' in the Eiser case;62 as falling 'outside the acceptable range of legislative and regulatory behaviour' in the Novenergia case;63 as having 'crossed the line from a non-compensable regulatory measure to a compensable breach' in the Foresight case;64 and as being 'radically different' from the established framework in the Watkins case.65

In some cases, however, the tribunals found that investors could not reasonably have anticipated that the tariff regime would remain unchanged throughout the life of their investments, and that Spain's reforms had not breached its obligation of ensuring a return at reasonable levels.66 In Isolux, for example, the tribunal noted its view that '[t]he only legitimate expectation of the Claimant was a reasonable return on its investment'.67 The same view was expressed more recently by the tribunals in Infracapital68 and Sevilla Baheer.69 In the FREIF case,70 the tribunal also accepted the reasonable return approach and, referring to the investor's due diligence, found that 'FREIF took the risk in making its investment that there could be a re-adjustment in the tariff regime. It may have lost the opportunity to earn higher profits, but it did not lose the expectation of a reasonable return'.

VI Conclusions

The twin principles of contributory fault (that a tribunal may take into account the extent to which a claimant's own conduct contributed to its loss) and mitigation (that a claimant has a duty to mitigate that loss) are perhaps the longest standing of the defences against a damages claim, and appear to be the most settled.

The other defences covered in this chapter – corruption (that the agreement under which the original investment was made may have been procured corruptly) and investor risk (that the actions from which the investor has suffered reflect normal commercial risk) – appear to have seen more development during recent years. The conclusions of some of the tribunals involved in the solar energy cases – to the effect that investors' legitimate expectations cannot be of anything more than a reasonable return on their investment – are potentially far-reaching in their application and seem set to be the subject of debate for some time.