As the title 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 generally take place prior to the breach that forms the substance of a claim (or prior to the claimant reasonably having knowledge of that breach); while 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 recent 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


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'; while Article 39 recognises that:3

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.

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) to 50 per cent (MTD).

The MTD case4 involved the development of a large-scale real estate project in Chile; and required rezoning permits to be secured from the Chilean Ministry of Housing and Urban Development, which were not granted. The tribunal found Chile (the Foreign Investment Commission (FIC), who were the counter-party 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 diligence;5 and therefore reduced damages by 50 per cent.6

The tribunal did not explain their reasons for choosing a 50 per cent reduction rather than any other number, but the 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 their ultimate loss was found to be significant. The tribunal noted:8

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.

The Occidental case9 was lengthy and complex; and defies straightforward summarisation. 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).10

The reasons for the split (as against any other number) 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.

In a strongly worded dissenting opinion, however, Professor Brigitte Stern, referring to the same case, disagreed with the split adopted by the majority:11

It is interesting to note that in the MTD case both the tribunal and the ad hoc committee have endorsed a 50/50 split on the sole ground that the claimant had acted imprudently from a business point of view though not illegally. Here the split 50/50 would have been even more justified, as the Claimants have acted both very imprudently and illegally.

In the Yukos suite12 of arbitrations, the tribunals13 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'.14

The tribunal also noted that the contribution 'must be material and significant' and, following MTD, that '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. Those are:15

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.

The MTD case was highlighted by the tribunal as being of the first type (unrelated to the wrongdoing); while 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:16

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.

Subsequent cases in which the issue of contributory fault have been considered include Bear Creek,17 Burlington Resources18 and Perenco.19 None of these led to a finding of contributory fault. In Bear Creek, the tribunal noted the principles involved in such a finding:20

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.

A partial dissenting opinion from Professor Philippe Sands QC disagreed on this point – suggesting that the claimant's actions in and around their mining operations in Peru had themselves contributed to the social unrest in the region that had in turn 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.21

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 their 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.22 The tribunal appear to have relied in part on the commentary to Article 31 of the ILC Articles:23

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”.

The fact that there was a (firm) dissention 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, the tribunals' understanding of the underlying principles (as set out in summary form above).

There appear to have been no such difficulties in the Perenco case, in which Ecuador argued, again, that the claimant had contributed to its own losses, by a failure to pay the same taxes that were at issue in the Burlington Resources case (but instead paying the amounts due into an escrow account). In this instance, the non-payment occurred only after the initiation of arbitration by Perenco; and turned out to be compatible with provisional measures ordered by the tribunal after the arbitration commenced. The tribunal concluded that 'it is wrong to equate a party's zealous protection of its legal rights and interests with wilful conduct or contributory negligence within the meaning of the ILC Articles'.24


The duty to mitigate arises after a breach, and is therefore 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).25

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/Bilcon v. Canada,26 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'.27

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 recent award: the Inicia case.28 There, 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'.29

It is perhaps unsurprising 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.

That may be why, although discussions of the duty to mitigate feature regularly in arbitral awards, actual findings of a failure to mitigate are rare.30 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 the Clayton/Bilcon case:31 'By its nature, the duty to mitigate is a restriction on compensatory damages'.

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 such 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 in principle be mitigable if alternative opportunities were available. Such losses will in principle be claimable 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);32 but such 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).33 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 below.


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.34

The very nature of corruption and bribery, namely disguise and opacity, makes their identification difficult, however. Several awards made during 2019 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 2019.35

The Niko Resources case related to 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'.36

In their 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. The tribunal noted also that, had they been upheld, the effect would have been an unjust advantage to the respondent (which had received gas for which it had not paid). In the tribunal's view 'granting such advantage to the alleged victims of corruption cannot be the purpose of the fight against corruption'.37

A second case in which corruption issues were addressed is that of Lao Holdings, for which the award was issued in early August 2019.38 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 their view as to the appropriate standard of proof required to reach a finding of corruption:39

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.

The tribunal in the Glencore/Prodeco case followed similar reasoning, referring to a 'time-honoured methodology to establish truth'.40 On the corruption issue, the tribunal found against the respondent. They made a point, however, of explaining their overall views on corruption, noting not just that it is 'morally odious' but also that it is:41

economically deleterious: it restrains economic development and subdues nations into under-development and poverty, as bribes enriching well-connected civil servants or politicians are financed via inflated prices paid or reductions in income suffered by the poorest citizens. Scarce public funds are misdirected by enriching privileged individuals, at the expense of the common good.

In a very recent decision, in November 2019, the Hague Court of Appeal overturned an International Chamber of Commerce (ICC) award against a subsidiary of Venezuela's state-owned oil and gas company, PDVSA, 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'.42

Recent decisions, therefore, show some consistency in applying a flexible approach to rules of evidence in arriving at a decision where 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.


The MTD award noted above, in relation to contributory fault, made the point that 'the 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 discounted cash flows: the discount rate is often adjusted to reflect country risk;43 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. A recent award in the ConocoPhilips case included some discussion on this point.44

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'.45 The rationale follows in the following paragraph: 'The inclusion of unlawful state action would result in higher discount rates and thus allow Venezuela to benefit from its own unlawful acts'.46

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.47

In concluding, the tribunal clarified that while 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';48 and therefore modified the position put forward by the claimants:49

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 tampered. 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 the solar cases; which concern the legality or otherwise of measures taken by EU governments 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.50 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 '[T]he 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'.51

They also, however, agreed that 'the right to regulate must be subject to limitations if investor protections are not to be rendered meaningless'.52 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'.53

In all of 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;54 as falling 'outside the acceptable range of legislative and regulatory behaviour' in the Novenergia case;55 and as having 'crossed the line from a non-compensable regulatory measure to a compensable breach' in the Foresight case.56

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.57 In Isolux, for example, the tribunal noted their view that '[t]he only legitimate expectation of the claimant was a reasonable return on its investment'.58


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 2019. The recent conclusions of some of the tribunals involved in the solar 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.