Compensation in Energy Arbitration

This chapter focuses on compensation issues that arise in arbitrations in the energy sector.[2] Arbitral tribunals resolving energy disputes typically use the same valuation approaches that are used in disputes relating to other sectors. Compensation in energy arbitration, however, deserves particular consideration. Energy-related investments are often high-value and long-term, subject to price volatility and typically more susceptible to political risk, not least as a result of ‘public-interest concerns’ relating to energy security and calls for resource-nationalism. Each of these factors may have a significant impact on quantum.

The chapter proceeds in four parts. First, we set out two valuation techniques that are typically adopted by tribunals to determine quantum in energy arbitrations – the predominant discounted cash flow (DCF) approach and the market-valuation approach. Second, we discuss the appropriate valuation date for determining damages, which can have a significant impact on quantum in light of fluctuating energy prices. Third, we analyse current approaches to awarding interest, another factor that may have a substantial impact on quantum and yet is considered a ‘vastly under-pleaded area’.[3] Finally, we provide a few concluding thoughts.

Valuation methodologies

Before considering which methodologies are best suited to determining quantum, it is necessary first to consider the appropriate standard of compensation to be applied. In commercial arbitration, this will depend on the governing law and type of breach. Treaties, however, do not always address these issues, particularly in circumstances where there has been an unlawful expropriation or cases involving other treaty violations. In such circumstances, tribunals will often look to principles of compensation in international law, most commonly citing the Chorzów Factory[4] standard of ‘full reparation’. The aim of the full reparation principle (in the absence of restitution) is effectively to determine the measure of quantum that would restore an aggrieved party to that it would likely have enjoyed, but for the wrongful act.

As the tribunal in CMS Gas v. Argentina[5] observed, the methods employed by tribunals to determine compensation depend on the circumstances. Often, however, the aim is to ascertain the fair market value (FMV) of the asset or assets in question. The description of FMV set out in the ASA’s International Glossary of Business Valuation Terms was described by the tribunal in CMS Gas v. Argentina as an ‘internationally recognized definition’:

the price, expressed in terms of cash equivalents, at which property would change hands between a hypothetical willing and able buyer and a hypothetical willing and able seller, acting at arms [sic] length in an open and unrestricted market, when neither is under compulsion to buy or sell and when both have reasonable knowledge of the relevant facts.[6]

The methodology employed in determining an asset’s FMV can have a significant impact on quantum and is often a source of dispute between parties to an investment treaty arbitration. In striving to ascertain the FMV, tribunals most frequently employ the DCF approach and, less frequently, the market-valuation approach. Specific issues arise in relation to each in the context of energy arbitration.

The DCF approach

The DCF approach measures value at a given date on the basis of projected future cash-flows that income-producing assets are expected to generate. These projected income streams are then appropriately discounted to produce a value at the given date that takes account of the risks inherent in such projections.

The DCF methodology is the ‘most widely used and accepted method in the oil and gas industry for valuing sales or acquisitions’.[7] It has been adopted by numerous tribunals, especially in the context of investment treaty arbitration,[8] and is widely accepted by international agencies, such as the World Bank, as a valid methodology for estimating the fair market value of assets in international disputes.[9] A strong recommendation for using DCF is that it is often employed by business people in the sector when making commercial decisions, reflecting that they identify it as a suitable tool for quantifying value in an industry subject to a degree of inherent uncertainty.

However, some tribunals have considered DCF methodologies inappropriate for valuing investments that are at an early stage, where there is insufficient historical data on which to base future projections.[10] Others have taken a different approach. Most recently, the majority in Process and Industrial Developments Ltd v. Federal Republic of Nigeria ordered Nigeria to pay US$6.6 billion in damages for wrongful repudiation of a gas supply and processing agreement.[11] The tribunal applied the DCF method despite the fact that the project had never come to fruition and that the claimant’s sunk costs did not exceed US$40 million.

Projecting future cash flows

A projection of future cash flows requires an assessment of future revenue streams and costs. Projecting future revenues relies on various inputs, each of which can be highly subjective and also have a significant impact on the quantification of future revenues.

First, a tribunal will need to assess likely future production, usually with reference to a volume and production profile (i.e., projecting the size of the reservoir or field and establishing the number of barrels or cubic metres that can be produced each year economically) and possibly incorporating risk adjustment factors to reflect the risk that certain reserve categories may not produce the projected amount of oil.[12] Generally, expert evidence is tendered to substantiate the parties’ assessments but they may also present (or seek disclosure of) internal projections or business plans that reflect contemporaneous assessments of reserves and production estimates.[13] In Occidental Petroleum Corp v. Republic of Ecuador, the tribunal was strongly persuaded by the fact that an assessment made by Petroamazonas (a unit of Ecuador’s state oil company) for business and reporting purposes substantiated the claimants’ expert’s estimates.[14] In Murphy Exploration & Production company – International v. Republic of Ecuador, however, the tribunal declined to rely on the claimant’s projections of oil production levels as these were based on assumptions that the tribunal deemed ‘highly speculative’.[15]

The second crucial input is found in the relevant commodity price projections. Forecasting future crude oil and gas prices is fraught with uncertainty given the volatility in prices, most recently highlighted by the price collapse in late 2014. As noted below, the valuation date may be extremely relevant in terms of price projections as a result. A tribunal will often seek to forecast prices based on market conditions as at the valuation date. One way of achieving this is by relying on actual futures prices as at the valuation date, which should, theoretically, price in the market’s forecast prices.[16] Alternatively, it is possible to rely on industry price forecasts (e.g., price forecasts published by SPEE), or expert evidence.

As well as estimating future revenues, based on a quantity/price assessment, the tribunal will have to consider cash flows out, namely operating expenditure and a portion of capital expenditure.

Discounting future revenues

Finally, the third important element of a DCF analysis is the discount rate. This is intended to reduce the anticipated future cash flows to a present value, accounting for the time value of money and risk. Future cash flows are inherently uncertain, and the discount rate has the effect of ‘neutralising’ that uncertainty.[17] There are different ways that the discount rate may be calculated. The ‘build up’ method essentially aggregates discount rates to reflect the time value of money and multiple risks. Another technique commonly employed by experts is the weighted average cost of capital (WACC) method, which is intended to reflect the relevant company’s cost of debt and equity. This supposedly reflects the risk attached to the company.

In the context of energy arbitration, the discount rate can be very important and may vary significantly depending on the type and characteristics of the project. Often the relevant contracts and operating life of the reservoir or field are long-term so the projected revenues stretch far into the future, amplifying the effect of the selected discount rate. Moreover, energy assets can be subject to a host of risks that may or may not be incorporated into the discount rate (e.g., political risk).

It is vital to properly contextualise the relevant project to identify specific risks associated with its inception, development and implementation. For example, a size premium,[18] or a project risk premium for ‘start-ups’[19] may be appropriate for some companies but not for others.[20]

Some inputs, however, are prevalent in virtually all discount rates, such as future taxes. However, in circumstances where there is a proven risk that the compensation awarded would be subject to tax, it may not be appropriate to include future taxes in the discount rate.[21]

One of the main, and perhaps most controversial, elements in any DCF analysis is country risk. This is the risk associated with making an investment in a particular geography that lacks one or more of the sophisticated legal, financial or constitutional systems evident in more developed economies. It incorporates myriad risks from political, sovereign and macroeconomic to the risks arising from poor or ageing infrastructure, constraints on access to health care or education and even the potential for natural disasters.

Country risk in a cost-of-capital calculation is normally expressed as a percentage premium to be added to the developed country cost of capital. While there is no consensus as to how country risk should be measured or even which features of a country’s political system, economy, business or regulatory environment should be reflected,[22] there are three main sources of country risk measurement: ratings and other agencies; political risk consultancies; and statistical databases that measure country risk based on the prices of publicly traded securities. Country risk can be calculated by examining the interest rate spread of the country’s sovereign bonds compared with ‘risk free’ countries or by reference to equity market deviations. While the latter might appear more appropriate for a foreign equity investor, equity market values for these purposes remain a matter of debate. The alternative is to build up a bespoke risk profile specific to the particular project or investment forming the subject matter of the damages calculation.

Country risk is controversial because it can be hard to calculate, it sometimes includes factors to which the particular investment may not be exposed, and it does not always reflect the economic reality of sophisticated and in particular diversified investors.

Certainly the treatment of country risk within a valuation analysis can significantly impact the quantum of damages ultimately awarded. The divergent outcomes in seven recent cases involving Venezuela are revealing:

In Gold Reserve Inc v. Bolivarian Republic of Venezuela,[23] the tribunal held that Venezuela had breached its fair and equitable treatment (FET) obligation in the Canada–Venezuela bilateral investment treaty (BIT) by terminating the mining company’s two concessions to operate a gold and copper mine in a series of acts and omissions between March 2007 and June 2010. The tribunal ordered Venezuela to pay US$713 million before interest and applied a country risk premium in 2008 of 4 per cent.

In Venezuela Holdings BV and others v. Bolivarian Republic of Venezuela,[24] Venezuela was held liable under the Netherlands–Venezuela BIT for expropriating assets of ExxonMobil subsidiaries in two oil projects and violating the FET obligation by implementing certain production and export curtailment measures. The tribunal awarded the claimants US$1.6 billion plus interest, having applied a global discount rate of 18 per cent. Based on the parties’ submissions, this suggests a country risk premium of approximately 9 per cent. The award has subsequently been partly annulled, although for reasons unrelated to country risk premium.[25]

In Flughafen Zürich AG and Gestión e Ingenería IDC SA v. Bolivarian Republic of Venezuela,[26] Venezuela was found in breach of the Chile–Venezuela and Switzerland–Venezuela BITs. The tribunal unanimously concluded that Venezuela had expropriated the claimants’ investment in the Isla Margarita airport, and held by a majority that Venezuela was also liable for a denial of justice. Venezuela was ordered to pay US$14.8 million as compensation for the expropriation plus interest. The Flughafen tribunal held that Venezuela’s country risk premium was 7.9 per cent during the course of the claimants’ investment in 2004 and 2005.

In OI European Group BV v. Bolivarian Republic of Venezuela,[27] the tribunal found that Venezuela’s undisputed expropriation of OI European Group’s investment in two glass production facilities in Venezuela was unlawful because it lacked due process and there had been an excessive and unjustified delay of over four years in the payment of just compensation. The tribunal also held Venezuela in breach of its FET obligation, arising from the illegality of the expropriation itself and from Venezuela’s occupation of the glass production plants following the expropriation decree. The claimant was awarded US$372.5 million plus interest, the tribunal having assessed Venezuela’s country risk premium in 2010 at 6 per cent.

In Tidewater Investment SRL and Tidewater Caribe CA v. Bolivarian Republic of Venezuela,[28] the tribunal held that Venezuela’s undisputed expropriation of the claimants’ Venezuelan maritime oil support services business was lawful under the Barbados–Venezuela BIT, meeting all the treaty conditions for a permissible expropriation except the payment of prompt, adequate and effective compensation. The tribunal assessed damages as US$46.4 million plus interest having assessed Venezuela’s country risk premium in 2009 at 14.75 per cent.

In Tenaris SA v. Bolivarian Republic of Venezuela,[29] the tribunal awarded US$137 million in damages for Venezuela’s expropriation of the claimant’s investments. The tribunal assessed Venezuela’s country risk premium in 2008 at 4.6 per cent, based on the delta between Venezuelan and US bond yields.[30]

In Saint-Gobain v. Bolivarian Republic of Venezuela,[31] the tribunal found that Venezuela had expropriated the claimant’s investment and failed to pay adequate compensation. The tribunal assessed Venezuela’s country risk premium in 2010 at 10.26 per cent.

These divergent outcomes may seem surprising, particularly given that some of these awards concerned the same expropriatory measures, the same country and sometimes were issued within a short period of each other. In part, this may be due to the specific characteristics of the projects involved, or the arguments and evidence tendered to the different tribunals on the question of country risk premium. As the tribunal in OI observed, the calculation of a discount rate in one case:

may not be mechanically extrapolated as regards another situation – even though one is dealing with a company in a similar sector, and one in the same country . . . in each case, the decision of a tribunal is predetermined by the arguments and counter-arguments made and the proof brought forward by the experts and the parties.[32]

For the most part, however, these divergent outcomes are explained by the decision of several of these tribunals to include the risk of expropriation by the host state in its country risk analysis.

In Gold Reserve – the first reported ICSID case expressly considering whether the country risk premium in a DCF analysis should account for expropriation risk – the tribunal found that it was ‘not appropriate’ to increase the country risk premium to reflect the market’s perception of a heightened risk of expropriation in a given country’s political climate.[33] The tribunal considered that only those political risks ‘other than expropriation’ could be factored into the country risk premium.[34] The tribunal adopted a country risk premium (as of the valuation date of April 2008) of 4 per cent, resulting in a total discount rate of 10.09 per cent.

The Exxon award, which was issued less than three weeks later, reached the opposite conclusion. That tribunal held that the risk of expropriation should be taken into account when calculating the discount rate applicable to the compensation due for an expropriation. It reasoned that, under the standard of compensation it held applicable in that case, compensation must correspond to the amount that a willing buyer would have been ready to pay at a time before the expropriation had occurred. Because ‘it is precisely at the time before an expropriation . . . that the risk of a potential expropriation would exist,’ the tribunal held that the willing buyer would have taken this risk into account when determining the sum he would be willing to pay at that time.[35]

The Flughafen, OI and Tidewater tribunals followed the approach taken in Exxon, rejecting the claimants’ arguments that legal, regulatory and political risk must be excluded from the country risk premium. Those tribunals reasoned that: the claimants were aware of the existing political and legal uncertainties in Venezuela when they invested in the country;[36] an emerging country had a ‘disadvantage’ arising from a hypothetical investor’s preference to invest in a developed country, which the country risk premium was intended to capture;[37] and the country risk premium quantifies the ‘general risks, including political risks, of doing business in a particular country’.[38]

This approach was most recently endorsed in the ConocoPhillips award. The tribunal stated that a proper but-for scenario must reflect the country risk exposure in full, ‘even the possibility of Venezuela adopting measures affecting the Projects including but not limited to expropriatory measures’.[39] Consistent with this approach, the tribunal in Karkey refused to accept the claimant’s view that it was unnecessary to include country risk in the valuation analysis when such a risk was already accounted for in Karkey’s insurance policies and applicable bilateral investment treaties. According to the tribunal, ‘an insurance policy does not eliminate all social, political and legal risks’; otherwise ‘investors could disregard any such risks in a host-State when investing, so long as they are able to obtain insurance’.[40]

In addition to the uncertainty created by these divergent awards, the inclusion of expropriation risk raises real questions as to the appropriateness of country risk premiums in investor–state disputes. Many investors consciously structure their investments with treaty coverage and could reasonably be said to have hedged against risks such as unlawful expropriation through such structuring.

The inclusion of expropriation risk by tribunals ignores the particular circumstances of such an investment. Its inclusion potentially fails to safeguard against the state being rewarded for engaging in conduct that increases the hypothetical buyer’s perception of country risk, even in circumstances in which the state has agreed by treaty either to refrain from or to provide compensation in respect of such action. The tribunal in Flughafen, recently observed that states ought not to be allowed to benefit from an increase in country risk where the increase is caused by the post-investment conduct adopted by the state, although on the facts before the tribunal, it found that had not been the case.[41]

Allowing conduct that is in breach of a treaty to operate as an effective discount on compensation might suggest that no value is attributed to such treaty-protections beyond the creation of a cause of action and access to international arbitration (rather than any local courts). On this view, no protection appears to be afforded to an investor’s ability to actually recover the full value of their investment. On another view, completely excluding expropriation risk from country risk might arguably be viewed as unrealistic, excluding risks that both the market and the investor have priced into the asset in question. Such an approach could be viewed as over-compensating the investor.

The tribunal in Saint-Gobain considered the issue of the appropriate country risk premium at some length. A significant source of the parties’ disagreement turned on ‘whether the risk of being expropriated without the payment of (sufficient) compensation as required by the Treaty should be excluded from the country risk premium applicable to Venezuela.’[42]

The majority took the view that the treaty and recourse to arbitration did not provide insurance against the general risks of investing in Venezuela, which a willing buyer would necessarily take into account.[43] The majority went on to hold that the country risk premium for Venezuela ‘must reflect all political risks associated with investing in Venezuela, including the alleged general risk of being expropriated without payment of (sufficient) compensation.’[44]

The ConocoPhillips tribunal agreed with the general proposition that the risk of expropriation by the host state must be included in its country risk analysis and that an investment treaty does not ‘eliminate’ country risk. However, the tribunal observed that the availability of investment treaty protections can ‘mitigate’ that risk:

An investment treaty does accord some foreign investors additional protection against state measures that other investors even within the same affected industry) might not otherwise have. The ICSID Arbitration initiated by the Claimants attests to that fact. The point is not that an applicable treaty would have eliminated regulatory or country risk, but rather that it would have managed or mitigated it.[45]

Notably, the ConocoPhillips tribunal emphasised that the task of properly assessing the political risk of doing business in a particular state is economic in nature, not legal. According to the tribunal, the lawfulness of the underlying public measure is therefore ‘irrelevant’. On that basis, the tribunal rejected the claimant’s view that including the risk of expropriation into the country risk analysis would amount to ‘rewarding [Venezuela’s] violations of international law [or] creat[ing] an incentive for [it] to take property in violation of its international obligations’[46] and would effectively reward the state for its ‘propensity to engage’ in unlawful conduct as irrelevant to the analysis.[47]

The ConocoPhillips and other recent awards dealing with the question of country risk premium have shed light on a number of important issues pertaining to this issue. The debate, however, continues to evolve, and a significant divergence of opinion on the issue of country risk premium remains. The concurring and dissenting opinion of Judge Charles N Brower in the Saint-Gobain case illustrates the point:

To reduce the recovery to the injured Claimant by applying a ‘fair market value’ that incorporates the very risk of which the Claimant purportedly is being relieved by the Tribunal is to deny the Claimant the full compensation to which it is entitled. It is like undertaking to restore to the owner of a severely damaged automobile a perfectly repaired and restored vehicle but then leaving parts of it missing because it just might be damaged again in the future.[48]

Outside the investor–state arena, country risk remains an appropriate albeit controversial aspect of an overall damages assessment.

An alternative to DCF: the market-valuation approach

Unlike the DCF approach, the market-valuation approach is outward looking insofar as it appeals to the market to discover the price at which willing buyers and willing sellers are prepared to transact for comparable assets, businesses, companies or shares from which a valuation of the particular asset or company can be extrapolated. There are two main types of market-valuation technique: the comparable transactions approach, which looks at arms-length transactions involving comparable assets, and the market capitalisation (for publicly-traded companies) and comparable companies (for private companies) approach, which looks at the value of the company (or comparable companies) on the stock market.

In practice, it is rare to find appropriate comparables in the oil and gas industry because ‘each oil and gas property presents a unique set of value parameters’.[49] They are likely to be of different sizes, host unique geological features, yield different quality of oil, be governed by specific contractual arrangements, be subject to different environmental and remedial obligations, and face different political and legal risks. Moreover, as commodity prices fluctuate, so too will the valuations of assets and companies. The task of identifying and selecting potential assets for comparison and determining whether they constitute appropriate comparators, involves a significant degree of judgment and can be a source of dispute. Tribunals are generally reluctant to adopt the market-valuation approach ‘due to the restricted reliability of sound sets of comparables in the local market’.[50] This approach is often used as an ancillary method of assessing loss[51] to assess or ‘sense-check’ the accuracy of other methods.

Nonetheless, the market valuation approach is increasingly being used as the primary valuation method, including in energy arbitrations.[52] For example, in BG Group Plc v. Republic of Argentina, the tribunal relied on two comparable transactions, taking place either side of the relevant egregious legislation, from which the tribunal could assess the impact of the legislation and awarded the difference between the two comparable transactions: US$185 million.[53] In Ioannis Kardassopoulos and Ron Fuchs v. Georgia, the tribunal was presented with three offers related to the claimant’s shares and held that ‘a completed or seriously contemplated transaction offers the best evidence of an asset’s FMV’.[54] The tribunal adopted the claimant’s approach of weighting the comparable transactions based on the ‘relative maturity of each transaction or offer’.[55]

The tribunal in Crystallex v. Venezuela[56] relied on two market-based approaches that yielded similar outcomes: a stock market approach and a relative market multiple approach. Under the stock market approach, the claimant sought to have the value of its investment assessed by reference to Crystallex’s market capitalisation prior to the impact of the expropriatory measures, adjusted to take account of the performance of other gold companies that had not been affected by the measures. One of the factors that led the Crystallex tribunal to accept this approach was the fact that Crystallex was effectively a one-asset company – its investments in Las Cristinas, an area reported to contain one of the world’s largest undeveloped gold deposits, were its only asset. This, combined with the fact that Crystallex’s shares were actively and heavily traded on two stock exchanges, persuaded the tribunal that the stock market approach was an especially appropriate and reliable valuation method in this case.[57]

The relevant market multiple approach assessed the ratio of the enterprise values to gold reserves of publicly traded gold companies, and then applied this ratio to the amount of Crystallex’s in situ gold reserves at Las Cristinas. The tribunal was satisfied that the claimant and its experts had identified sufficiently comparable companies.[58] The two approaches led to results that were largely consistent with each other, which provided the tribunal with further comfort.[59]

Undoubtedly the most high-profile and significant use of the market valuation approach was in Yukos Universal Limited (Isle of Man) v. Russian Federation (set aside by the Dutch courts on unrelated grounds in 2016). Rather than relying on comparable transactions, the tribunal used the market valuation of comparable companies as ‘the most tenable approach to determine Yukos’ value’.[60] It selected the RTS Oil and Gas index, an index that included nine Russian oil and gas companies. The tribunal then used the movements in the index to adjust Yukos’ value as of 21 November 2007 (the date of the breach) to the award date (the valuation date) and awarded over US$30 billion.[61]

Alternatively, rather than being used as the primary valuation method, the market-valuation approach may prove to be a useful check on a DCF valuation approach. Thus, in Enron Creditors Recovery Corporation v. Argentine Republic, the tribunal referred to the company’s market capitalisation (given that the company’s shares were traded on the Buenos Aires and New York stock exchanges) to find that it was ‘satisfied that the figures resulting from DCF do not show unreasonable differences with those resulting from the verification done in light of the stock market value’.[62]

Other alternatives to DCF

Other valuation methods have also been applied by tribunals in circumstances where DCF or market-valuation approaches were considered inappropriate. For example, in Rusoro v. Venezuela,[63] the tribunal held that the best approach to determining the FMV of expropriated gold mines was a weighted combination of three methodologies. The first was described as the ‘maximum market value’, in effect, the enterprise value of the assets at their peak, prior to the measures taken by Venezuela. Noting that this value was reached only for a very short period, three years prior to the date of expropriation, the tribunal gave it a 25 per cent weighting. The second was the book value of Rusoro’s assets on the last day of the quarter in which the expropriation took place. Noting that this did not reflect the increase in gold prices between the investment and expropriation, nor Rusoro’s development of the mining properties, the tribunal gave this value a 25 per cent weighting. Finally, the tribunal gave a 50 per cent weighting to what was described as the adjusted investment valuation, which was derived by adjusting the original investment consideration for the increase in the price of gold and value of gold producing companies.[64]

It is worth noting that the Rusoro tribunal opted for this complex, three-step analysis after finding that the market-valuation approach was not appropriate because Rusoro’s share price had already been depressed, or ‘contaminated’ before President Chávez announced the expropriation, due to a suite of adverse governmental acts. According to the tribunal, ‘[t]he Final Market Valuation is deeply contaminated by the Bolivarian Republic’s policy decisions affecting the gold sector, adopted after Rusoro had made its investments, and by the increased export restrictions imposed by way of the 2010 Measures’.[65] While the tribunal refused to rely on market value to determine the ‘genuine value’ of Rusoro’s investment, it did use Rusoro’s peak market capitalisation as part of the weighted combination it ultimately employed.

As these examples seek to underscore, tribunals make good use of the various valuation methods to determine an investor’s compensable loss. Ultimately, the choice of valuation method is highly fact dependent and subject to a number of variables. While generalising is perilous, particularly in the area of damages, the method best supported by evidence and one that yields most accurate and reliable results will likely be preferred by the tribunals.[66]

Valuation date

The valuation date, namely the date on which the market value of a claimant’s assets is established for purposes of determining compensation, may have a significant impact on quantum. In energy arbitration in particular, the date on which a party’s investment is valued may be of critical importance given the price volatility that has characterised energy markets in recent years.

As an initial matter, choice of valuation date may not be available to claimants in commercial arbitrations premised on a breach of contract. Instead, ‘[b]y far, the common legal approach’ is to use a valuation date ‘as of the date of harm’.[67]

However, the issue is more complicated in investor–state disputes where the host state is alleged to have expropriated the claimant’s assets. Expropriations may be lawful or unlawful. The appropriate valuation date for a lawful expropriation is ‘well-settled’ and refers to ‘the date of expropriation, or to the date before the impending expropriation became public knowledge, whichever is earlier’.[68]

On the other hand, tribunals in cases of unlawful expropriation (and sometimes other treaty breaches)[69] have referenced the full reparation standard, applicable to remedy illegal acts under international law, to justify departure from this approach. These tribunals have instead valued the claimant’s expropriated assets as of the date of the award.[70] The justification for a date-of-award valuation is often linked to international law’s preference for restitution:

investors must enjoy the benefits of unanticipated events that increase the value of an expropriated asset up to the date of the decision, because they have a right to compensation in lieu of their right to restitution of the expropriated asset as of that date. If the value of the asset increases, this also increases the value of the right to restitution and, accordingly, the right to compensation where restitution is not possible.[71]

This approach was recently adopted by the majority in Quiborax v. Bolivia, which concluded that assessing the value of the investment on the date of the award allowed the tribunal to consider ex post data:

Using actual information is better suited for this purpose than projections based on information available on the date of the expropriation, as it allows to better reflect reality (including market fluctuations) when attempting to ‘re-establish the situation which would, in all probability, have existed if that act had not been committed.’[72]

Similarly, having already found that the appropriate valuation date was the date of the Award, the majority in Burlington Resources v. Ecuador held that ‘if, in the circumstances of the particular case, the use of ex post information is relevant, reasonable and reliable … it should be preferred to ex ante information’.[73]

The tribunal in Murphy v. Ecuador adopted a different approach. It held that in the circumstances, it was not appropriate to assess the value as at the date of the award on the basis of ex-post data. The tribunal found that such an approach would not have reflected ‘what the situation would have been in a but-for scenario’ because, following the sale to Repsol, an entirely new contractual framework had been agreed that substantially changed the financial dynamics of the investment.[74] Similarly, the tribunal in Karkey recently concluded that the ex-post valuation advocated by the claimant was ‘too often speculative and too often based on insufficient evidence’ and applied instead Pakistan’s ex-ante approach, which it found more reliable in the circumstances.[75]

It is therefore important to understand when an expropriation may be characterised as lawful or unlawful, which will often be dictated by the relevant investment treaty. For example, the Energy Charter Treaty (ECT) provides that lawful expropriation entails that the taking is in the public interest, not discriminatory, carried out under due process of law, and ‘accomplished by the payment of prompt, adequate and effective compensation’ and satisfaction of other requirements.[76] Recent cases have concluded that non-payment of compensation is insufficient to render an otherwise lawful expropriation unlawful, where the state has negotiated payment of compensation in good faith.[77]

As noted above, assessing damages as of the date of the award may have a major impact on quantum. For example, the tribunals’ decision to value the claimants’ expropriated assets as of the date of the award in the Yukos arbitrations accounted for a difference of US$44.7 billion. As a result of the significant increase in oil prices over the past decade, the value of the assets had grown from US$21.9 billion on the date of the alleged expropriation in 2004, to US$66.6 billion when the award was issued in 2014.[78]

Of course, claimants should not automatically opt to have their damages assessed as of the date of the award where their investments have been unlawfully expropriated, especially if the value of the relevant assets has declined in the interim. In this regard, the Yukos tribunals suggested that an investor is entitled ‘to select either the date of expropriation or the date of the award’.[79]

It is important to distinguish between the valuation date and the use of information that might post-date the treaty breaches.[80] Even if the chosen valuation date is immediately before the expropriation or treaty breach (as opposed to the date of the award), tribunals have sometimes looked to ex-post information and data in their assessment of damages. This approach arguably allows ‘a more precise assessment of the damages’.[81] In addition, where the market value of the expropriated asset at the time of the award confirms the assumptions underlying a party’s valuation model, the tribunal may look more favourably on that party’s valuation.[82]

An interesting example of the use of ex-ante projections and ex-post data can be found in the recent ConocoPhillips award. This long-standing dispute related to two extra-heavy crude oil projects in Venezuela’s oil-rich Orinoco belt pursuant to certain association agreements between the state-owned Petróleos de Venezuela, SA and two local subsidiaries of US oil major ConocoPhillips. In the context of determining compensation for losses suffered by the subsidiaries as a result of Venezuela’s expropriation of their stakes in the two projects, the tribunal had to decide whether to use ex-ante or ex-post data to perform the valuation. The tribunal declined the claimant’s invitation to use a combination of ex-ante projections (as regards production volumes and costs) and some elements of ex-post data (particularly in relation to oil prices). The tribunal found the ex-post valuation model was more appropriate in the circumstances, not least because most of the evidentiary record, including the expert reports, was based on that approach. The tribunal nonetheless observed that ex-ante data could still be used where ex-post data was ‘either unsubstantiated or unreliable’.[83]

Awards of interest

Awards of pre-award interest (accruing from the date of breach through the date of the award) and post-award interest (accruing from the date of the award through the date of payment) on principal damages are provided for in many national legal regimes, and in international law.[84] Today, it is common practice for international arbitral tribunals to award one or both types of interest, and to set the mode by which interest will be calculated where post-award interest is ordered, although some commentators have bemoaned the lack of a ‘rational and uniform approach for evaluating interest claims’.[85]

Before turning to consider the rules and principles that tribunals have applied to calculate interest payments, it is important to keep in mind the functions that awards of interest are typically believed to fulfil. In essence, payment of interest reflects that the damages owed by the respondent to the claimant are a debt.[86] As such, the claimant is entitled to compensation for the time value of money, preventing the unjust enrichment of the defendant, who improperly derived benefit from the ‘borrowed’ money, and encouraging the timely resolution of disputes by attaching financial consequences to the time elapsing between the defendant’s breach and the resolution of the dispute and payment of damages.[87] Interest may also be viewed as compensating the claimant for the payment risks attached to the respondent.[88]

An arbitral tribunal will look first to the relevant treaty or contract to determine whether interest should be paid, and in what amount. If the treaty or contract is silent, the tribunal may resort to the relevant national law rules regarding payment of pre and post-judgment interest,[89] or general principles of international law.[90] Tribunals in investor–state disputes have moved away from the application of domestic law rules, towards an international law rule of full reparation and compensation for the injured party.[91] Claimants and respondents should survey the contractual provisions, relevant national laws and international law to determine which regime provides the most advantageous rules.

Interest can have a significant impact on quantum, especially where compounded interest at a commercially reasonable rate is ordered.[92] The Tenaris tribunal recently awarded pre-award interest in a sum almost as much as the calculated loss of investment.[93] Having applied an interest rate of 9 per cent compounded semi-annually, the pre-award interest calculated on the loss of US$87.3 million amounted to approximately US$85.5 million. Subject to a six-month grace period, post-award interest was ordered on the same basis.

Accrual of interest

It is generally accepted under international law that interest begins to accrue from the date on which the loss is suffered, usually the date of the breach: ‘Interest runs from the date when the principal sum should have been paid until the date the obligation to pay is fulfilled’.[94] While this may be an easy rule to apply in the case of an outright expropriation, it may be more difficult to discern the date of breach in cases of ‘creeping’ expropriations or, for example, where the respondent has failed to accord fair and equitable treatment to the claimant’s investment.

Compound or simple interest

Awards of compound interest have become the prevailing practice in international arbitration over the course of the last decade.[95] For example, in 2007, the tribunal in Compañiá de Aguas del Aconquija SA and Vivendi Universal SA v. Argentine Republic conducted a ‘non-exhaustive review of investment arbitration cases’ and concluded that seven awards granted compound interest and three granted simple interest: ‘[t]o the extent there has been a tendency of international tribunals to award only simple interest, this is changing, and the award of compound interest is no longer the exception to the rule.’[96] As the tribunal in Koch v. Venezuela observed recently: ‘in modern times, it would be “abnormal” for interest to be limited to simple interest’.[97]

However, an award of compound interest should not be taken for granted. Despite recognising that ‘the awarding of compound interest under international law now represents a form of ‘jurisprudence constante’ in investor–state expropriation cases’,[98] the Yukos tribunal concluded that ‘it would be just and reasonable to award Claimants simple pre-award interest’ in the circumstances of that case, although it did order that post-award interest be compounded.[99] Similarly, the tribunal in Mr Franck Charles Arif v. Republic of Moldova also considered the award of simple interest as ‘more appropriate’.[100] Earlier this year, the tribunal in Olin Holdings Ltd v. Libya declined the claimant’s request for compound interest stating that ‘interests’ have already been accounted for and ‘awarded by the Tribunal for the past losses incurred by the Claimant’.[101] It therefore is important for a claimant to provide reasoned argument in favour of an award of compound interest, highlighting the recognition that compound interest is not punitive, but rather intended to ensure appropriate compensation;[102] and the fact that compound interest may better reflect the economic ‘realit[ies] of financial transactions’ or ‘contemporary financial practice’.[103]

Interest rate

If the governing contract or treaty specifies the appropriate interest rate, the tribunal generally will apply that provision. In the absence of an explicit provision governing the rate of interest (or where provision is simply made for interest to be paid ‘at a commercial rate established on a market basis’),[104] tribunals have typically considered payment at:

  • a reasonable commercial rate, typically reflecting a risk-free rate or benchmark interest rate;
  • the respondent’s unsecured ‘borrowing’ rate;
  • the claimant’s borrowing rate; or
  • the cost of the claimant’s own capital.

Tribunals will most often order interest to be paid at a commercially reasonable rate, although they have diverged in determining how such a rate should be calculated. The majority of tribunals have looked to risk-free rates or benchmark interest rates when awarding interest to claimants. The underlying theory is that as a result of the breach, the claimant no longer bears the risk attached to its investment and it therefore should not be compensated for those risks. Instead, the award of interest simply compensates the claimant for the time value of money for the period of the dispute. Commonly used risk-free rates derive from US or German government bonds, although a tribunal will have to decide which short-term or long-term rate is most appropriate depending on the particularities of the case. In BG Group Plc. v. The Republic of Argentina, the tribunal applied the rates of US Treasury six-month certificates of deposit, compounded semi-annually.[105] Benchmark rates will typically take the form of LIBOR plus a spread. For example, in PSEG Global Inc. and Ors v. Turkey, the tribunal ordered interest to be paid at a rate of LIBOR +2 per cent.[106] The tribunal in Guaracachi and Rurelec v. Bolivia, on the other hand, awarded interest at ‘the annual interest rate reported on the website of the Central Bank of Bolivia for USD commercial loans’.[107]

The claimant’s borrowing rate, namely the ‘average interest rate which Claimant would have paid to borrow’ has been applied by tribunals, on the assumption that ‘it is appropriate and realistic to assume that Claimant would have applied the sums received either to eliminate existing debt or avoid incurring additional debt’.[108]

Awarding interest at the claimant’s cost of capital, or at the rate of return available on investment alternatives, reflects the notion that the claimant would have made productive use of the damages owed to it by the respondent by profitably investing in its own business. This approach has been criticised on the grounds that it is economically unsound, not least because ‘the claimant never undertook the alternative investments nor did it run the associated risks, yet pre-award interest at the claimant’s return would provide compensation on the presumption that the claimant did and that the investments turned out as expected’.[109] Tribunals are divided on the acceptability of an award of interest at the claimant’s cost of capital. While the tribunal in Vivendi v. Argentina accepted that pre-award interest represented a ‘reasonable proxy for the return Claimants could otherwise have earned on the amounts invested and lost’,[110] many tribunals have rejected this argument. For example, the tribunal in Guaracachi and Rurelec v. Bolivia found that the claimant’s weighted average cost of capital did not constitute ‘interest at a normal commercial or legal rate’, and left the claimants overcompensated because ‘the WACC includes an ex ante allowance for forward-looking business risks which should not be applied ex post’.[111]


While every case for damages in the energy sector will turn on its facts, a few lessons of general import can be taken from the discussion above. In particular, there are a number of legal and factual issues that must be considered and properly pleaded to maximise (for the claimant) or minimise (for the defendant) quantum. It will almost always be appropriate to engage at least one expert to support any damages analysis submitted in energy arbitration, with appropriate experience in the relevant sector. The expert or experts will need to work closely with counsel as well as internal business people, who will often be best-positioned to provide evidence and information regarding, for example, the production profile for the specific assets at issue.

The first decision to be made will be selecting the valuation model to support the compensation claim. A DCF model will be appropriate in the majority of situations, but it is worth considering whether a market-valuation approach may provide a viable alternative (or whether it might provide comfort to the tribunal as a check on the reasonableness of a DCF valuation). It is important to spend time thinking about the appropriate evidence to support the cash flow inputs for a DCF analysis, namely production and price, or the appropriateness of the comparables selected for a market-valuation approach. Likewise, serious thought should be given to an appropriate and credible discount rate that properly reflects the risk to which the assets in question are subject, including country risk. Where there is potential in investment treaty arbitration for a claim of illegal expropriation, this may provide an opportunity to value the assets as of the date with the most advantageous price point and projections, whether that is the date of breach or the date of the award. Finally, requests for interest on the principal damages sum should be fully articulated and economically sound, to ensure that the claimant is compensated for the full scope of its loss.


[1] Samantha J Rowe is a partner, Aimee-Jane Lee is international counsel, and Maxim Osadchiy is an associate in the international disputes resolution group at Debevoise & Plimpton LLP. The views expressed in this chapter are those of the authors alone. The authors are grateful for the assistance of Debevoise associate Fiona Poon in the preparation of this chapter.

[2] The term ‘compensation’ is sometimes distinguished from ‘damages’, the former being more closely associated with the nature and kind of remedy applicable in circumstances involving permitted or lawful expropriations, as codified in investment treaties. The term ‘damages’ by contrast is sometimes more closely associated with unlawful conduct more generally and specifically breaches of contract. See Walde, T.W. and Sabahi, B., ‘Compensation, Damages and Valuation in International Investment Law’ in Compensation and Damages in International Investment Arbitration (ed. I. Marboe), OUP Oxford (20 August 2012), p. 2. The primary focus of this chapter concerns the methodologies employed by tribunals in evaluating quantum in energy arbitrations and the terms ‘damages’ and ‘compensation’ are used interchangeably.

[3] Clemmie Spalton, ‘An unexpected interest in interest,’ Global Arbitration Review (12 May 2015).

[4] Factory at Chorzów, Merits, Judgment (13 September 1928), P.C.I.J., Series A, No. 17.

[5] CMS Gas Transmission Company v. Argentine Republic, ICSID Case No. ARB/01/08, Award (12 May 2005), 402.

[6] International Glossary of Business Valuation Terms, American Society of Appraisers, ASA website, 6 June 2001, p. 4 as cited in CMS Gas Transmission Company v. Argentine Republic, ICSID Case No. ARB/01/08, Award (12 May 2005), 402.

[7] Occidental Petroleum Corp v. Republic of Ecuador, ICSID Case No. ARB/06/11, Award (5 October 2012) 779 (Occidental). See also, Abdala, Manuel A., ‘Key Damage Compensation Issues in Oil and Gas International Arbitration Cases’, American University International Law Review 24, no.3 (2009), 548.

[8] Occidental 779; CMS Gas Transmission Company v. Argentine Republic, ICSID Case No. ARB/01/08, Award (12 May 2005); El Paso Energy International Company v. Argentine Republic, ICSID Case No. ARB/03/15, Award (31 October 2011) (El Paso); Enron Creditors Recovery Corporation (formerly Enron Corporation) and Ponderosa Assets, L.P. v. Argentine Republic, ICSID Case No. ARB/01/3 (22 May 2007) (Enron); Hulley Enterprises Limited (Cyprus) v. Russian Federation, PCA Case No. AA 226, Final Award (18 July 2014), Veteran Petroleum Limited (Cyprus) v. Russian Federation, PCA Case No. AA 228, Final Award (18 July 2014); Yukos Universal Limited (Isle of Man) v. Russian Federation, PCA Case No. AA 227, Final Award (18 July 2014) (together, Yukos). See also, Quiborax SA v. Plurinational State of Bolivia, ICSID Case No. ARB/06/02, (16 September 2015), 344. ‘The DCF method is widely accepted as the appropriate method to assess the FMV of going concerns with a proven record of profitability’.

[9] The World Bank Group, Legal Framework for the Treatment of Foreign Investment: Guidelines, 41-42 (1992).

[10] For example, see Windstream Energy LLC v. Government of Canada, PCA Case No. 2013-22, Award (27 September 2016), 475-476, in which the tribunal found that the DCF approach was not an appropriate method for valuing a wind farm that was in early development, instead preferring to rely on the evidence relating to comparable transactions involving early-stage projects; see also Copper Mesa Mining Corporation v. Republic of Ecuador, PCA No. 2012-2, Award (15 March 2016), 7.26, in which the tribunal was tasked with assessing the value of mining concessions that were at an early exploratory stage. In the circumstances, the tribunal found that the most reliable, objective and fair method was to base its assessment of the claimants’ proven expenditure; Tenaris S.A. and Talta – Traiding e Merketing Sociedade Unipessoal Lda. v. Bolivarian Republic of Venezuela, ICSID Case No. /11/26, ARB/11/26 59, Award (29 January 2016) (Tenaris 1) in which, notwithstanding that both parties agreed DCF was the most appropriate valuation methodology [520], the tribunal rejected the DCF methodology, partly on the basis that the history of operations only ran to approximately 42 months and had been characterised by uncertainties [525-527]. The tribunal also rejected the market multiples approach on the basis that it was not satisfied that the companies put forward as most comparable provided reliable guidance [532]. Instead, the tribunal held that the price that had been paid for the interest in the plant almost three and a half years prior to expropriation was an appropriate reflection of the FMV of Talta’s interest in the plant, without any adjustment [566-567]; Khan Resources Inc., Khan Resources B.V., and Cauc Holding Company Ltd. v. The Government of Mongolia, UNCITRAL, Award on the Merits (2 March 2015), 392-393, in which the tribunal declined to apply the DCF method to assess the value of a company holding exploration and mining licences in Mongolia because the DCF approach was ‘unattractive and speculative’ in light of a number of factors and uncertainties, including, inter alia, that it was ‘far from certain’ that the mine would have actually reached production.

[11] Process and Industrial Developments Ltd v. Federal Republic of Nigeria, et al., UNCITRAL, Award (January 2017).

[12] For example, a tribunal may apply different adjustment factors for proved reserves, probable reserves and possible reserves.

[13] See, for example, Antin Infrastructure Services Luxembourg Sàrl and Antin Energia Termosolar BV v. Kingdom of Spain, ICSID Case No. ARB/13/31 Award (27 July 2018), 724, in which the tribunal relied on the claimants’ quantum reports, which it found were ‘well-reasoned and convincing’. Gold Reserve Inc. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB(AF)/09/1, Award (22 September 2014) (Gold Reserve) 830, in which the tribunal accepted the claimant’s experts DCF analysis noting that the DCF method was appropriate because of the commodity nature of the product and the detailed mining cashflow analysis previously performed.

[14] Occidental, 714.

[15] Murphy Exploration & Production company – International v. The Republic of Ecuador, Partial Final Award, UNCITRAL, PCA Case No. AA434 (6 May 2016), 492. The importance of tendering reliable and accurate evidence in support of the proposed damages assessment cannot be emphasised enough. In Koch Minerals Sàrl and Koch Nitrogen International Sàrl v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/11/19 Award (28 November 2017) (Koch), 9.217, the majority of the tribunal ultimately refused to accept either party’s valuation assessments and instead opted for an independent third-party valuation, noting that Venezuela’s valuation was ‘far too low’ and the claimants’ ‘too high’.

[16] Id. 752-758.

[17] Kai F Schumacher and Henner Klonne, ‘Discounted Cash Flow Method ’ in Contemporary and Emerging Issues on the Law of Damages and Valuation in International Investment Arbitration (ed. CL Beharry), Brill Nijhoff (3 April 2018) p. 218.

[18] A size premium is commonly understood as additional returns that might be required for investing in the shares of small companies.

[19] Karkey Karadeniz Elektrik Uretim AS v. Islamic Republic of Pakistan, ICSID Case No. ARB/13/1, Award (22 February 2018) (Karkey), 423.

[20] See, for example, Valores Mundiales, SL and Consorcio Andino SL v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/13/11, Award (25 July 2017) 815. The tribunal refused to apply a size premium to the DCF analysis on the basis that such premium was not justified because the market information tendered by Venezuela’s experts related to US companies and thus was not instructive as to the size of the local companies owned by the claimant that were subject to disputed measures.

[21] Karkey, 846. The tribunal rejected Karkey’s argument that including taxes would create the risk of Karkey having to pay the same amount twice (first as part of the discount rate and then when receiving proceeds of the award). While acknowledging some risk of double-taxation, the tribunal found that Karkey had failed to prove it, let alone demonstrate at what rate the compensation awarded would be taxed. Some tribunals have adopted a more flexible approach to accounting for taxes. See, for example, Valores Mundiales, SL.and Consorcio Andino SL v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/13/11, Award (25 July 2017) (the tribunal declared that the compensation was awarded ‘tax-free’ on the basis that taxes had already been taken into account in valuing the companies).

[22] James Searby, ‘Measuring Country Risk in International Arbitration’, in Contemporary and Emerging Issues on the Law of Damages and Valuation in International Investment Arbitration (ed. CL Beharry), Brill Nijhoff (3 April 2018) p. 260.

[23] Gold Reserve.

[24] Venezuela Holdings, B.V. & Others v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/07/27, Award (9 October 2014) (Exxon).

[25] Id. Decision on Annulment (9 March 2017). In the annulment proceedings, Venezuela argued that the property rights acquired by the investors were circumscribed in a manner that limited the compensation due to the investors in respect of the Cerro Negro Project (the price cap). The ad hoc committee found that the tribunal had not given consideration to the relevance of the price cap to the application of the mandatory criteria set out in the BIT for compensation. In particular, there had been no discussion of what effect the price cap may have had on the market value of the investment or the amount a hypothetical willing buyer might be willing to pay [184]. The award was partly annulled on a number of bases, including that: the tribunal had manifestly exceeded its powers by determining that customary international law regulated the determination and assessment of compensation in place of the provisions of the BIT; the tribunal had failed to state reasons in relation to those parts of the award that purported to be based on the existence of a justiciable obligation to compensate regardless of the provisions of the BIT in respect of compensation; and the tribunal’s dismissal of the relevance of the compensation provisions under the price cap that were said by Venezuela to have formed part of the investors’ investment was unsupported by analysis, based on contradictory reasoning and hence amounted to a failure to state reasons [188].

[26] Flughafen Zürich A.G. and Gestión e Ingenería IDC S.A. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/10/19, Award (18 November 2014) (Flughafen).

[27] OI European Group B.V. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/11/25, Award (10 March 2015) (OI).

[28] Tidewater Investment SRL and Tidewater Caribe, C.A. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/10/5, Award (13 March 2015) (Tidewater).

[29] Tenaris S.A. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/12/23 594, Award (12 December 2016) (Tenaris 2).

[30] Tenaris 2, 461-462.

[31] Saint-Gobain Performance Plastics Europe v. Bolivarian Republic of Venezuela, Decision on Liability and Principles of Quantum (30 December 2016) (Saint-Gobain).

[32] OI, 763.

[33] Gold Reserve, 841.

[34] Id.

[35] Exxon, 365.

[36] Flughafen, 907.

[37] OI, 780.

[38] Tidewater, 184, 186.

[39] Phillips Petroleum Company Venezuela Ltd, Conocophillips Petrozuata BV v. Petroleos De Venezuela, SA, Corpoguanipa, SA, PDVSA Petroleo, SA, ICC Case No. 20549/ASM/JPA (C-20550/ASM) Final Award (24 April 2018) (ConocoPhillips), 1083.

[40] Karkey, 809.

[41] Flughafen, 905-907.

[42] Saint-Gobain, 703.

[43] Saint-Gobain, 719.

[44] Saint-Gobain, 723.

[45] ConocoPhillips, 1077. The tribunal, however, offered little guidance as to the weight such a mitigating factor should assume in the analysis of a country risk premium.

[46] ConocoPhillips, 1083.

[47] ConocoPhillips, 1021.

[48] Saint-Gobain, Concurring and Dissenting Opinion of Judge Charles N. Brower (21 December 2016), 3.

[49] Occidental, 781, 787.

[50] El Paso, 711.

[51] PriceWaterhouseCoopers, International Arbitration damages research update 6 (2017),

[52] Id.

[53] BG Group Plc v. Republic of Argentina, UNCITRAL Award (24 December 2007) (BG) 444.

[54] Ioannis Kardassopoulos and Ron Fuchs v. Georgia, ICSID Case No. ARB/05/18 and No. ARB/07/15, Award (3 March 2010), 595.

[55] Id. 600-603.

[56] Crystallex International Corporation v. Bolivarian Republic of Venezuela, ICSID Case No. ARB(AF)/11/2 Award (4 April 2016) (Crystallex).

[57] Crystallex, 889-890.

[58] Crystallex, 901.

[59] Crystallex, 901.

[60] Yukos, 1787.

[61] The tribunal adjusted the value to both the expropriation date and the award date. The tribunal had earlier identified that the claimants were entitled to select either the expropriation date or the award date as the valuation date.

[62] Enron, 428.

[63] Rusoro Mining Limited v. The Bolivian Republic of Venezuela, ICSID Case No. ARB(AF)/12/5 Award (22 August 2016) (Rusoro).

[64] Rusoro, 787-790.

[65] Rusoro, 779.

[66] See, for example, Windstream Energy LLC v. Government of Canada, Award (27 September 2016), PCA Case No. 2013-22, 476, in which the tribunal preferred a comparable transactions approach over other valuation methods because ‘the evidence relating to comparable transactions [was] the best evidence before it.’

[67] Kantor, Mark, Valuation for Arbitration, Kluwer Law International (2008), p. 61.

[68] Ripinsky, Sergey and Williams, Kevin, Damages in International Law, British Institute of International and Commercial Law, pp. 243-244.

[69] Murphy, 425.

[70] See generally, e.g., ADC Affiliate Limited and ADC & ADMC Management Limited v. Republic of Hungary, ICSID Case No. ARB/03/16, Award (2 October 2006), Siemens A.G. v. The Argentine Republic, ICSID Case No. ARB/02/8, Award (17 January 2007), Yukos.

[71] Yukos, 1767. See also Draft Articles on Responsibility of States for Internationally Wrongful Acts, adopted by the International Law Commission at its fifty-third session (2001), (ILC Draft Articles) Articles 35–36.

[72] Quiborax SA v. Plurinational State of Bolivia, ICSID Case No. ARB/06/02, (16 September 2015) at 379, citing Factory at Chorzów, Merits, 1928, P.C.I.J., Series A, No. 17, p. 47.

[73] Burlington Resources v. Ecuador, ICSID Case No. ARB/08/05 Decision on Reconsideration and Award (7 February 2017).

[74] Murphy v. Ecuador, 483-484. Following the introduction of the measure that ultimately led to the sale of Murphy Ecuador to Repsol YPF, Repsol negotiated a new contractual framework that incentivised investment (which Repsol did, leading to increased production). By contrast, under the old contractual framework in place with Murphy, the incentives were to wind down investment. In the circumstances, the tribunal employed an FMV methodology with a valuation date coinciding with the sale, rather than with the date of the award [486]. The tribunal did, however, make use of information that became known after the valuation date in determining whether the assumptions on the development of oil prices and production levels should be adjusted and concluded that it was ‘neither necessary nor justified to adjust the ex-ante calculation of damages’ [487].

[75] Karkey, 670.

[76] ECT, Article 13(1).

[77] ConocoPhillips Petrozuata B.V., ConocoPhillips Hamaca B.V. and ConocoPhillips Gulf of Paria B.V. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/07/30, Decision on Jurisdiction and Merits (3 September 2013), 401 (‘The Tribunal accordingly concludes that the Respondent breached its obligation to negotiate in good faith for compensation for its taking of the ConocoPhillips assets in the three projects on the basis of market value as required by Article 6(c) of the BIT, and that the date of the valuation is the date of the Award.’); Mobil Corp. et al v. Venezuela, ICSID Case No. ARB/07/27, Award (9 October 2014), 301, 305–306 (‘[T]he mere fact that an investor has not received compensation does not in itself render an expropriation unlawful. An offer of compensation may have been made to the investor and, in such a case, the legality of the expropriation will depend on the terms of that offer . . . [T]he evidence submitted does not demonstrate that the proposals made by Venezuela were incompatible with the requirement of “just” compensation of Article 6(c) of the BIT. Accordingly, the Claimants have not established the unlawfulness of the expropriation on that ground.’).

[78] Yukos, 1826.

[79] Id. 1763.

[80] See Steven Ratner, ‘Compensation for Expropriations in a World of Investment Treaties: Beyond the Lawful/Unlawful Distinction’ (2017) 111 American Journal of International Law 1, 13 (noting that tribunals striving to assess the full value of investment on the date of expropriation have not been consistent in whether to only use information available at the date of the expropriation or information that has become available since that time, in particular when calculating the DCF of the investment).

[81] Flemingo DutyFree Shop Private Ltd v the Republic of Poland, UNCITRAL Award (12 August 2016), 899. See also Amco Asia Corporation and others v. Republic of Indonesia, ICSID Case No. ARB/81/1 Award (20 November 1984), 613-14, in which the tribunal assessed the value expropriated property as of the date of expropriation using ex-post information.

[82] For example, in Amco Int’l Finance Corp. v. Iran, the tribunal referred to actual post-valuation oil prices to check the reasonableness of forecasts adopted by experts. Amco Int’l Finance Corp. v. Iran, Iran Award 310-56-3, 15 Iran-U.S.C.T.R. 189, (14 July 1987), 181, 237. In Occidental the tribunal availed itself of post-valuation data to check the parties’ assumptions. Occidental, 726.

[83] ConocoPhillips, 579.

[84] ILC Draft Articles on State Responsibility, Article 38(1). (‘Interest on any principal sum . . . shall be payable when necessary in order to ensure full reparation. The interest and mode of calculation shall be set so as to achieve that result.’).

[85] Gotanda, John Y. ‘Awarding Interest in International Arbitration’ (1996) 90 AJIL 40, 40; Walde, T.W. and Sabahi, B., ‘Compensation, Damages and Valuation in International Investment Law’ in Compensation and Damages in International Investment Arbitration (ed. I. Marboe), p. 46.

[86] Harris, D., Caldwell, R. and Maniatis, M. Alexis, ‘A Subject of Interest: Pre-award Interest Rates in International Arbitration’ (1 June 2015),

[87] Gotanda, John Y., ‘Compound Interest in International Disputes’ (2004) Oxford U. Comparative L. Forum at

[88] Harris, D., Caldwell, R. and Maniatis, M. Alexis, ‘A Subject of Interest: Pre-award Interest Rates in International Arbitration’ (1 June 2015),

[89] Gotanda, John Y., ‘Awarding Interest in International Arbitration’ (1996) 90 AJIL 40, 51 (describing the process of ascertaining a national law to be applied to an interest claim as ‘particularly complex’).

[90] Id. at 53.

[91] Compañía de Desarollo de Santa Elena, S.A. v. The Republic of Costa Rica, ICSID ARB/96/1, Final Award (17 February 2000) (Santa Elena); Wena Hotels v. Egypt, 41 I.L.M., pp. 933, 945 (2002).

[92] For example, the Occidental tribunal awarded the claimant pre-award interest on principal damages of US$1.769625 billion at the rate of 4.188 per cent per annum, compounded annually over six years, and post-award interest at the US six-month LIBOR rate, compounded monthly, accruing from the date of the award in September 2012. Ecuador still has not paid the award while an annulment application is pending. Occidental, 876.

[93] Tenaris 1, 625.

[94] ILC Draft Articles on State Responsibility, Art. 38. See also Ioan Micula and others v. Romania, ICSID Case No. ARB/05/20, Award (11 December 2013), 1273 (‘Interest must be calculated from the date on which the loss was suffered. This is usually the day on which the breach occurs.’).

[95] This was not always the case; in 2007, two commentators noted that ‘[t]he norm, until recently, has been to award simple interest.’ See Walde, T.W. and Sabahi, B., ‘Compensation, Damages and Valuation in International Investment Law’ in Compensation and Damages in International Investment Arbitration (ed. I. Marboe), OUP Oxford, p. 45.

[96] Compañiá de Aguas del Aconquija SA and Vivendi Universal SA v. Argentine Republic, ICSID Case No. ARB/97/3, Award (20 August 2007) (Vivendi), 9.2.4. See also Walde, TW and Sabahi, B, ‘Compensation, Damages and Valuation in International Investment Law’ in Compensation and Damages in International Investment Arbitration (ed. I. Marboe), OUP Oxford, p. 45.

[97] Koch, 11.10. See aslo PriceWaterhouseCoopers, International Arbitration damages research update 7 (2017), (noting that of 21 new cases surveyed for the period from January 2015 to January 2017, compound interest was awarded in 19 cases).

[98] Yukos, 1689.

[99] Id.

[100] Mr. Franck Charles Arif v. Republic of Moldova, ICSID Case No. ARB/11/23, Award (8 April 2013), 619.

[101] Olin Holdings Ltd v. Libya, ICC Case No. 20355/MCP Final Award (25 May 2018), 534.

[102] Santa Elena, 104.

[103] Azurix Corp. v. The Argentine Republic, ICSID Case No. ARB/01/12, Award (14 July 2006), 440; LG&E Energy Corp. & Ors v. Argentina, ICSID Case No. ARB/02/1, Award (25 July 2007), 103.

[104] Id., Art. 13(1).

[105] BG, 455.

[106] PSEG Global Inc. and Konya Ilgin Elektrik Üretim ve Ticaret Limited Sirketi v. Turkey, ICSID Case No. ARB/02/5, Award (19 January 2007), 348 (PSEG).

[107] Guaracachi America, Inc. and Rurelec PLC v. The Plurinational State of Bolivia, UNCITRAL, PCA Case No. 2011-17, Award (31 January 2014), 615 (Guaracachi).

[108] National Grid P.L.C. v. The Argentine Republic, UNCITRAL, Award (3 November 2008), 294.

[109] Harris, D., Caldwell, R. and Maniatis, M. Alexis, ‘A Subject of Interest: Pre-award Interest Rates in International Arbitration’ (1 June 2015),; see also Abdala, Manuel A., Lopez Zadicoff, Pablo D., Spiller, Pablo T., ‘Invalid Round Trips in Setting Pre-Judgment Interest in International Arbitration’, World Arbitration and Mediation Review, 2011 Vol. 5 No. 1, p.11.

[110] Vivendi, 9.2.8. Note, though that the tribunal halved the ‘anticipated 11.7 per cent rate of return on investment’ sought by the claimant to a 6 per cent interest rate. Id.

[111] Guaracachi, 609, 615. See also PSEG 348 (rejecting the claimant’s cost of equity, Turkish bond yields and US Treasury bills in favour of LIBOR +2 per cent, compounded semi-annually); Enron, 451–52 (rejecting use of the claimant’s WACC and applying instead an interest rate of 6-month LIBOR +2 per cent).

Get unlimited access to all Global Arbitration Review content