Despite the variety of investment treaty disputes involving assets in the Post-Soviet jurisdictions in Central Asia, assessment of damages in each particular case is often heavily debated by the parties, experts and tribunals. In many instances, selecting an appropriate valuation method is the cornerstone of the tribunal’s decision-making on damages. This article provides an overview of some publicly available investment treaty awards in relation to assets in Kazakhstan, Uzbekistan, Kyrgyzstan and Tajikistan, and analyses the key considerations of tribunals in relation to damages.


An Analysis of Tribunals’ Considerations in Damages Assessment


  1. The Discounted Cash Flows (DCF) Method

DCF valuation is based on the concept that value can be assessed by reference to expected future cash flows.

In performing a DCF valuation, it is necessary to consider the expected financial performance of the subject asset. In doing so, valuers may have regard to projections prepared by management or other stakeholders (such as banks, investors, equity analysts) as at the valuation date, and also consider the historical performance of the asset (although the latter is not necessarily relevant in estimating the future performance).

A survey of publicly available awards rendered for investment disputes in Central Asia demonstrates that the majority of tribunals treat this method with caution, principally because of the uncertainties they face in its application:

  • In AIG Capital Partners et al. v Kazakhstan (2003) [¶¶ 12.1.9-12.1.10], the tribunal referred to the speculative nature of DCF analysis due to the asset not being a going concern and the absence of a track record or advance customer orders to assess future cash flows reliably;
  • In Al-Bahloul v Tajikistan (2010) [¶¶ 71-73] and Caratube International Oil Company et al. v Kazakhstan et al. (2017) [¶¶ 1087, 1098, 1101, 1106-1107, 1119, 1131, 1151], the tribunals raised the issues of going concern, proven track record and “sufficient certainty” of profitability in commenting on the DCF method; and
  • Also in Al-Bahloul v Tajikistan (2010) [¶ 96], the tribunal questioned the applicability of the DCF method if the financing of the asset was uncertain.

Notwithstanding the above, some tribunals have determined that the DCF method is an appropriate approach to valuation, even when the subject business has no track record or limited data is available as to its historical and/or expected financial performance:

  • The tribunal in Al-Bahloul v Tajikistan (2010) [¶ 75] acknowledged that determination of future cash flows from hydrocarbon exploration projects “need not depend on a past record of profitability…and sufficient data allowing future cash flow projections should be available” and a minority arbitrator in Caratube International Oil Company et al. v Kazakhstan et al. (2017) [¶ 1089] stated “that the valuation of a concession or a contract for the exploration of an oil field is calculated by reference to the reserves (and not to the actual profit)”;
  • In Sistem Muhendislik v Kyrgyzstan (2009) [¶ 164], the tribunal stated that DCF may be applied even if available data is scarce, if both parties considered it to be sufficient for the DCF approach;
  • In Rumeli Telekom et al. v Kazakhstan (2008) [¶ 810], the tribunal found that “DCF valuation would likely have formed one of the measures which would have informed a discussion between a willing seller and a willing buyer”, but nevertheless the DCF method “must be understood as an approximation which is dependent on the validity of the assumptions, and not as a mechanical calculation which yields a value whose validity is not open to question”.


  1. The Market Approach / Comparable Multiples Method

Whilst tribunals acknowledge the applicability of the market approach (sometimes referred to as the comparable multiples method), and in some instances even prefer this method over the DCF method (e.g., Stati et al. v Kazakhstan (2013) [¶ 1625]), the market approach has its limitations. It provides an indication of value by comparing the subject asset with identical or similar assets for which price information is available, in particular by considering the latter’s prices as a multiple of their financial and/or operating metrics. Commonly used multiples include enterprise value to EBITDA and price to earnings.

The key shortcoming of the market approach, as identified by tribunals in the past, is that it is limited by the degree of comparability between the subject asset and the benchmark / comparable assets (or between the economic characteristics of the subject asset over time).

Nonetheless, tribunals have concluded that despite (sometimes) limited comparability, circumstances might exist where there are no alternative ways to value the subject asset:

  • In Belokon v Kyrgyzstan (2014) [¶ 312], the tribunal noted that the comparable multiples method to valuing a Kyrgyz bank was “in some respects mechanistic, and unlikely to be wholly consonant” and “the circumstances are not free from difficulty or doubt”, but the “final numbers reflect what the arbitrators believe to be prudent approximations derived from the best available information to them”. Due to the absence of comparable quoted banks in Kyrgyzstan or Central Asia generally, the multiple ultimately applied to value the bank in question was based on the prices of similar banks in Central and Eastern Europe, which had different economic characteristics (not least due to the different country risk levels);
  • In Caratube International Oil Company et al. v Kazakhstan et al. (2017) [¶¶ 1133-1135], the tribunal rejected multiples derived from comparable asset sales and purchase offers for a variety of reasons, including differences in asset location, stage of development, size and arm’s length nature of the bids and actual transactions;
  • In Sistem Muhendislik v Kyrgyzstan (2009) [¶ 162], the tribunal found no adequate basis for application of the comparable multiples approach as the majority of comparables were in more developed markets than Kyrgyzstan (e.g., UK, US, Sweden). The tribunal considered that proposed application of a 30% discount to those multiples to adjust for the conditions of the Kyrgyz market “involve[d] a large measure of speculation”.


  1. Role of Past Transactions in the Subject Asset

The awards analysed suggest that tribunals place significant weight on transactions in the subject asset itself (and even in non-binding or indicative bids that do not result in transactions), and any value indications derived from these. In some instances, tribunals relied on transactions despite them having taken place when the asset’s performance and/or general economic conditions were materially different to those at the damages assessment date. I note that using past transactions to value the subject asset is a variation of the comparable multiples method discussed above.

For example:

  • In Rumeli Telekom et al. v Kazakhstan (2008) [¶¶ 813-817], the tribunal awarded just above 50% of the claimed DCF value without any detailed calculation of the awarded amount, but with reference to transactions in and offers for the subject telecom business that: (a) occurred after the valuation date (i.e., the benefit of hindsight was used); (b) were rejected (in particular, the tribunal states it “does not regard them as relevant to the market value of the shares” at the valuation date); (c) were non-binding; and (d) were made at the “time the very rapid market growth in the market…had not become established”;
  • In Stati et al. v Kazakhstan (2013) [¶¶ 1746-1748], instead of relying upon valuation methods adopted by the parties’ experts, including DCF, comparable multiples and wasted costs, the tribunal considered “the relatively best source for the valuation…accepted by the Tribunal are the contemporaneous bids that were made for the LPG Plant [i.e., one of the subject assets] by third parties after Claimants’ efforts to sell the LPG Plant both before and after” the valuation date. The tribunal considered actual bids for the asset and ultimately awarded the amount offered by a state-owned entity. In other words, the tribunal relied purely on factual inputs (despite those appearing to be non-binding) as opposed to valuation expert opinions.


  1. Third-party Valuations of the Subject Asset

Another reference point that tribunals appear to consider are contemporaneous third-party valuations performed outside of the dispute context.  For example, in Stati et al. v Kazakhstan (2013) [¶ 1643], the tribunal refers in its award to a third-party valuation performed by a bank for a state-owned entity independently of the dispute and which corroborates the claimant’s valuation.


  1. Sunk or Wasted Costs

Notwithstanding the pros and cons of the valuation methods discussed above, the so called “sunk or wasted costs” method appears to have a material impact on tribunals’ awards in investment arbitrations. This method follows the replacement cost approach, which follows the economic principle that a buyer will pay no more for an asset than the cost to obtain an asset of equal utility, whether by purchase or by construction, unless undue time, inconvenience, risk or other factors are involved. In practice, tribunals appear to treat wasted costs as a reference point for replacement cost. This consideration is particularly prominent where tribunals are concerned with the speculative nature of other valuation methods, inputs or modelling assumptions.

For example:

  • In AIG Capital Partners et al. v Kazakhstan (2003) [¶¶ 12.1.9, 15], the tribunal criticised the DCF method as applied by the claimant because the fair market value of the investment on the DCF basis was more than 4 times the amount invested at the relevant date. On that basis, the amount invested up to the valuation date was awarded;
  • In Caratube International Oil Company et al. v Kazakhstan et al. (2017) [¶¶ 1087, 1151, 1161, 1164, 1166], the tribunal found that the value of lost future profits did not provide a basis for damages that was sufficiently certain and, therefore, “sunk investment costs best express in monetary terms the damages incurred…as a result of the unlawful expropriation”;
  • In Stati et al. v Kazakhstan (2013) [¶¶ 1687-1688], the tribunal accepted the amount invested in an oil field exploration project as damages, but stated that damages claimed for lost profit / opportunity “provide a much higher threshold for Claimants’ burden of proof…both legally and factually”, referring to a requirement for “track record of profitability rooted in a perennial history of operations, or…binding contractual revenue obligations in place that establish the expectation of profit at a certain level over a given number of years”.

Nonetheless, some awards recognise that cost is not necessarily representative of value, even if easier to establish. For example, the tribunal in Sistem Muhendislik v Kyrgyzstan (2009) [¶ 161] found that, in the context of expropriation, “replacement cost is less helpful than a valuation based upon expected profits…in contrast, because buyers of businesses can be expected to value them according to the profit that they will generate, rather than the cost of creating them, the “multiple deals”…and the DCF method, appears more appropriate”.


Comments and Concluding Remarks

Based on the above analysis, it appears that tribunals in Central Asian investment arbitrations have concerns with the DCF and market approaches due to questions of reliability of the assumptions underpinning the former and comparability of benchmarks in the latter. Tribunals viewed transactions in or bids for the subject asset and wasted costs as valuation reference points or, at least, as helpful cross-checks.

In relying upon historical transactions or bids, it is necessary to consider (a) the similarity in the economic characteristics of the asset at the relevant times (e.g., an asset in the development stage may not be comparable to the asset when it is more mature); (b) changes in external market conditions over time (e.g., interest rates, regulatory regime); and (c) whether the transactions were conducted at arm’s length and, in the case of bids, whether or not they were binding.

As to wasted costs, despite the above raised criticisms of the DCF and market approaches, they are generally accepted business valuation approaches and, as noted in Rumeli Telekom et al. v Kazakhstan (2008) [¶ 810], one would expect these approaches to be considered by a willing buyer and a willing seller in reaching an arm’s length transaction price. This is particularly relevant given that the cost (a historic measure) to build an asset may be substantially different to the asset’s fair market value, which in many instances reflects investor’s expectations as to future benefits that could be derived from that asset.


The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions, position, or policy of Berkeley Research Group, LLC or its other employees and affiliates.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here.

Kluwer Arbitration

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.