The Investment Treaty Arbitration Review: Choosing the Appropriate Valuation Approach for Damages Assessment

I Introduction

Investment treaty arbitration is often driven by a claim for financial compensation – specifically, claims for the value of an investment that has been lost or diminished. The claims, particularly in expropriation cases, typically focus on the value of an asset; for example, an operating company. Furthermore, the value of the asset needs to be estimated when there is no obvious or reliable market data; for example, market prices for traded equity shares. Tribunals (assisted by damages experts) are tasked with assessing the value of these assets, which can be a complicated and technical exercise. The first step in the assessment of value is choosing the appropriate approach or methodology.

This chapter focuses on choosing the appropriate approach for valuing a specific asset at a given valuation date, as is often relevant for an expropriation case or other treaty violations. Expropriation claims may require the valuation of an entire asset prior to the expropriation. Other treaty claims may require the comparison of two valuations in order to assess the loss in value due to the allegations. In either situation, choosing the appropriate valuation approach is the necessary first step in the assessment of damages. This chapter is based on accounting and economic expertise and should not be considered to offer any legal expertise or opinion.

II Standard of value

The standard of value is a legal matter and typically an instruction to experts. The instruction may be directed by the language in the investment treaty or legal arguments. The standard of value, especially in expropriation cases, is often fair market value,2 which is defined in the Guidelines on the Treatment of Foreign Direct Investment from the World Bank as:

an amount that a willing buyer would normally pay to a willing seller after taking into account the nature of the investment, the circumstances in which it would operate in the future and its specific characteristics, including the period in which it has been in existence, the proportion of tangible assets in the total investment and other relevant factors pertinent to the specific circumstances of each use.3

Similarly, Black's Law Dictionary defines fair market value as 'the price that a seller is willing to accept and a buyer is willing to pay on the open market in an arm's length transaction; the point at which supply and demand intersect'.4

The important elements of the various definitions of fair market value are:

  1. the price in a hypothetical arm's-length transaction;
  2. willing and able hypothetical buyer and seller; and
  3. free and open market.

The idea of fair market value is that it should be a figure that equates to a price that would be acceptable to both buyer and seller in a hypothetical market transaction. Fair market value should, as much as possible, approximate a price in a transaction in a non-contentious situation.

Furthermore, the definition of fair market value typically assumes the buyer and seller are hypothetical, meaning there are no specific characteristics or circumstances that should be attributed to either the buyer or the seller. For example, one should not assume that the buyer is a specific company that can benefit from specific synergies5 with existing assets or operations.

Similarly, one must assume that the seller of the asset is under no compulsion to sell and would only accept an offer that provides at least as much value as continuing to hold the asset.

Further, a concept known as 'highest and best use' of the asset is often also included in fair market value.6 This means that the asset should be valued assuming that it will be operated efficiently and for the highest profit that can reasonably be achieved. These elements of fair market value and highest and best use, taken together, should approximate the price in a negotiated transaction in an open market where it is reasonable to assume that the highest bidder determines the price.

III Approaches to valuation

There are four general approaches to valuing an asset:

  1. discounted cash flows (DCF, sometimes referred to as an income approach);
  2. relative valuation (also referred to as market-based valuation);
  3. cost or replacement value; and
  4. option pricing.

The first three approaches are commonly used in valuations in arbitration. We discuss each of these approaches at a high level.7 Option pricing, while common for some specific assets such as derivatives, is not often seen in the valuation of private companies and other assets that are typically the subject of investment treaty arbitrations, and so is not discussed here.

i Discounted cash flows

The DCF approach is the fundamental basis of valuation and forms the foundation for other approaches, such as relative valuation. DCF is based on the principle that 'the value of any asset is the present value of expected future cash flows on it'.8 In other words, a DCF approach seeks to measure the fundamental value specific to the asset or company in question, by measuring the expected cash flows to be generated by the asset and discounting the cash flows by a discount rate related specifically to the risk of the asset.

There are two major components of a DCF analysis: (1) projected free cash flow (estimates of the cash available to debt and equity holders after all cash expenses of the business have been met); and (2) the discount rate (a measure of the risk of the projected cash flows, with higher rates for riskier assets and lower rates for less risky assets).9 When valuing a firm, the discount rate is typically measured as the weighted average cost of capital (i.e., the average return required by both debt and equity investors).

There are a range of DCF models that can be used to value the entire assets of a company or the equity in a company. However, the principal elements are the same in each case.

ii Relative valuation

Relative valuation (sometimes referred to as a market-based approach) involves comparison with similar assets with known market values. 'In a relative valuation, the value of an asset is derived from the pricing of comparable assets, standardized using a common variable such as earnings, cash flows, book value, or revenues.'10

Effectively, a relative valuation is performed on the basis that the market has correctly valued similar assets and the market prices of those similar assets constitute a suitable proxy for the value of the asset in question, scaled for earnings, revenues, etc. The approach is not a fundamental analysis of the intrinsic value of an asset.

The two most common types of relative valuations use (1) comparable company values and (2) comparable transactions. The first approach relies on identifying publicly traded companies that are similar to the asset being valued. It uses market data and financial metrics of the comparable companies to determine valuation multiples such as price to earnings (P/E) or enterprise value (EV) to earnings before interest, taxes, depreciation and amortisation EBITDA (EV/EBITDA), and then applies those multiples to the earnings or EBITDA of the company being valued. The comparable transactions method takes a similar approach but relies on transactions between parties as the basis of value (e.g., acquisitions of entire companies) instead of share prices from equity markets.

In reality, the relative valuation approach is commonly relied on in transactions but the shortcomings of this approach become more pronounced in the context of arbitrations because there is often a scarcity of sufficiently comparable assets or transactions.

iii Cost

The cost approach, sometimes referred to as investment cost or book value, is essentially a measure of how much has been invested in a project or an asset. Alternatively, the approach can look at how much it would cost to replicate or replace the asset. This is typically useful for fixed assets but less helpful when assessing the fair market value of an operating company, unless the time frame between investment and the valuation date is short, with no significant changes in microeconomic or macroeconomic factors or other developments that would affect the asset.

IV Choosing a valuation approach

Choosing which valuation approach to use depends on the characteristics of the asset and the information available. We explain below the factors that must be assessed when considering each approach. We also provide some examples of how each approach has been viewed in arbitration awards.

i Suitability of the DCF approach

From an economic perspective, assessing the fundamental or intrinsic value of an asset using a DCF approach is almost always preferred if possible. There are several characteristics of a DCF analysis that are useful, particularly in a contentious setting.

A DCF approach is specific to the asset in question, allowing for consideration of the operations and projected financial performance of the firm being valued. Understanding and testing the cash flow projections makes it possible to determine whether the expectations of the company are reasonable.

A DCF analysis is transparent, providing a framework that is easily used by any party. The analysis can be adjusted to see the sensitivity to specific assumptions, allowing the users to focus on those assumptions that have a substantial effect on the valuation and those assumptions that may be in dispute. That transparency allows a tribunal to understand the fundamental difference between experts' opinions.

A DCF approach is well suited to understanding the effect on a valuation of allegations, where the asset has not been expropriated or destroyed entirely or where there are multiple allegations that may need to be separated.

Although it may be a preferred approach on principle, from a practical perspective a DCF approach requires reliable estimates of expected future cash flows and discount rates. If these two main inputs are too difficult to estimate, a DCF approach cannot produce a reliable or robust result. The DCF approach 'is easiest to use for assets (firms) whose cash flows are currently positive and can be estimated with some reliability for future periods, and where a proxy for risk that can be used to obtain discount rates is available'.11

This is echoed in the World Bank Guidelines, which state that a DCF approach can be deemed reasonable 'for a going concern with a proven record of profitability'.12 A proven record of profitability makes the estimation of future cash flows easier and less speculative.

Tribunals have agreed and have accepted the DCF approach as an appropriate basis for valuation. The tribunal in the CMS v. Argentina case stated:

This leaves the Tribunal with the DCF method and it has no hesitation in endorsing it as the one which is the most appropriate in this case. [The asset] was and is a going concern; DCF techniques have been universally adopted, including by numerous arbitral tribunal, as an appropriate method for valuing business assets; as a matter of fact, it was used by ENARGAS in its 1996/97 tariff review. Finally, there is adequate data to make a rational DCF valuation of [the asset].13

For early-stage companies, this proven track record may not be in place. Early-stage companies may not have proven technology or operations, may not have an accessible market and may not have a full picture of the operational risks. In this situation it can be difficult to estimate the future cash flows with any degree of certainty, meaning the resulting valuation could be considered speculative.

There may be an exception to the early-stage rule – companies or projects at an early stage of operation but where the valuation inputs are well understood and reliable. For early-stage resource projects, particularly in the energy or mining sectors, a DCF analysis is often the approach used in the real world. This is because technical information is often available about the reserves in place, and the required infrastructure costs and the markets for such commodities are well understood. For example, oil has a ready and liquid market that can provide pricing data.

The World Bank Guidelines discuss the use of DCF for a 'going concern' on the basis of generating reliable data:

For a going concern, i.e. and enterprise consisting of income-producing assets and already in existence for a sufficient period of time to generate the data necessary for proving its profitability and the calculation, with reasonable certainty, of its income in future years . . . discounted cash flow may represent an acceptable method of valuation.14

This suggests that the requirement for a track record of profitability is driven by the need, at least partially, for data so as to produce reliable projections. In resource projects, some of that data may be available in other forms, not simply from historical performance of the specific project.

This is demonstrated in the Gold Reserve v. Venezuela award, in which damages were determined based on a DCF model despite the gold mine having never operated:

Although the [project] was never a functioning mine and therefore did not have a history of cashflow which would lend itself to the DCF model, the Tribunal accepts the explanation of both [experts] that a DCF method can be reliably used in the instant case because of the commodity nature of the product and detailed mining cashflow analysis previously performed.15

Even in this situation, the DCF must still be considered a specific analysis, and the underlying characteristics for the asset must be assessed (e.g., whether reliable reserves estimates are available and whether cost studies have been undertaken to understand the specific infrastructure that would be required).

ii Suitability of the relative valuation (market-based) approach

Compared to a DCF approach, a relative approach is more simplistic and has the benefit of being grounded in market data, which represents the unbiased valuations of many buyers and sellers. A relative approach is most useful when it is possible to identify a sample of companies that are sufficiently comparable to the asset in question to generate useful valuation multiples (e.g., P/E and EV/EBITDA). Further, the comparable companies should have reliable market data on which to base the valuation multiples.

The difficulty with adopting a relative valuation is that truly comparable companies with available market and financial data can be difficult to find. It is necessary to consider the operations of potentially comparable companies, geographical location, growth prospects and financial performance. In many cases the multiples will vary substantially across the potentially comparable companies. This is a clear signal that the companies are not sufficiently comparable.

Although a relative valuation is fundamentally a forward-looking valuation approach, it is a very blunt tool. Typically, the chosen P/E or EV/EBITDA multiple is applied to only one year of financial data, meaning that any medium-term or long-term expectations about revenues or profits of the asset being valued are truncated. This is of particular concern if the valuation date occurs during a period of uncertainty. In this case the contemporaneous data cannot be assumed to represent a reasonable proxy for the continuing expectations of the company.

In Hulley v. Russia, the tribunal adopted the comparable companies methodology (as a starting point with a number of subsequent corrections) only after it found the claimant's DCF analysis unreliable (not the DCF approach itself), and also dismissed the comparable transactions analysis:

the Tribunal concludes, for the reasons set out below, that the corrected comparable companies figure is the best available estimate for what Yukos would have been worth on 21 November 2007 but for the expropriation.
The Tribunal finds that neither of the other two primary valuation methods put forward by Claimants is sufficiently reliable to ground a determination of damages for this case. On balance, the Tribunal was persuaded by [Respondent's expert's] analysis of Claimants' DCF model, and is compelled to agree that little weight should be given to it. The Tribunal observes that Claimants' expert admitted at the Hearing that his DCF analysis had been influenced by his own pre-determined notions as to what would be an appropriate result. Similarly, the Tribunal can put little stock in Claimants' calculations based on the comparable transactions method, since both Parties agree that, in fact, there were no comparable transactions, and thus no basis that would allow a useful comparison.16

We also note, however, that the relative valuation approach may be the preferred approach when valuing banks or other financial institutions (e.g., using a price-to-book ratio). This is because, unlike other operating companies, the book value of a bank's assets and liabilities should be similar to the market value of those assets and liabilities, making the book value a more reliable basis for comparison across firms.

iii Suitability of the cost approach

A cost approach is generally only useful if the amount invested in the asset represents the current market value of that asset. This is reflected in the award in Rusoro v. Venezuela:

Under certain conditions, the historic amounts invested by an investor may be a relevant yardstick to establish the correct fair market value of the investment at the time of expropriation. If
- the investment consisted in the arms' length acquisition from non-related third parties of existing enterprises,
- the time period between investment and expropriation is not unreasonably long,
- no major micro- or macroeconomic disruptions have occurred between investment and expropriation, and
- the fundamental parameters of the enterprise have not been subverted (e.g. in an investment in the natural resources sector, no significant new findings have been made),
- the actual amount invested may bear a relationship with the fair market value at the time of expropriation and, with the appropriate adaptations for the passage of time and the change in market conditions, may yield a relevant yardstick for the valuation.17

It is typically not the case that the historical cost reflects current market values. However, cost or replacement value can be a useful measure of market value for tangible items such as buildings, fixed assets or inventory. It also has the benefit of being objective, with the amount invested being supportable by evidence and not a matter of expert opinion.

In practice, except in specific circumstances such as fixed assets, this method is often submitted as an alternative to the primary valuation method (typically based on DCF or relative valuations) or where there are specific legal reasons.18 Likewise, tribunals appear to resort to this basis in the absence of a more reliable, forward-looking valuation.

In Copper Mesa v. Ecuador, the tribunal found that the proven expenditure incurred provided the most reliable, objective and fair method for valuing the claimant's investment, having dismissed the DCF valuations proffered on the basis that the claimant's concessions 'remained in an early exploratory stage with no actual mining activities, still less any track record as an actual mining business; and particularly as regards [one of the ] concessions, that the Claimant's chances of moving beyond an exploratory stage were . . . slender'.19

Similarly, in Vivendi v. Argentina, the tribunal found that the 'investment value' appears to offer the 'closest proxy, if only partial, for compensation sufficient to eliminate the consequences of the [Respondent's] actions'.20

V Choosing more than one approach

Many experts choose to use more than one valuation approach, thereby providing the tribunal with valuations based on a range of information. There are several reasons why this can be useful.

Fair market value seeks to determine the price in a hypothetical transaction. In actual transactions (i.e., the acquisition of an operating company), it would be typical to consider several valuation approaches and several circumstances. Advisers to buyers and sellers of companies often provide both DCF analyses and relative valuations based on comparable companies or comparable transactions, thereby providing a range of values of the asset.

Relying on both a DCF analysis and a relative approach has the benefit of including an analysis of the specific cash flow projections and risk of the company (through the DCF) and incorporating market pricing data (through the comparable company multiples). Owing to the simplicity of a relative valuation, it is often used as a benchmark for a more substantive approach (e.g., a DCF). Seeking to extract the benefits of both approaches is echoed in Damodaran's textbook, Investment Valuation:

Philosophically, there is a big gap between discounted cash flow valuation and relative valuation. In discounted cash flow valuation, we take a long-term perspective, evaluation a firm's fundamentals in detail, and try to estimate a firm's intrinsic value. In relative valuation, we assume that the market is right on average and estimate the value of a firm by looking at how similar firms are priced. There is something of value in in both approaches, and it would be useful if we could borrow the best features of relative valuation while doing discounted cash flow valuation, or vice versa.21

In the context of a transaction, using more than one approach provides a basis for negotiation between the buyer and seller. In contrast, an arbitral tribunal is required to determine and award a specific figure. In that sense, it is necessary to provide guidance or weight to the alternative valuation approaches employed, which must be based on expertise and an assessment of the reliability of each approach.

VI Conclusions

The choice of approach does not dictate the magnitude of value

From an economic perspective, the choice of a valuation approach should be dictated by the characteristics of the asset and the information available. The DCF approach is the natural first consideration as it is the fundamental basis of valuation. It has been gaining traction with arbitral tribunals over the years. A study of arbitral awards by the International Chamber of Commerce found that tribunals accepted the use of DCF approach 85 per cent of the time it was proposed.22 However, the growing acceptance of DCF as the basis of damages claims has faced a recent backlash.23

The criticism of DCF seems often to stem from the magnitude of claims (supported by DCF analysis) relative to substantially lower sums invested (i.e., investment cost).24 However, as noted above, the approach to valuation does not dictate the valuation output, whether large or small. The chosen valuation approach is merely a tool, which can be used appropriately or inappropriately. Given that the DCF approach should provide a transparent and working valuation framework, identification of any erroneous inputs that may cause either undervaluation or overvaluation, should be possible. Furthermore, high-level assessments should be considered. That is, it should be possible to address the fundamental driver of value, such as patents, specific expertise and monopoly rents, among other things. If the fundamental drivers are not evident, it is likely that flawed underlying assumptions are responsible for large valuations relative to the amounts invested.

The choice of an approach must be appropriate for the circumstances, but it is the assumptions and inputs that drive the valuation result and are often the cause of large discrepancies between experts. We should keep in mind that, no matter the chosen approach, the output will only be reliable if the inputs are robust and reasonable. If the inputs and assumptions are not robust, the assessed value will not be reliable regardless of the approach taken – hence the saying 'garbage in, garbage out'.


Footnotes

1 Jessica Resch, Maja Glowka and Tim Giles are partners at Independent Economics & Finance LLP.

2 The World Bank Group, 'Guidelines on the Treatment of Foreign Direct Investment', para. 40.

3 ibid., Section IV, para. 5.

4 Black's Law Dictionary (n2), 844.

5 Note that we approach the concept of fair market value from an economic perspective but understand that there is a debate about whether any idiosyncratic synergy value should be included in a valuation for the purpose of investment treaty arbitration. See, e.g., Sergey Ripinsky and Kevin Williams, Damages in International Investment Law (British Institute of International and Comparative Law), p. 185–86.

6 See, e.g., IFRS 13 (International Financial Reporting Standards Foundation): 'A fair value measurement of a non-financial asset takes into account a market participant's ability to generate economic benefits by using the asset in its highest and best use or by selling it to another market participant that would use the asset in its highest and best use.'

7 See also Chapter 29 of this publication.

8 Aswath Damodaran, Investment Valuation (3rd ed., Wiley Finance), p. 11.

9 ibid., p. 12.

10 ibid., p. 19.

11 ibid., p. 17.

12 World Bank, 'Guidelines on the Treatment of Foreign Direct Investment', Section IV, para. 6.

13 CMS Gas Transmission Company v. The Republic of Argentina Award, ICSID Case No. ARB/01/8 (12 May 2005), para. 416.

14 World Bank, 'Guidelines on the Treatment of Foreign Direct Investment', Introductory Remarks, para. 42.

15 Gold Reserve Inc v. Bolivarian Republic of Venezuela Award, ICSID Case No. ARB(AF)/09/1 (22 Sep. 2014), para. 830.

16 Hulley Enterprises Limited (Cyprus) v. Russian Federation, PCA Case No. ARB(AF)/09/1 (22 Sep. 2014), paras. 1784-785).

17 Rusoro Mining Limited v. The Bolivarian Republic of Venezuela (Rusoro v. Venezuela), Award, ICSID Case No. ARB(AF)/12/5, para. 772.

18 For example, in Rusoro v. Venezuela, the Nationalization Decree specifically provided for book value to be paid as compensation – para. 689.

19 Copper Mesa Mining Corporation v. Republic of Ecuador Award, PCA No. 2012-2 (15 Mar. 2016), para. 7.24.

20 Compañiá de Aguas del Aconquija S.A. and Vivendi Universal S.A. v. Argentine Republic Award, ICSID Case No. ARB/97/3 (21 Nov. 2000), para. 8.3.13.

21 Aswath Damodaran (op. cit. note 9, above), p. 930.

22 'Damages awards in international commercial arbitration: A study of ICC awards' (Dec. 2020), PwC and Queen Mary University of London, p. 15. We note that this was a study of ICC awards as opposed to investment treaty awards, but from an economic perspective we see no distinction between the two for the purposes of valuation for damages.

23 As seen in the proposals outlined in a note prepared by the UNCITRAL Secretariat to the Working Group III on the possible ISDS reform of the assessment of damages and compensation (https://uncitral.un.org/sites/uncitral.un.org/files/media-documents/uncitral/en/assessment_of_damages_and_compensation.pdf), and responses thereto (including https://www.iisd.org/system/files/2021-11/ccsi-iisd-iied-unictral-damages-november-2021.pdf) (web pages last accessed 28 Mar. 2022).

24 See, e.g., Sergey Ripinsky and Kevin Williams (op. cit. note 5, above), p. 206.

The Law Reviews content