Valuation of Expropriated Property under Investment Treaty Law – On the Distinction between Lawful and Unlawful Expropriation Kandidatnummer

Valuation of Expropriated
Property under Investment
Treaty Law
– On the Distinction between Lawful and Unlawful Expropriation
Kandidatnummer: 639
Leveringsfrist: 25.4.2015
Antall ord: 17264



Table of Contents
1 INTRODUCTION ……………………………………………………………………………………………. 1
1.1 Short Introduction to Investment Treaty Law on Expropriation ……………………………….. 1
1.2 Object and Purpose ……………………………………………………………………………………………. 2
1.3 Research Methodology and Limitations ……………………………………………………………….. 3
1.4 Structure of the Thesis ……………………………………………………………………………………….. 4
EXPROPRIATION ………………………………………………………………………………………….. 5
3.1 The Standard in International Investment Agreements……………………………………………. 9
3.2 Fair Market Value ……………………………………………………………………………………………. 10
3.3 The Date of Valuation ………………………………………………………………………………………. 11
4 VALUATION OF FAIR MARKET VALUE …………………………………………………… 13
4.1 Choosing a Valuation Method …………………………………………………………………………… 13
4.2 Valuation Methods …………………………………………………………………………………………… 16
4.2.1 The Income-Based Approach ………………………………………………………………… 16
4.2.2 The Market-Based Approach…………………………………………………………………. 19
4.2.3 The Asset and Cost Based Approach ……………………………………………………… 24
EXPROPRIATIONS ……………………………………………………………………………………… 29
5.1 Customary International Law…………………………………………………………………………….. 29
5.2 Full Reparation ………………………………………………………………………………………………… 31
5.2.1 The Value of Restitution……………………………………………………………………….. 33
5.3 Heads of Damage …………………………………………………………………………………………….. 35
5.4 The Date of Valuation for Unlawful Expropriation ………………………………………………. 37
5.4.1 Introduction ………………………………………………………………………………………… 37
5.4.2 Chorzów’s Date of Valuation ………………………………………………………………… 37
5.4.3 Creeping Expropriations – The Date of Expropriation? …………………………….. 41
6 FINAL REMARKS ………………………………………………………………………………………… 43
7 REFERENCES ………………………………………………………………………………………………. 45
7.1 Bibliography……………………………………………………………………………………………………. 45


7.2 Table of Cases …………………………………………………………………………………………………. 49
7.3 Table of Treaties ……………………………………………………………………………………………… 51


1 Introduction
1.1 Short Introduction to Investment Treaty Law on Expropriation

Provisions on expropriation of foreign investments are a common feature of bilateral and mul-
tilateral investment treaties (hereinafter investment treaties).1 These provisions seek to bal-
ance two potentially conflicting interests; (i) the investor’s property rights and (ii) the host
state’s need for regulatory freedom in the pursuit of public purposes.

Recognizing the interests of both parties, investment treaty provisions afford the state with a
right to expropriate if compensation is paid to the investor. In addition, lawful expropriations
have to serve a public purpose, be done in accordance with due process of law, and be con-
ducted in a non-discriminatory manner. If the state does not comply with the treaty require-
ments for a lawful expropriation, the state commits an unlawful act for which it is obligated to
make full reparation for the injury caused.2

The compensation standard for lawful expropriation in investment treaties is generally the
standard of fair market value.3 The fair market value of an expropriated investment is the
price that the investment would have traded at in a hypothetical commercial transaction. The
standard is therefore detached from the investor and can be regarded as an objective standard
of compensation.

The primary remedy for making reparation for an unlawful expropriation is restitution in kind;
however, this is in practice rarely claimed and awarded due to the problems associated with
enforceability of awards against the state.4 Instead tribunals award damages equal to the in-
vestor’s loss. Reparation is thus a subjective standard aimed at wiping out the consequences
of the unlawful act.

Investment treaties also contain dispute settlement provisions that are a standing offer to the
investor to initiate arbitral proceedings against the state for claims based on the treaty provi-
sions. If successful, investors are then left with a pecuniary award, which can normally be
enforced in accordance with either ICSID Convention or the New York Convention.5 In-

1 According to UNCTAD (2014) p. 114 there were 3,236 investment treaties in force at the end of 2013
2 Marboe (2007) p. 725
3 Ibid p. 730
4 See Ripinsky (2008) p. 57-59 for an analysis on why restitution is rarely claimed.
5 ICSID Convention Article 54(1) and New York Convention Article III both leave the investor with a binding
award that can only be denied recognition in limited circumstances


vestment treaties thus provide the investor with an important dispute settlement mechanism,
which ensures that the investor’s rights are effective and not illusory.

1.2 Object and Purpose

The aim of this thesis is to give an account of how investment arbitration tribunals arrive at
the ultimate sum awarded for a claim based on a finding of an expropriation. This necessitates
separate studies of the different standards for compensation and damages.

The thesis supposes that there has been a taking of a foreign investment that is attributable to
the state, and for which an arbitral tribunal is to award either compensation or damages. Con-
sequentially the thesis does not consider the jurisdictional requirements of who qualifies as an
investor and what qualifies as an investment in the treaty, nor does it attempt to answer when
a state act passes the threshold for expropriation. Mitigation, causation and contributory fault,
although important when considering compensation and damages, also fall outside of the
scope of this thesis as the focus here is on the primary loss caused. The same goes for moral
damages as these are not financially assessable. Interest will only be considered as a head of
damage as the issue of interest for lawful expropriation is an unsettled and complicated topic
that warrants a more thorough examination than is permitted here.

With regards to damages there has been a development in recent case law towards awarding
the economic difference between the investor’s actual present day position and the hypothet-
ical position of the investor if the state had not acted unlawfully. This means that if the gen-
eral economic conditions of the investment improved in the interim period between the ex-
propriation and the award, the state has to pay damages reflecting this change. Additionally, it
seems that the tribunals have accepted that the value of the investment at the date of the ex-
propriation and any consequential loss represents the lower limit of damages due.6 The conse-
quences of such a view are that states now bear the risk of both favorable and unfavorable
changes in the fair market value of an unlawfully expropriated investment. This has rendered
the distinction between lawful and unlawful expropriation more important, and naturally, the
cases that have taken this approach will be afforded quite a bit of attention in this thesis.

Furthermore, the thesis will look at the valuation methods used in investment arbitration prac-
tice for awarding compensation adhering to these standards.

6 See e.g. ADC v Hungary, Kardassopoulos v Georgia, Yukos Owners v Russia, Unglaube v Costa Rica


For investors contemplating dispute settlement, the concern is not just the legal basis of a
claim but maybe more so the amount of compensation that they can expect to receive after a
lengthy and costly legal battle.7 Precise estimates of how much an investor can expect to
achieve, if successful, through investment treaty arbitration is of vital importance in order to
assess the risk of such a process. As the sum ultimately awarded is largely based on methods
of valuation the most commonly used methods are considered.

Sound valuation methods are also vital to the well-functioning of investment treaties. On the
one hand, too high valuations of losses affect the state and ultimately the people of the state.
Excessive awards can also lead to less than efficient number of expropriations as states might
refrain from expropriating in cases where it would have been efficient if compensation was
set correctly. Furthermore, excessive awards can create an incentive for the costly and slow
process of investment arbitration rather than alternative means of dispute settlement.8 On the
other hand, too low valuation of losses can erode faith in property rights and lead to opportun-
istic expropriation with the likely effects of reduced foreign investment, which is contrary to
the goal of investment treaties.9

1.3 Research Methodology and Limitations

This thesis is an analytical study of primarily recent case law in investor-state arbitration,
where the investor has been awarded either compensation or damages. The study is limited to
claims forwarded under the dispute-settlement provisions of bilateral and multilateral invest-
ment treaties. An important limitation to the material is that only publicly available cases are
analyzed. Another limitation of the analysis is that quite often some of the submitted material
used for valuation purposes often is not available. In addition to case law, I will also be rely-
ing on some influential scholarly works including those of Ripinsky with Williams,10 Mar-
boe,11 and for valuation purposes, Kantor.12

In studies of case law in international investment law there are two important limitations to
the findings. Firstly, the lack of a doctrine of stare decisis or binding precedents in interna-
tional law means that the future tribunals are not obliged to follow past practice.13 Secondly,

7 Marboe (2009) p. 2
8 Wells (2003) p. 478
9 Ibid p. 481
10 See Ripinsky (2008)
11 See Marboe (2009)
12 See Kantor (2008)
13 Commission (2007) p. 134 with reference to Statute of the International Court of Justice Article 59


tribunals cannot go beyond the claims submitted by the parties, which might lead to awards
being rendered contrary to the tribunal’s true view on the law.14

Despite the lack of stare decisis in international law, tribunals and parties to investment dis-
putes are to an increasing degree referring to past awards. Over time this might lead to de fac-
to precedents in international investment law. The view of tribunals on previous awards is
well described in an oft-quoted paragraph found in the case of Saipem v Bangladesh:

“The Tribunal considers that it is not bound by previous decisions. At the same time, it is of
the opinion that it must pay due consideration to earlier decisions of international tribunals.
It believes that, subject to compelling contrary grounds, it has a duty to adopt solutions estab-
lished in a series of consistent cases. It also believes that, subject to the specifics of a given
treaty and of the circumstances of the actual case, it has a duty to seek to contribute to the
harmonious development of investment law and thereby to meet the legitimate expectations of
the community of States and investors towards certainty of the rule of law.”15

The rationale behind this passage explains why this thesis is primarily a study of past cases.

1.4 Structure of the Thesis

The remainder of this thesis divided into five chapters: (i) Chapter 2 provides a short discus-
sion on the legal distinction between a lawful and an unlawful expropriation, (ii) Chapter 3
examines the compensation standard for lawful expropriations in detail, (iii) Chapter 4 deals
with the valuation techniques used in arbitration practice for awarding compensation adhering
to the standard for lawful expropriations, (iv) Chapter 5 examines the standard of compensa-
tion for unlawful expropriations, and (v) In Chapter 6 I provide some final remarks.

14 Ripinsky (2008) p. 57
15 Saipem v Bangladesh paragraph 90


2 The Distinction between Lawful and Unlawful Expropriation

States enjoy sovereignty over their territories and resources.16 Their sovereignty is limited by
international law and obligations that the state has accepted through treaties. In investment
treaties states generally accept obligations not to “nationalize or expropriate an investment of
an investor of another Party in its territory or take a measure tantamount to nationalization
or expropriation”,17 or that investments cannot be “expropriated, nationalized or subjected to
other measures having a similar effect”, unless the state complies with the treaty requirements
for expropriating.18 What constitutes an expropriation is normally not defined in any detail in
the treaties. Essentially, an expropriation entails a seizure by the state of the investment and/or
a transfer of legal title to the investment to the state or a state-mandated third party.19 Nation-
alization is a large-scale expropriation of an entire industry or sector coupled with a transfer
of the legal title to the property to the state. Expropriations and nationalizations are normally
referred to as direct expropriations.

Measures that have “a similar effect” or are “tantamount to nationalization or expropriation”
are generally referred to as indirect expropriation. Indirect expropriations are measures that
leave the legal title to the property intact but interfere with the investor’s ownership rights
through regulatory acts and omissions, to the extent that it has the same effect as an expropria-
tion of the investment. Indirect expropriations often take place in the form of creeping expro-
priation. Stern explains creeping expropriation as “a process extending in time and comprising
a succession of measures that, taken separately, do not have the effect of dispossessing the
investor but when taken together do lead to such a result.”20

Whether a state act is held to be a direct expropriation or an indirect expropriation is of little
legal importance because both are normally covered by the investment treaty text and are as
such just subspecies of expropriation. However, in practice, indirect expropriations rarely
comply with the due process and compensation requirements and are, as will be discussed
below, unlawful.21

If a tribunal holds that a state has expropriated an investment, the tribunal has to determine the
lawfulness of the expropriation. When deciding on the legality, the tribunal has to pay atten-

16 Sornarajah (2010) p. 119
17 NAFTA Article 1110(1)
18 Norway-Russia bilateral investment treaty (BIT) Article 5
19 UNCTAD (2011) p. 6-7
20 Stern (2008) p. 36
21 Marboe (2009) p. 61


tion to the specific wording of the investment treaty. Notwithstanding that treaties are worded
differently, the substance of provisions on expropriation is to a degree standardized. This is
shown by the comprehensive survey done in 2007 by the United Nations Conference on Trade
And Development (UNCTAD):

“Most agreements include the same four requirements for a lawful expropriation, namely
public purpose, non-discrimination, due process and payment of compensation. Furthermore,
most BITs have similar provisions regarding the standard of compensation. Notwithstanding
some variations in language, the overwhelming majority of BITs provide for prompt, ade-
quate and effective compensation, based on the market or genuine value of the investment.”22

The requirement of a public purpose essentially means that the taking “must be motivated by
the pursuance of a legitimate welfare objective, as opposed to a purely private gain or an
illicit end.”23 The public purpose must be present at the time of the taking but does not depend
on the ultimate achievement of the goal. Conversely, the requirement is not fulfilled if the
taking of property initially had no public purpose but is later used to serve a public purpose.24
Tribunals have tended to afford states a wide margin of appreciation in determining if an ex-
propriation has a public purpose.25 A finding of lack of public purpose is thus rare but not
unheard of.26

The requirement of non-discrimination means that the expropriation cannot be based on the
foreign national belonging to “a specific racial, religious, cultural, ethnic or national
group.”27 This requirement is not violated simply because the expropriation targets a foreign
investor; the expropriation must be motivated by one of the investor’s specific traits.

The due process requirement is formulated differently in the treaties and the fulfillment of this
requirement will necessarily depend on the investment treaty formulation. In general, the due
process requirement will require that the expropriation complies with the procedures of the
domestic legislation and internationally recognized principles on due process, and that the
investor is afforded with a right to an independent review of the case and the compensation
due.28 Case law on the requirement is limited, but one notable case is that of ADC v Hungary.

22 UNCTAD (2007) p. 52
23 UNCTAD (2011) p. 28-29
24 Ibid p. 31
25 Newcombe (2009) p. 371
26 See ADC v Hungary, paragraphs 429-433
27 Newcombe (2009) p. 373
28 Reinisch (2008) p. 191-193, UNCTAD (2011) p. 36


The International Centre for Settlement of Investment Disputes (ICSID) Tribunal in ADC
explained that the requirement requires:

“an actual and substantive legal procedure for a foreign investor to raise its claims against
the depriving actions already taken or about to be taken against it. Some basic legal mecha-
nisms, such as reasonable advance notice, a fair hearing and an unbiased and impartial ad-
judicator to assess the actions in dispute, are expected to be readily available and accessible
to the investor to make such legal procedure meaningful.”29

For indirect expropriations such procedures will rarely be available to the investor because the
expropriatory measure(s) are normally regulatory acts that are not directed towards the inves-
tor. The absence of even the possibility to challenge the expropriation means that indirect ex-
propriations normally are unlawful.30

A majority of investment treaties provide for prompt, adequate and effective compensation or
phrases that are generally interpreted as having the same meaning.31 Adequate compensation
is normally understood as the fair market value of the investment and will be dealt with in
chapter 3 and 4. Many treaties also contain detailed rules on the precise methods that are to be
used for calculating the compensation due.32 A payment of compensation is normally effec-
tive if the payment is made in “in convertible or freely useable currency”.33

With regards to the promptness of the payment of compensation and its importance for the
legality of the expropriation there does not seem to be an established consensus in investment
arbitration practice. Often a tribunal will conclude that an expropriation was, regardless of the
fulfillment of the other requirements, unlawful “because no compensation had been paid.”34
In another case one might see a tribunal state that even though no compensation had been paid
prior to the arbitral proceedings “it suffices to conclude that the present expropriation was
lawful, since it wants only compensation”35

It has been suggested that if there has been a non-payment of compensation this should not by
itself render an expropriation unlawful. A more nuanced approach has been suggested where-

29 ADC v Hungary paragraph 435
30 Marboe (2009) p. 61, Reisman (2004) p. 137 footnote 104
31 UNCTAD (2007) p. 48
32 Reinisch (2008) p. 195-196
33 UNCTAD (2011) p. 40
34 Vivendi v Argentina paragraph 7.5.21
35 Tidewater v Venezuela paragraph 146


by only non-payment for an unreasonable period of time and/or negotiation in bad faith
should affect legality. However, if there has been a non-payment but the requirements for
good faith are fulfilled, the act of non-payment itself should not render the expropriation un-
lawful as even good faith application of accepted valuation guidelines can lead to diverging
results, which in turn could necessitate third party mediation.36An example of a tribunal sup-
porting this nuanced approach can be found in the case of ConocoPhillips v Venezuela:37

“The requirements for prompt payment and for interest recognise, in accordance with the
general understanding of such standard provisions, that payment is not required at the pre-
cise moment of expropriation. But it is also commonly accepted that the Parties must engage
in good faith negotiations to fix the compensation in terms of the standard set, in this case, in
the BIT, if a payment satisfactory to the investor is not proposed at the outset.”38

In this case the ISCID Tribunal concluded that Venezuela had not negotiated in good faith by
relying on book value as the method of valuation because this would not comport to the in-
vestment treaty standard of fair market value. The Tribunal thus held that Venezuela had act-
ed unlawfully.39

There are several compelling arguments to support this nuanced approach. First of all valua-
tion is not a science, as it necessarily requires complicated judgements on value. Furthermore,
this gives tribunals the necessary flexibility to distinguish between egregious acts of confisca-
tion and bona fide expropriations only wanting compensation. This being said there still
seems to be conflicting views on the importance and content of the requirement.

36 Ripinsky (2008) p. 67-69
37 Tidewater v Venezuela and Mobil v Venezuela are also cases that take a similar approach to the requirement
38 ConocoPhillips v Venezuela paragraph 394
39 Ibid paragraphs 394, 400-402


3 The Standard of Compensation for Lawful Expropriations
3.1 The Standard in International Investment Agreements

The standard of compensation in customary international law for lawful expropriation has
been a hotly debated topic until recently. Traditionally, developed states argued for a standard
of “full compensation”, which entails compensation equal to the fair market value of the taken
investment. Developing countries have argued that expropriation demands a standard of na-
tional treatment or a standard that provides for less than fair market value.40

The issue of the compensation standard in customary international law has lost a lot of its
importance due to rapid expansion of investment treaties. Investment treaties that contain pro-
visions on expropriation are recognized by arbitral tribunals as “a lex specialis whose provi-
sions will prevail over rules of customary international law.”41 This makes the relevant stand-
ard of compensation for lawful expropriations in investment arbitration the specific binding
treaty language.42

The standard of compensation for lawful expropriations that can be found in investment trea-
ties are on their surface different. Examples include ‘real value’43 ‘market value’,44 ‘prompt
adequate and effective’,45 and ‘genuine market value’,46 to name a few. In sum it might be fair
to say that most treaties provide for compensation equal to the fair market value of the in-
vestment.47 In investment arbitration practice, terms such as ‘value’,48 ‘genuine value’,49
‘market value’50 and ‘actual value’51 have been interpreted as a reference to fair market value.

Some investment treaties could be interpreted as providing for something less than fair market
value. These are not dealt with in this thesis because the standard of fair market value is the
prevalent standard, and since it will necessarily depend on the circumstances of the particular

40 Newcombe (2009) p. 377
41 ADC v Hungary paragraph 481
42 Wälde (2008) p. 15-16
43 Australia-Hongkong BIT Article 6(1)
44 Greece-South Africa BIT Article 4(1)
45 Australia-Czech Republic BIT Article 6(1)(c)
46 Italy-Tanzania BIT Article 5(1)
47 Ripinsky (2008) p. 78-79
48 Siemens v Argentina paragraph 353
49 CME v Czech Republic Partial Award paragraph 624
50 Tidewater v Venezuela paragraphs 151-152
51 Vivendi v Argentina paragraph 8.2.10


case and thus be hard to generalize about how to value an expropriation according to such a
standard. The focus in the following is therefore on fair market value.

3.2 Fair Market Value

In the context of compensation it is clear that when an investment treaty makes reference to a
value then it is the economic value of the investment. The economic value of an investment is
not possible to determine by its characteristics but has to be determined by reference to a dis-
tinct perspective. Marboe has explained the concept of value in the following way:

“Value, however, is not an objective quality of things. It always depends on a specific rela-
tionship between the particular object and a subject. As Immanuel Kant pointed out, value
may only be understood as appreciation by persons. Without the needs and affections of peo-
ple, things would not have any value. ‘Value’, therefor, is a relative concept.”52

The value a specific person puts on an object depends, as the quoted passage suggests, on the
preferences of that person. Another way of describing this is that the economic value to a per-
son is equal to a person’s reservation price, which is determined by that person’s disposable
income, preferences and the prices of substitute goods. If a market for a property exists, the
subjective perspectives of value by the large numbers of buyers and sellers are balanced into
an objective value for a property.53 One definition of fair market value is:

“The estimated amount of which property should exchange on the date of valuation between a
willing buyer and a willing seller in an arm’s length transaction after proper marketing
wherein the parties had each acted knowledgeably, prudently, and without compulsion.”54

As the quoted definition suggests, the concept of fair market value is to find the price the
property would trade at in a hypothetical commercial transaction between a willing seller and
a willing buyer. Certain assets/businesses are not traded in the market, which means that there
in reality does not exist a market for the asset/business. The valuators, and ultimately the arbi-
tral tribunal, will in such circumstances have to rely on a valuation method that determines the
price that the asset would trade at in a hypothetical market. The concept of fair market value
is in these cases a legal construct.55

52 Marboe (2009) p. 22 – Footnotes omitted
53 Ripinsky (2008) p. 183 with reference to Marboe (2007) p. 735-736
54 Kantor (2008) p. 31
55 Ibid p. 58-59


A common misinterpretation of the standard of fair market value is that it is synonymous with
the price paid for the property. The price paid for a property is merely the subjective valuation
of a particular buyer for the property and, therefore, just a historic fact. The standard of fair
market value is an objective valuation. Consequently, the standard does not consider what is
called special value. Special value refers to features of the property that makes it particularly
attractive to a specific buyer.56 A typical source of special value is synergistic value, such as
operational synergy and financial synergy, which is created by combining two or more as-
sets.57 This means that the possible special value an investment has to an investor is not com-
pensated under the fair market value standard.

When considering the value of a property the principle of highest and best use applies. Simply
put, the principle of highest and best use means that the valuation of a property is not depend-
ent upon the current use of the property, but has to be determined according to its best use and
thus the highest value that it could be put to use in. The ICSID Tribunal in Unglaube v Costa
Rica, which concerned the expropriation of a beachfront property, explained this principle as

“If, as Claimants’ expert has suggested, it is appropriate, in determining fair market value, to
identify the highest and best use of this particular property, it seems plain to the Tribunal that
that can only be the highest and best use subject to all pertinent legal, physical, and economic
constraints.”58 (my underscore)

In the Unglaube case the Tribunal looked at the specifics of the property in question and held
that the property had to be valued based on a usage “appropriate to the environmentally-
sensitive surroundings” and “with a density comparable to that permitted by the guidelines set
forth in the 1992 Agreement”.59 Even though the property in the case potentially could have
been used to build a large-scale hotel on the beach, the principle of highest and best use ex-
cluded this for valuation purposes, as it was not legally feasible.

3.3 The Date of Valuation

The valuation of compensation or damages for an expropriated investment is highly depend-
ent on the date of valuation. Value changes constantly as new information on an investment
and its operating conditions is revealed. Consider a company that discloses a lawsuit against it

56 Marboe (2009) p. 177
57 Damodaran (2006b) p. 1-3
58 Unglaube v Costa Rica paragraph 309
59 l.c.


claiming $100 in damages, which potential buyers perceive as having a 90% chance of suc-
ceeding. A rational willing buyer would incorporate this information into his valuation of the
company and reduce the value by $90, which is the expected value of the lawsuit (0.9 ×
$100). As this example shows, value changes as new information is disclosed and willing
buyers price in this information in their valuations.60

Since the goal of a valuation is to determine the value as of the date of valuation, only infor-
mation preceding the date of valuation (ex-ante information) is relevant. Information on
events that occur after the valuation (ex-post information) would not be available to buyers on
the date of valuation and should therefore be disregarded. As the discussion below will show,
ex-post information is sometimes used in order confirm the reasonableness of assumptions
made in the submitted valuations.61 This is because ex-post information on actual earnings
often can be the most reasonable assumption on the expectations of buyers and sellers as of
the date of valuation.62

With regards to lawful expropriations, determining the date of valuation is normally a
straightforward exercise. Most investment treaties provide that whichever is earlier, the mo-
ment of expropriation or the moment that the expropriation became public knowledge, is the
date of valuation.63 In practice, this will normally be identifiable by a government decree ex-
pressing the intent to expropriate or actions such as an outright seizure of the investment. The
rationale behind this rule is to exclude any effect that disclosure of information on a state’s
intent to expropriate would have on the market value of an investment. If this information was
included the state would benefit from its own actions through a lower valuation of the invest-
ment.64 From the date of valuation states expropriating businesses assume the equity risk of
the investment when expropriating lawfully, as subsequent events that decrease or increase
the value of the investment are irrelevant to the valuation.

60 Ripinsky (2008) p. 243
61 This is discussed in chapter 4.2.1 on the income-based approach
62 Abdala (2007) p. 3-4
63 Ripinsky (2008) p. 243-244
64 Ibid p. 244


4 Valuation of Fair Market Value
4.1 Choosing a Valuation Method

The focus in this chapter is on some of the most common valuation methods that can be used
in order to determine the fair market value of an investment. If there is an active market for
the property in question, there would not be a need for a valuation method as the price at
which the investment trades at would be the market value. A tribunal would of course rely on
this price in the award.65 However, in many circumstances, if not most, it is not possible to
fetch a price from the market as the investment is unique and/or not traded in the market. This
necessitates the use of valuation methods.

There is no single method that is appropriate to use for all kinds of property. Some investment
treaties specify the accepted methods but since no method fits all tribunals are afforded dis-
cretion on the choice of valuation method.66 The choice of valuation method is primarily de-
termined by the property being valued. In this regard, it is useful to distinguish between meth-
ods used for valuing businesses and methods used to value individual assets. Since most tak-
ings affect entire businesses, the weight of the discussion here is put to valuation methods that
are appropriate to value these. Some of the factors that affect the likely choice a tribunal
might make on valuation method are addressed below.

The most common methods used in investment arbitration practice are the income-based ap-
proach, which is also referred to as the discounted cash flow method (DCF), the market-based
approach and the asset and cost based approach.67 Underlying the common choices by tribu-
nals is also the notion that willing buyers in fact use these methods in practice.68 Additionally,
valuations have on several occasions been based on historic cost. This rarely reflects market
value but is still discussed below due to its usage in practice.69

For businesses that generate income it is appropriate to use methods that value the business on
the worth of the future cash flows to the business. The rationale behind this is that a hypothet-
ical buyer values the business based on the expected future inflows of profit rather than on the
historic cost of the business, since this is the benefit that the buyer can expect to get from the
purchase.70 As these methods value the business on future income they are generally referred

65 Ibid p. 189
66 Ripinsky (2008) p. 192, See for instance Italy-Bosnia and Herzegovina BIT Article 5(4)
67 Marboe (2009) p. 186
68 Ripinsky (2008) p. 187
69 See e.g. Metalclad v Mexico paragraph 122, Vivendi v Argentina paragraph 8.3.13
70 Ibid (2008) p. 195


to as forward-looking methods. The preferred method of tribunals for forward-looking valua-
tions is the DCF method.71

In order for a tribunal to apply the DCF method the investment in question has to pass what
many tribunals have considered the entry requirement for applying the DCF method, which is
that the business is a going concern.72 It is important to note here that a different definition of
going concern is used in investment arbitration than in accounting. In accounting, the term is
normally used for a business that is expected to continue operating for the foreseeable future.
If not then the business is in liquidation. Kantor explains the use of the term in investment
arbitration: “Tribunals employing the term ‘going concern’ to mean several years of profita-
bility are in reality worried about establishing forward-looking compensation ‘with reasona-
ble certainty’.”73 As the quote illustrates tribunals are reluctant to award market value based
on future income if it is unlikely that the business would have earned profit in the future. The
case of Metalclad v Mexico is a good illustration of tribunals’ attitude towards the DCF meth-
od. The ICSID Tribunal noted:

“where the enterprise has not operated for a sufficiently long time to establish a performance
record or where it has failed to make a profit, future profits cannot be used to determine go-
ing concern or fair market value.”74

In the case the Tribunal rejected the DCF method because “the landfill was never operative
and any award based on future profits would be wholly speculative.”75 Instead, the Tribunal
based the valuation on the actual investment. The case suggests that tribunals are unlikely to
apply the DCF method in cases where operations have not begun.

In Asian Agricultural Products v Sri Lanka, the ICSID Tribunal stated that “‘goodwill’ re-
quires the prior presence on the market for at least two or three years, which is the minimum
period needed in order to establish continuing business connections.”76 The minimum period
prescribed here of 2-3 years is perhaps just an indicative period if compelling evidence of
likely future profit is provided. In this regard the case of Vivendi v Argentina, and the more
nuanced take on the evidence necessary for profitability that was expressed there, is of inter-
est. The ICSID Tribunal noted:

71 See e.g. Gold Reserve Inc v Venezuela paragraph 831, Marboe (2009) p. 397
72 Kantor (2008) p. 91
73 Ibid p. 95
74 Metalclad v Mexico paragraphs 119, 121
75 Ibid paragraph 122
76 Asian Agricultural Products v Sri Lanka paragraph 103


“that in an appropriate case, a claimant might be able to establish the likelihood of lost prof-
its with sufficient certainty even in the absence of a genuine going concern. For example, a
claimant might be able to establish clearly that an investment, such as a concession, would
have been profitable by presenting sufficient evidence of its expertise and proven record of
profitability of concessions it (or indeed others) had operated in similar circumstances.”77
(my underscore)

The underscored parts of the quote highlight that past profit is not an absolute requirement for
a going concern valuation. Proof that makes it likely that future profit would be earned such
as expertise and proven record of performance can in a particular case be sufficient for such
valuation. With regards to the standard of evidence the Tribunal in this stated that “convincing
evidence” for profitability in the circumstances faced would have to be provided.78 This high-
lights that the going concern requirement is based on an overall assessment of the probability
for future profit. In most cases, this will require that the claimant can display proof of past

Market-based methods can also be used for valuing profit-generating businesses. The market-
based method is based on the notion that the market value of a business might be adequately
reflected by the value of comparable businesses. If the comparable business is valued on its
profit-generating potential, the method will incorporate that potential in the valuation of the
business.79 The benefit of the method is that is relatively easy to use, however, it can be a
challenging process of identifying comparable companies and relevant performance metric(s)
that drive the value of both companies. Do the businesses being compared have to be in the
same industry? The same country? Of the same size? Can a private company be compared to
a public company? How old can the metrics on the compared business be? These are just
some of the questions that valuators could have to look at in order to justify a valuation based
on the relative value to that of other businesses.80

If forward-looking methods are not considered appropriate asset and cost based methods will
have to used. These do not cover the additional value that a business can have beyond the
value of the sum of the assets. Therefore, these methods should generally only be used when
the business is not worth more than its assets.

77 Vivendi v Argentina paragraph 8.3.4
78 Ibid paragraph 8.3.8
79 Kantor (2008) p. 15
80 For a discussion on this see Kantor (2008) p. 119-130


If the expropriation targets only at a specific asset the likely choice of valuation method de-
pends on the whether the asset is unique or if it is traded in the market. Real estate is in a
sense always unique and is well suited for the use of a market-based method such as the com-
parable transactions method. If the asset is identical or very similar to other assets traded in
the market, the tribunal is likely to use an asset-based approach such as replacement value.
Highly specialized assets, for which there are few or comparable assets traded, will likely lead
to the use of an asset-based method such as adjusted book value.

One of the most important factors affecting the choice of valuation method is the submission
of the parties. If tribunals are not assisted by their own valuation experts, they will rarely have
any other valuations available. Ripinsky points out that “Tribunals are rarely proactive in
using valuation methods; they usually limit themselves to the examination of the methods,
calculations and evidence produced by the parties.”81 Yukos v Russia is a good example of
this as the Permanent Court of Arbitration (PCA) Tribunal eventually relied on one of the
claimant’s corrected submission since it had no other valuation available that had not been

4.2 Valuation Methods
4.2.1 The Income-Based Approach

The goal of the income-based approach/DCF method is to arrive at the present value of a
business’ anticipated future cash flows. The reasoning underlying this method is that the like-
ly price paid in the market for a business is reflected by what the business is expected to earn
the holders of equity interest in the business. In fact the DCF method is often used by willing
buyers in the market, and thus the result reached of such an analysis would be likely to be
close to the fair market value if the assumptions made in the calculation are similar to those of
willing buyers.83

The mechanics of the DCF method is a two-step process. Firstly, the valuator has to calculate
the expected year-by-year cash inflows to the business for the forecast period. If the business
is terminated after the forecast period because, i.e., it is based on a concession that ends then,
there is no need for additional calculations. However, if the company in theory could exist
beyond the forecast period or even indefinitely, the terminal value of the business has to be
forecast and discounted back. Secondly, each yearly figure has to be discounted down to pre-
sent value by applying a discount rate. The discount rate reflects two factors that affect the

81 Ripinsky (2008) p. 234
82 Yukos Owners v Russia paragraphs 1782-1787
83 Ripinsky (2008) p. 195


present value of future cash flows; Firstly, it adjusts for the time value of money since money
received years from now has less real value than the same nominal amount received today. A
payment of a dollar today can immediately start earning interest and is therefore worth more
than a dollar paid ten years from now. Secondly, it adjusts for the risk associated with a future
cash flow.84

Forecasting future cash flows is not an easy exercise as it requires an assessment of the busi-
ness’ likely performance over a period where the operating conditions are constantly chang-
ing. Some of the factors that need to be analyzed include sales, costs, general and industry-
specific conditions and the competition faced. In order to do so the valuator has to consider
the past performance of the business and its value drivers, and make an assumption as to the
likely performance in changing economic circumstances.85 With regards to information, the
general rule is that only information preceding the valuation date is relevant as ex-post infor-
mation would not have been available to a willing buyer then. From the date of valuation until
the date of the award information on what the actual conditions of the business would have
been is revealed. On several occasions tribunals have used this information used in order to
test the reasonableness of the assumptions made in the valuation,86 and as pointed out earlier
ex-post information can often be the most reasonable assumption on the expectations of will-
ing buyers as of the date of valuation.87

Once the cash flows have been determined the figures arrived at have to be adjusted by the
relative sizes of equity and debt and be discounted. In order to do so the valuator could use
either the indirect equity method or the direct equity method. The indirect equity method cal-
culates the present value of the firm by discounting future cash flows to the firm by the
weighted average cost of capital (WACC) rate. In order to reach the equity value, debt then
has to be subtracted. The direct equity method calculates the present value of cash flows to
equity (after having paid all expenses, taxes, interest and principal payments) by discounting
them by the cost of equity.88 Both methods can be expressed by this formula:

Present value = Cash flow/(1+r)t

84 Ibid p. 196-197
85 See Marboe (2009) p. 219-231
86 Ripinsky (2008) p. 254-255 with reference to Santa Elena v Costa Rica, paragraph 84 & Starrett Housing v
Iran, paragraph 18
87 Abdala (2007) p. 3-4
88 Kantor (2008) p. 173-175


r here represents the discount rate and has to be added by one and compounded by a power
equal to t, which represents the number of years. This calculation has to be performed for all
years in the forecast range and include terminal value of applicable. The aggregate of these
calculations is the net present value of the future cash flows.

The WACC rate is calculated as follows:89

WACC = Weight of equity × cost of equity + weight of debt × Cost of debt

The formula for calculating the weight of equity is:

Equity ÷ (debt + equity)

If equity in the numerator is replaced by total amount of debt then the same formula can be
used for calculating the weight of debt.

The cost of debt is based on the effective rate the company has on its current debt. This rate
will have to be calculated based on the business’ outstanding debt. As interest paid on debt
often is tax deductible, calculating this rate might require an adjustment for the deductibility.

Cost of equity can be calculated in several ways. The two most common methods are the
build up procedure and the capital asset pricing model (CAPM). Under the build up procedure
the discount rate is comprised of a base rate and a subjective risk component. The base rate
consists of a risk free component and systematic risk for equity investments. Generally, gov-
ernment bonds of developed countries are regarded as free of risk, and as a sufficient indicator
for the risk free rate.90 The systematic risk is the additional risk of making an equity invest-
ment rather than a debt investment and can be a source of controversy.91 As investment trea-
ties are meant to protect investors against certain political risks, such as the risk of expropria-
tion and unfair and unequitable treatment, these should not be included in the discount rate. In
the case of Gold Reserve Inc v Bolivia the ICSID Tribunal noted:

“However, the Tribunal also considers that the country risk premium adopted by Mr Kaczma-
rek (Navigant) is too low, as it takes into account only labor risks and not other genuine risk
that should be accounted for – including political risk, other than expropriation.”92

89 Ibid (2008) p. 159-163
90 Ripinsky (2008) p. 197
91 Kantor (2008) p. 147-153
92 Gold Reserve Inc v Venezuela paragraph 841


The subjective risk covers the risk that distinguishes an investment in that company from a
generic investment. This risk includes technical risk, financial risk, exchange rate risk, etc.
The calculation of this component is a source of controversy in proceedings as states will tend
to see this risk as very high and investors will view the risk as low.93

CAPM is comprised of the risk free rate and an equity market premium adjusted by a beta
factor. The equity risk premium is based on historical data on the excess risk of making an
equity investment rather than a risk-free one.94 The beta reflects the volatility of the company
and can be based either on statistical analysis of the company’s share prices or on the beta for
comparable companies.95 CAPM assumes that subjective risk can be eliminated through di-
versification, which might necessitate changes to the forecasted cash flows in order to account
for subjective risk.96

As the discussion on this method shows the income-based method is a complicated method of
valuation, which necessitates a range of assumptions both on the size of possible future profit
and the risks faced by the business. With the benefit of hindsight, valuators can to a certain
degree limit the speculative element of the method, however, tribunals can naturally not look
beyond the proceedings, which often makes assumptions about the future a necessary exer-

4.2.2 The Market-Based Approach In General

The market-based approach seeks to determine the value of an asset based on various methods
of comparison of the relative value of the asset to that of other comparable assets. The market-
based approach entails a three-step process. The first step is finding comparable assets that
have been priced by the market. If the assets are not identical in size or units, a second step of
scaling the market price to a common variable is necessary. The final step is to adjust for dif-
ferences in the quality of the asset.97

93 Kantor (2008) p. 155-159
94 Ibid p. 147-149
95 Ibid p. 166
96 Ibid p. 163-165
97 Damodaran (2006a) p. 0-1


Two methods for relative valuation are discussed here; the comparable transactions method
and the comparable companies method. Additionally, stock prices and prior information on
the market value of the property being valued is discussed. Relative Valuation – Comparable Transactions Method

The comparable transactions method is only dealt with through a simple example here. The
method can be made more complex and accurate by including more comparable transactions
in the comparison material and averaging the valuations derived from each comparison. The
example follows the three-step process outlined above.

Let us say that you are contemplating a sale of your apartment and want to have a valuation of
your apartment. If your neighbor’s apartment was sold yesterday, the sales price could be
considered a comparable transaction for the value of your apartment, as the apartment is in the
same neighborhood and even the same building. If your apartment is slightly smaller, then an
adjustment for the difference in number of square meters is appropriate. Finally, an adjust-
ment might also be appropriate if you recently refurnished your apartment whilst your neigh-
bor’s apartment was sold without any improvements in the past 20 years. Relative Valuation – Comparable Companies Method

The comparable companies method seeks to calculate a value based on the value of one or
more comparable businesses. Because the compared company has been priced by the market a
relative valuation of the property will also encompass future profit, as the compared compa-
ny’s market price would be based on the future income.98 Following the three-step process
described above the valuator first has to find a relevant company for comparison. Secondly,
the valuator has to calculate a multiple by dividing the market value of the comparable busi-
ness on a metric that is considered relevant to the value of both businesses. The multiple then
has to be multiplied by the metric for the valued business. Finally, adjustments might be ap-
propriate if the valued company for some reason is deemed to be worth more or less than the
comparable company.

A simple example of the method example could be a comparable company with stock market
capitalization of $100 million and Earnings Before Interest, Taxes, Depreciation and Amorti-
zation (EBITDA) of $5 million. The EBITDA multiple then is 100 ÷ 5 = 20. If the company
being valued has an EBITDA of $20 million then this method leads to a stock value of 20 ×

98 Marboe (2009) p. 203


$20 million = $400 million. This method can in theory be applied to any business that has a
comparable business with a known market price and a common metric.

This example supposes that the value of both companies is 100 % determined by the chosen
metric. Consequently, there is a risk that the value of a company is not fully reflected by the
chosen metric and that the valuation is incorrect. This can to a certain degree be mitigated by
including more metrics and calculating and averaging the results derived from each calcula-
tion. Another way to improve the accuracy of the valuation might be to include more compa-
nies in the multiple calculation by averaging the multiple from each calculation.

The only case that has directly applied this method so far is the case Yukos Owners v Russia.
Interestingly this case resulted in the award of over US$ 50 billion in damages – the largest
award known in investment arbitration. However, the case is not a clear-cut example of the
method as it also relied on a DCF valuation for the performance metrics of Yukos. This makes
the approach taken by the Tribunal a hybrid approach.

In the Yukos case only the claimants had submitted valuations of Yukos. Their valuations
based on the DCF method and the comparable transactions method were rejected. The Tribu-
nal chose the comparable companies method, which the respondent had corrected for flaws in
the comparison material. The corrections added by the respondent are not available in the fi-
nal award, however, the respondent’s valuation expert – Mr. Dow – wrote an expert report on
the Tribunal’s methodology afterwards. This report also sheds light on how the initial valua-
tion of the claimant was calculated.

According to Mr. Dow claimant had used the following four multiples:

? The ratio of the company’s enterprise value to its EBITDA.
? The ratio of the company’s market price to its earnings.
? The ratio of the company’s enterprise value to its proven oil and gas reserves.
? The ratio of the company’s enterprise value to its oil and gas production.99

The reference companies chosen by the claimants were both Russian and international oil
companies.100 Mr. Dow submitted a report excluding Rosneft because “‘the market perceives
Rosneft as having higher value by virtue of being a majority state-owned and that oil devel-
opment assets are accordingly valued more in the hands of Rosneft than they are in the hands

99 Dow (2014) paragraph 19
100 Yukos Owners v Russia paragraph 1715


of the other Russian companies.'”101 Furthermore, he excluded the major international oil
companies as he believed that these were not comparable and gave equal weights to the re-
maining comparable companies. This had the effect of reducing the value by 32% or about
USD $ 31.7 billion.102 A large part of this reduction had to have been caused by the 70%
weighting of Rosneft. This clearly shows the importance of both the weighting of metrics and
the choice of comparable companies can have on the final value.

After these corrections the weighted average multiple values for the compared companies was
calculated. The average was then multiplied by the metrics derived from the DCF valuation
for Yukos, in order to determine an implied enterprise value for Yukos based on each metric.
The four results calculated were then averaged resulting in a valuation of USD 61.076 billion.
This was the value that the Tribunal held was the value of the company in 2007.103 Data on the Asset

Past information in the form of offers or transactions on the asset being valued can provide
the tribunal/parties’ valuation experts with useful information for the valuation. The prerequi-
sites for applying such information is, as the ad hoc Tribunal in CME v Czech Republic ex-
plains it, that the transaction:

“was negotiated at arms-length and that the valuation of CNTS reflects the valuation of a
willing buyer and a willing seller at the point of time relevant for this arbitration.”104 (my

The underscored parts of the quote highlight some of the necessary requirements for past data
on the asset being a reliable indicator of current value. The first three address the require-
ments for an actual market transaction. If these are not present then the offer is likely affected
by their absence. The requirement of a relevant point of time is important because the longer
the period between the valuation date and the date the information stems from is; the more
likely it is that an intermediate event has affected the value of the asset.

101 Dow (2014) paragraph 25
102 Ibid paragraph 26
103 Yukos Owners v Russia paragraph 1783-1784 – This figure still had to be adjusted to the relevant valuation
dates as the submitted valuation was not of a relevant date of valuation
104 CME v Czech Republic paragraph 560


In the CME v Czech Republic, the Tribunal held that the date of expropriation was 5 August
1999.105 The agreement on the merger transaction was concluded 29 March 1999. This short
time lapse made the Tribunal confident that this data was the fair market value of CNTS.106

Information on a prior offer to purchase the investment was used in Ascom v Kazakhstan. The
date of expropriation was 30 April 2009.107 The Tribunal based its valuation of damages on an
offer of USD 199 million made in September 2008.108

As both cases show, prior information on the value of the asset can be used if the time be-
tween the offer/purchase and the date of valuation is not too long. If such information is used
it might be appropriate to take into account possible events that have affected the market val-
ue of the asset by making adjustments for intermediate events.109 Stock prices

Another method that is based on data on the valuated asset itself is the market capitalization
method. This method takes the stock price of a single share in a company and then multiplies
it by the total amount of stocks being valued. In expropriation cases, this often means all
stocks issued. This method is rather straightforward if the stock price is a reliable indicator of
the price that would be paid for the entire business; however, there are a number of circum-
stances that could lead to discrepancies between market capitalization and fair market value.

Firstly, stocks are often traded on a speculative basis. Buyers of stock might hope to earn
money through increases in stock price rather than dividends, which could lead to the stock
price reflecting other factors than real value.110

Secondly, stock prices are based on smaller transactions where the buyers normally do not
acquire controlling interests in the company. The absence of control means that these stocks
are sold at a discount that reflects the absence of some or all of the powers of control.111 This
helps to explain why, even looking away from potential synergy effects, buyers of entire
companies often offer and pay more than the market capitalization value of the company. The

105 Ibid paragraph 492
106 Ibid paragraph 534
107 Ascom v Kazakhstan paragraphs 1497-1499
108 Ibid paragraphs 1707, 1747-1748
109 Marboe (2009) p. 188
110 Ibid p. 190
111 Kantor (2008) p. 256


reason investors do so is that they believe that there is a value in controlling the company be-
cause they are able to run it better. Generally, the premium will be greater the poorer the man-
agement of the company is. It also depends on the probability of implementing changes in the
management of the company.112 If market capitalization is to be used to value an entire com-
pany, adjustments for the control premium will often be necessary.

Finally, adjustments for lack of marketability can in be appropriate for illiquid shares such as
privately held shares. Liquidity entails the relative ease that an asset can be sold at. Ease of
sale represents a value in its own, and if a share cannot be sold quickly then adjustments to the
share price can be appropriate.113

The possibility for inaccurate valuations due to lack of marketability was accepted in Enron v
Argentina. The ICSID Tribunal agreed with the claimant’s point that when a market is illiquid
or the volume of transactions limited, the market capitalization method might lead to distorted
valuations. The Tribunal noted that this could be mitigated by taking longer time periods into
consideration.114 The Tribunal did not elaborate further on how this ought to be done but pos-
sible ways could be to apply the median or the arithmetic mean of the transactions during the
period. Instead of relying on the market capitalization value for the valuation, the Tribunal
used the market capitalization value “to verify the outcome of the DCF method so as to estab-
lish whether the variables used in the latter reflect reasonable assumptions.”115 Due to the
aforementioned pitfalls of the market capitalization method the Tribunal’s use of the method
seems to have been reasonable given the circumstances in the Argentinian stock market at the

4.2.3 The Asset and Cost Based Approach Introduction

Asset based methods value businesses based on the net value of the business’ assets. Conse-
quentially, this method cannot lead to an award of value that exceeds the sum of the individu-
al assets and therefore disregards the value that can be created by combining assets.117 The
advantage of the method is that it is less speculative than the income-based and market-based
approaches since there is no need to “look into the future” or determine comparable transac-

112 Damodaran (2012) p. 494-495
113 Kantor (2008) p. 258-259
114 Enron v Argentina paragraph 383
115 Ibid paragraphs 424, 437
116 Ibid paragraph 425 – In the Tribunal’s view the Argentinian crisis had led to “wide speculation” at the time
117 Ripinsky (2008) p. 218-219


tions or companies. The method is thus an easier way of establishing value than income-based
methods but does so at the cost of accuracy.

The discussion here primarily focuses on the asset-based method of book value as a tool for
valuation purposes. This section also considers liquidation value and the cost-based methods
of replacement value and historic cost. Book Value

Book value is an accounting concept and refers to the value as per the business’ balance
sheets. The book value of an asset is the book value less depreciation.118 This value is normal-
ly recorded on the balance sheet at acquisition cost. The book value of a business is the net
sum of all assets carried on the business’ balance sheet, which is equal to assets less liabilities
and depreciation.119 The values assets are carried at on the balance sheet are for various rea-
sons rarely measures of fair market value. The balance sheet is a tool used for tax purposes
and financial reporting. In accounting, there is a principle of conservatism that leads to an
understatement of the value of assets and overestimation of liabilities.120 Additionally, record-
ed depreciation does not necessarily reflect genuine economic depreciation, which can lead to
further discrepancy between book value and fair market value.121 Businesses might often also
own assets that have value but are not carried on the balance sheet.

For these reasons, the balance sheet is generally not accepted by valuation experts as a meas-
ure of the real value of the business. It is thus not surprising that tribunals have been reluctant
to use book value as a measure of compensation in regards to businesses.122 The method has
been favored more for valuation of individual assets such as machinery and equipment.123

In order to address the shortcomings of book value the valuators use a method called adjusted
book value. The first step when using this method is to adjust the book values to the market
value. In addition, assets that are not found on the balance sheet but still have a market value,
such as assets that are fully depreciated for accounting purposes but still have economic value,
internally developed intellectual property and goodwill have to be added to the adjusted book

118 Marboe (2009) p. 268
119 World Bank (1992) p. 43
120 Marboe (2009) p. 269
121 Ripinsky (2008) p. 221
122 Marboe (2009) p. 270
123 Ibid p. 273


value. Similarly, liabilities such as contingent liabilities and liabilities for corporate capital
gains taxes on appreciated assets have to be added.124

In Siemens v Argentina the ICSID Tribunal adjusted the book value by removing a tax credit
because the tribunal found that the company would not make any profit that it could use the
tax credit on.125 The Tribunal also removed an entry on the books for risks related to contract
termination as the tribunal awarded the investor damages for consequential losses, which the
Tribunal said would lead to double counting.126 As this case shows the process of adjusting
the book value is necessarily dependent upon the concrete circumstances of the valued busi-
ness and its assets, which requires some judgment by the valuator. Liquidation Value

If a company is not expected to make profit in the future, then the principle of highest and
best use might lead to a valuation based on the liquidation value of the business.127 A compa-
ny’s liquidation value “means the amounts at which individual assets comprising the enter-
prise or the entire assets of the enterprise could be sold under conditions of liquidation to a
willing buyer less any liabilities which the enterprise has to meet.”128 A company in liquida-
tion is under financial pressure and will have to terminate its operations over a limited period
of time. This reduces the chances of finding the buyer that is willing to pay the highest price,
which in turn means that the sale price will carry a discount.129 The size of the discount de-
pends on the urgency of the situation. Orderly liquidation value is appropriate for a business
that has some time to field offers from many willing buyers in order to get the best price for
the asset. Often this means that the seller can sell of individual assets rather than the all the
assets at one time. Distress liquidation value is justified if the company is desperate to liqui-
date its assets. Under such circumstances the business often has to sell all of its assets at once
at a heavily discounted price.130 In recent arbitration practice this method has rarely been

124 Kantor (2008) p. 232
125 Siemens v Argentina paragraph 373
126 Ibid paragraph 374
127 Marboe (2008) p. 287
128 World Bank (1992) p. 42-43
129 Ripinsky (2008) p. 224
130 l.c.
131 Ibid p. 226

27 Replacement Value

Replacement value should generally be used for specific assets and not to value going con-
cerns.132 The replacement value of an asset can be defined as the sum that would have to be
paid for an asset of similar kind, utility and condition.133 If there are no equivalent assets be-
ing traded, the replacement would have to be of the closest equivalent utility to the expropri-
ated asset.134 If the asset is new and unused it is often easy to establish the replacement value
– the purchase of the asset might be even be used as evidence of value.135 However, if the
asset is either highly specialized or has been impaired to a degree due to usage, a tribunal
would have to determine what the cost would be to purchase an asset with equivalent utility if
this method is to be applied. Additionally, economic depreciation and differences in the asset
being valued and the replacement cost asset will have to be factored into the valuation if ap-
plicable. Historic Cost

Historic cost relies on the historic amounts spent by the investor. The method is distinguished
from book value, as it does not consider depreciation.136 Investment arbitration tribunals have
on many occasions relied on the historic amounts spent by the investor as a measure of fair
market value.137 This, however, is rarely an indicator of fair market value as willing buyers
are primarily concerned with what profit a business can generate or the market price for
equivalent assets rather than historical cost. The expenses paid would normally only reflect
fair market value if they are spent on an acquisition of assets under market conditions just
prior to the expropriation, in which case the method is similar to a use of historical data on the
asset for valuation purposes. The effect of awarding the historic cost of the investment is that
the investor is put in a situation where the investment was not made, which is comparable to
the reliance interest in contract law.

Two short examples can be illustrative of why this method often does not reflect fair market
value. In the case of a startup business the value of the business depends on the efficiency of
the investor at setting up the business and the contribution of know-how, technology and

132 Marboe (2009) p. 283
133 Ripinsky (2008) p. 219
134 Marboe (2009) p. 283
135 This would make this method similar to the method discussed in
136 Marboe (2009) p. 278-279
137 See for instance Metalclad v Mexico paragraph 122, Vivendi v Argentina paragraphs 8.3.12-8.3.13, Azurix v
Argentina paragraphs 424-425


management skills. Depending on these factors the fair market value of the business could be
more or less than the money spent. The method also fails to account for value in high risk
undertakings.138 Consider a business that is set up to develop a new treatment for cancer. If
the project fails, a willing buyer would probably be willing to pay close to zero. If it succeeds,
the market value might be many times that of the historic cost.

The tendency in practice has been to use this method for businesses when forward-looking
methods have been considered too speculative.139 Additionally, the popularity can also be
explained to a degree by the fact expenses can be ascertained with higher a higher degree of
certainty than future profits.140

The historic cost method was used in Siemens v Argentina. Kantor has suggested that the in-
vestor in this case might not have received sufficient compensation through the reliance on
historic cost.141 The case concerned a contract to provide a system for immigration control,
personal identification and electoral information system.142 Siemens had completed the sys-
tem engineering stage of the contract and had begun the operations stage.143 Through the en-
gineering stage of the project it is likely that Siemens had contributed with management
know-how and technical expertise that created value in excess of the expenses incurred by
Siemens. As only these amounts were awarded, Siemens probably did not receive the fair
market value of the investment. The case clearly illustrates the potential risks of under- and
overcompensation that relying on historic cost can have.

With regards to the eligibility expenses under this method only a theoretical starting point will
be provided here. First of all the funds sought recovered under the historic cost method have
to be spent for the purpose of the investment. Secondly, the expenses have to have a link to
the investor. Thirdly, the expense has to be reasonably incurred.144

138 Ripinsky (2008) p. 231
139 Ibid p. 227
140 Ibid p. 229
141 Kantor (2008) p. 238
142 Siemens v Argentina paragraph 81
143 Ibid paragraphs 86-91
144 For a detailed analysis on the recoverability of expenses see Ripinsky (2008) p. 266-273


5 The Standard of Compensation for Unlawful Expropriations
5.1 Customary International Law

For many years, the distinction between lawful and unlawful expropriations did not have a
great impact on the final sum awarded. Since modern valuation techniques are able to value
investments on their income-generating potential, lost profit was recoverable under both
standards. This made the main difference the chance of being awarded compensation for con-
sequential loss. The following quote from the partial award in CME v Czech Republic is good
illustration of the previous view on reparation:

“The Respondent is obligated to “wipe out all the consequences” of the Media Council’s un-
lawful acts and omissions… Restitution in kind is not requested by the Claimant… Therefore,
the Respondent is obligated to compensate the Claimant by payment of a sum corresponding
to the value which a restitution in kind would bear. This is the fair market value of Claimant’s
investment as it was before consummation of the Respondent’s breach of the Treaty in August

There are several possible explanations as to why tribunals previously focused exclusively on
the fair market value as of the date of the expropriation as the standard of compensation for
unlawful expropriations. The primary reason is probably that most investments tend not to
increase in value after expropriations. Secondly, party submissions might solely have been
focused on the date of expropriation. Thirdly, tribunals might have taken the narrow approach
to reparation and focused on restoring the status quo ante.146 Fourthly, Sabahi has argued that
it was because investment arbitration tribunals relied on the standard of compensation for law-
ful expropriation and for unlawful takings as well.147

To support his argument Sabahi points to Metalclad v Mexico and Tecmed v Mexico. The
above quoted partial award in CME v Czech Republic clearly shows that this was not always
the case and even in Metalclad v Mexico the ICSID Tribunal also recognized the principle of
reparation and noted that the amount awarded would also be consistent with this standard.148
Most likely, the exclusive focus on valuating the investment as of the date of expropriation
was caused by a combination of these reasons.

145 CME v Czech Republic Partial Award paragraph 618
146 The status quo ante approach is explained below in 5.2
147 Sabahi (2011) p. 95
148 Metalclad v Mexico paragraph 122


The approach to reparation seems to have changed starting with the award in ADC v Hungary
in 2006. The Tribunal in ADC held that the expropriation was unlawful because Hungary had
not fulfilled any of the requirements for a lawful expropriation contained in the Hungary-
Cyprus BIT Article 4. On the standard of compensation, the Tribunal held that:

“The BIT only stipulates the standard of compensation that is payable in the case of a lawful
expropriation, and these cannot be used to determine the issue of damages payable in the
case of an unlawful expropriation since this would be to conflate compensation for a lawful
expropriation with damages for an unlawful expropriation.”149

The Tribunal further held that the BIT did not contain rules on the standard of compensation
for an unlawful expropriation and stated that the standard contained in customary internation-
al law would apply to the question of damages.150 The Tribunal went on by awarding damages
in accordance with the standard of full reparation.

This approach has since been recognized in Siemens v Argentina,151 Vivendi v Argentina,152
Biwater Gauff v Tanzania,153 Siag ; Vecchi v Egypt,154Funnekotter v Zimbabwe,155 Kar-
dassopoulos v Georgia,156 Unglaube v Costa Rica,157 ConocoPhillips v Venezuela,158 and
Yukos Owners v Russia.159 The distinction only proved significant in a few of these cases.
Several of these cases will be dealt with in detail below in 5.4 as the “new” application of the
principle of reparation is largely about the date of valuation.

Only one case to date has explicitly denied the application of the principle of reparation for
unlawful expropriation. In Rumeli Telekom v Kazakhstan the ICSID Tribunal held that the
Chorzów standard only applies to other wrongful acts than unlawful expropriation.160 The
award is somewhat confusing in this respect, as the tribunal seems to have overlooked its

149 ADC v Hungary paragraph 481
150 Ibid paragraph 483
151 Siemens v Argentina paragraph 353
152 Vivendi v Argentina paragraphs 8.2.3-8.2.7
153 Biwater Gauff v Tanzania paragraph 775
154 Siag ; Vecchi v Egypt paragraphs 538-541
155 Funnekotter v Zimbabe paragraph 111
156 Kardassopoulos v Georgia paragraphs 512-515
157 Unglaube v Costa Rica paragraph 306
158 ConocoPhillips v Venezuela paragraph 342-343
159 Yukos Owners v Russia paragraphs 1766-1769
160 Rumeli Telekom v Kazakhstan paragraph 792


finding of a violation of the fair and equitable treatment provision in the treaty.161 The case is
thus perhaps not a good example of how future tribunals will proceed.

5.2 Full Reparation

If a state expropriates an investment that does not comport to the requirements set out in an
investment treaty, the state commits an internationally wrongful act that entails the interna-
tional responsibility of the state.162 According to ILC Article 31, the legal consequence of
international responsibility is to make “full reparation for the injury caused”. In Article 31(2)
injury is defined as “any damage, whether material or moral”. Moral damages are not consid-
ered here, as these are not financially assessable.

The basis of the customary international law on reparation stems from the frequently cited
judgment from the Permanent Court of International Justice (PCIJ) regarding the Factory at
Chorzów. The case concerned a nitrate factory in Upper Selisia that belonged to German na-
tionals, which was transferred to a Polish national through a ministerial decree. The PCIJ held
that the expropriation was unlawful as it was done contrary to Article 7 of the Geneva Con-
vention of 1922.163 On the principle of reparation, the PCIJ stated:

“The essential principle contained in the actual notion of an illegal act – a principle which
seems to be established by international practice and in particular by decisions of arbitral
tribunals – is that reparation must, as far as possible, wipe out all the consequences of the
illegal act and re-establish the situation which would, in all probability, have existed if that
act had not been committed. Restitution in kind, or, if this is not possible, payment of a sum
corresponding to the value which a restitution in kind would bear; the award, if need be, of
damages for loss sustained which would not be covered by restitution in kind or payment in
place of it – such are the principles which should serve to determine the amount of compensa-
tion due for an act contrary to international law.”164

From the quote it is clear that the objective of reparation is to “re-establish the situation which
would, in all probability, have existed if that act had not been committed.” In other words
reparation is a subjective standard, which in investor-state disputes aims at putting the specific
investor in the probable hypothetical situation he or she would have been in had the state not
breached the treaty obligations. This is comparable to the expectation interest in contract law.

161 Ibid paragraph 618
162 See ILC Article 1 – These articles will be used as expressions of customary international law in the following
163 Chorzów Factory paragraph 123
164 Ibid paragraph 125


Reparation can take the form of “restitution, compensation and satisfaction, either singly or in
combination”, cf. ILC Article 34. Satisfaction is of little practical importance in investor-state
disputes. On the relationship between restitution and compensation, it is clear from ILC Arti-
cle 36(1) that the state is only obligated to compensate for damage “not made good by restitu-
tion”, 165 which makes restitution the primary remedy for making reparation. However, resti-
tution is for various reasons rarely claimed in practice,166 which makes damages the de facto
primary remedy for unlawful expropriations.

The obligation to pay damages includes any “financially assessable damage including loss of
profits”.167 The state is therefore liable for any loss caused by the unlawful act. Damages
therefore have a compensatory objective aimed at wiping out all of the financial consequences
of the unlawful act. Typically, this will require an award of the value of investment, lost prof-
it, to the extent that this is not accounted for by the investments value, and any other conse-
quential loss that the unlawful act has caused. So far it seems that tribunals have viewed the
loss suffered by the investor for the taking of the investment as the fair market value of the
investment. This, however, is only the lower limit of the investor’s loss and it could lead to
less than full reparation for the investor if the investment has a special value to the investor.168

The case of CME v Czech Republic seems to be a good illustration of the difference between
fair market value and full reparation. The Tribunal awarded the investor the fair market value
of the investment, which they based on the sum paid by the claimant about 6 months prior to
the expropriation. From the sum paid the Tribunal deducted a sum relating to what was called
“the Zelezny Factor”.169 The Tribunal did not consider that this was related to the business
itself; however, as the Tribunal said that it was awarding “the value which a restitution in kind
would bear”,170 the special value this investment had to the investor meant that the investor’s
loss was not fully wiped out when this deduction was made. So far, this distinction between
subjective value and fair market value has been largely overlooked by tribunals when award-
ing damages.

165 ILC Article 36(1)
166 See Ripinsky (2008) chapter 3.2.3 for an analysis
167 ILC Article 36(2)
168 See chapter 3.2 for more details on special value
169 CME v Czech Republic paragraphs 538-549, 620
170 CME v Czech Republic Partial Award paragraph 618


5.2.1 The Value of Restitution

There are two possible approaches to the obligation of paying damages equal to the value of
restitution. Under the narrow approach the state has to wipe out all the consequences by com-
pensating the value of restitution prior to the unlawful act. From the Commentary to the ILC
Articles, it is clear that the ILC intentionally adopted the narrow definition.171 ILC Article 35
reflects this as the state’s obligation is to “re-establish the situation that existed before the
wrongful act was committed”. In the words of the PCIJ in Chorzów Factory, this would entail
an obligation of paying damages amounting to “the value of the undertaking at the moment of
dispossession, plus interest to the day of payment.”172

The broad approach to restitution entails an obligation to restore a present hypothetical situa-
tion in which the unlawful acts had not been committed. Under this approach the tribunal has
to award damages equal to the difference between the investor’s actual and hypothetical situa-
tion at the date of the award ‘but for’ the state’s unlawful expropriation. This often leads to a
valuation date of the investment being the date of the award. This approach will be referred to
in the following as the Chorzów standard.

Recently, the tribunals that have applied the Chorzów standard have held that the investment
value as of the date of the expropriation and any consequential loss constitute the lower limit
of compensation due for expropriation.173 The consequence of such a view is that states bear
both the risk of a decrease in value as the investment remains on their hand but still have to
pay damages according to the value as of the date of expropriation, and any increase in value
that the investor would have enjoyed ‘but for’ the unlawful act since this is to be awarded as

Only the PCA Tribunal in Yukos Owners v Russia specifically addressed the rationale behind
this lower limit. The Tribunal held that the investor in “absence of the expropriation…could
have sold the asset at an earlier date at its previous higher value.”174 As this case concerned a
direct expropriation through a seizure and auction of the company’s assets, the state did in
fact prevent such a sale of at least the auctioned assets through the expropriation.175

171 ILC (2001) Commentary to Article 35 paragraph (2)
172 Chorzów Factory paragraph 124
173 See e.g. Siemens v Argentina paragraph 352 ; Yukos Owners v Russia paragraph 1768
174 Yukos Owners v Russia paragraph 1768
175 Ibid paragraphs 98-102


However, for indirect expropriations, the investor’s title to the investment is by definition
unaffected, and the investor could, at least formally, have sold it off. Abdala & Spiller have
argued that if the state’s measures do not prevent the investor from divesting, the investor is in
any case to be awarded the difference between the hypothetical value and the actual value as
of the date of the award.176 Their argument is that if economic conditions deteriorate, com-
pensation limited to the value as of the date of the expropriation would give the investor a
windfall, as the state acts are not the cause of this loss. Whether tribunals accept this notion or
not remains to be seen, but other arguments have also been put forward in scholarly writing
for the lower limit approach.

Ripinsky and Williams have noted that “logically, compensation for unlawful expropriation
cannot be lower than that for a lawful one.”177 The writers do not elaborate on why this is
logical, but this would create a disincentive for acting unlawfully, which in turn would make
the rules on lawful expropriation more effective and prevent attempts at opportunistic expro-
priations. This does, however, bring a certain punitive element to the table rather than a com-
pensatory one.

On the other hand, there will rarely be an abundance of actual willing buyers due to the uncer-
tainty surrounding the future of the investment. In cases of indirect expropriation, the lower
limit approach has the benefit of shutting the door on a speculative exercise of determining
whether the investor would have divested or not at some point between the expropriation and
the award.

Whether a tribunal will award damages based on a decreased value of an indirectly expropri-
ated investment is unclear but the expressed views of tribunals so far seems to support that the
lower limit would be chosen. If the compensatory goal of wiping out the consequences is to
be fulfilled, an award based on the lower limit would lead to a windfall for the claimant, and
be contrary to this.

As the application of the broad approach mainly involves the determination of the relevant
date of valuation, a more detailed analysis of investment arbitration practice will be left for
5.4.2 below.

176 Abdala (2008) p. 118
177 Ripinsky (2008) p. 245, see also p. 256


5.3 Heads of Damage

The loss suffered by the investor between the date of expropriation and the date of the award
can be categorized into different heads of damage. These are as follows:

? The value of the investment less any residual value,
? Lost profit,
? Interest, and
? Incidental expenses

Depending on the date of valuation of the investment, different heads of damage are relevant
and can be compensated for.

If the date of valuation is the date of expropriation, the damage not made good by restitution
is the value of the investment to that particular investor, interest loss and any incidental ex-
penses. Incidental expenses are considered at the end of this section. If the investment was
indirectly expropriated any residual value will have to be deducted from the investment value
as this is not lost to the investor. If the value of the investment is calculated using a forward-
looking method, then the loss of profit will already be accounted for by the method, which
would obviate an award of this.178 If not, then lost profit could be awarded, however, this will
probably rarely be the case, as the tribunal will in a sense already have denied that any loss of
profit has occurred between the expropriation and the award through the choice of valuation
method. To the extent that it is necessary to ensure full reparation, the investor will also have
to be awarded interest on the investment value from the date of expropriation using an interest
rate and mode of calculation that wipes out the consequences of the unlawful act.179

If full reparation demands valuing the investment value as of a later date than the date of ex-
propriation, the investment’s value as of that date will have to be awarded. Interest will also
normally have to be awarded on the investment value if the date of valuation is set earlier than
the date of the award. For income-generating investments the tribunal will also have to award
past lost profit that the investor would have earned in the absence of the unlawful act in the
interim period between the date of the unlawful act and the date of valuation. Interest on past
lost profit should also be awarded to the extent that it is warranted to ensure full reparation. In

178 This is dealt with in detail in chapter 4.2
179 See ILC Article 38. The rate and mode of computation remains an unsettled topic in customary international
law and is not dealt with in this thesis, see Ripinsky (2008) p. 364-391 and Marboe (2009) p. 329-377 for a
thorough discussion


such cases the relationship between investment value and lost profit can pose a risk for double
counting. The case of Yukos Owners v Russia seems to be a good illustration of just that.

In Yukos Owners v Russia the Tribunal had to value dividends for a ten year period between
the date of expropriation and the date of the award. The Tribunal considered the parties’ sub-
missions on the issue and made a discretionary downward adjustment that took into account
risks of higher taxes and those associated with the company structure.180 The business was
valued using a comparable companies method coupled with a discounted cash flow valuation,
and the equity value of the investment was index adjusted to the present day value. This tech-
nique failed to account for company value and dividends being inversely related, as hypothet-
ically paid dividends cannot at the same time both contribute to the growth of equity and be in
the hands of the investor earning interest, which was also awarded on the amount. As Russia’s
valuation expert noted in a subsequent report, the Tribunal’s finding was that Yukos’ compa-
ny value mirrored the development of comparable companies, but paid, according to his cal-
culations, dividends of more than three and a half times that of the historical dividends paid
by the comparable companies.181 It is thus possible that the investors were awarded more than
their actual loss, which would be contrary to the compensatory principle underlying the Chor-
zów standard. Therefore, valuators should pay particular care when awarding both past lost
profit and investment value.

In order for reparation to wipe out all the consequences of the unlawful act, investors will
sometimes also have to be awarded incidental expenses caused by the act. This rule is sub-
jected to certain limitations as can be shown by the Commentary to ILC Article 36:

“It is well established that incidental expenses are compensable if they were reasonably in-
curred to repair damage and otherwise mitigate loss arising from the breach.”182

As can be seen by the quote incidental expenses can only be awarded if they were “reasona-
bly incurred” (1) to repair damage or (2) to mitigate loss. In the context of international in-
vestment law incidental expenses will normally be expenses incurred in other to mitigate fur-
ther loss. These expenses will be awarded if they were “reasonably incurred” and there is cau-
sation between the unlawful act and the expenses incurred.

In practice there have been a few cases that have considered such damages but general con-
cepts such as causation, remoteness and reasonableness are still unclear.183 A noteworthy case

180 Yukos Owners v Russia paragraphs 1805-1811
181 Dow (2014) paragraph 75
182 ILC (2001) Commentary to Article 36 paragraph (34)


in this respect is Siemens v Argentina. The investor claimed damages for costs incurred by
maintaining a ‘skeleton operation’, storage costs, training costs and technical support costs,
which were accepted by the Tribunal as justified. 184 The Tribunal did not provide any reasons
as to why these costs were justified; however, the tribunal must have viewed these as reason-
able losses caused by the unlawful act. Such losses will have to be awarded in order to put the
claimant in the position he would have been in ‘but for’ the unlawful act.

5.4 The Date of Valuation for Unlawful Expropriation
5.4.1 Introduction

For unlawful direct expropriations it is normally easy to determine the date of expropriation
since these normally are marked by either a definitive act such as a seizure or a decree that
deprives the investor of the investment. For indirect expropriations it can be a challenging
process of determining the exact date valuation unless the expropriation is the result of a sin-
gle act tantamount to expropriation. The date of valuation for indirect expropriations that do
not happen at one particular moment (creeping expropriations) is dealt with in 5.4.3 below.

The topic that is discussed next here is the justification for using a later valuation date than the
date of expropriation. Logically, if a later valuation date is to be used the investor would have
had to have retained the investment until that date. Additionally, any subsequent increase in
the overall value of the investment and lost profit must have benefitted the investor in order
for there to have been a loss. The following analysis looks at cases dealing with this issue.

5.4.2 Chorzów’s Date of Valuation

The first tribunal that chose a later valuation date was the ICSID Tribunal in the case of ADC
v Hungary. The case concerned the expropriation of an operating agreement for the Budapest
Ferihagy International Airport. The Tribunal held that the state measures constituted an un-
lawful expropriation, as it violated the requirements of public purpose, due process, non-
discrimination and because just compensation had not been paid.185 The value of the agree-
ment had risen markedly after the date of expropriation due to Hungary’s accession to the EU
and a substantial increase in the number of tourists that travelled to Hungary.186 The Tribunal
had to determine which party that would benefit from these events. A valuation based on the
date of expropriation would have considered these events as ex-post information, which

183 Ripinsky (2008) p. 306-307
184 Siemens v Argentina paragraphs 329, 386, 403
185 ADC v Hungary paragraph 476
186 Abdala (2007) p. 6


would have benefited Hungary. Conversely, a valuation of the hypothetical value as of the
date of the award would treat this information as ex-ante information, which would benefit the

The Tribunal held that the applicable standard of compensation was the Chorzów standard
and that the investor was to be awarded compensation based on the market value as of the
date of the award.187 Given the facts this approach seems to have merit. The main driver of the
overall value of the investment was the number passenger travelling through the airport. As
this number had increased and ADC is a professional airport operator, it seems probable that
the investment would have increased in value after the date of expropriation. Had the investor
not been compensated as of the date of the award, the compensatory objective of the Chorzów
standard would not have been achieved. Choosing an earlier valuation date would not have
put the claimant in the hypothetical situation he would have been in ‘but for’ the unlawful

Another interesting case in this respect is Vivendi v Argentina. The ICSID Tribunal held that
there had been an unlawful expropriation and that the relevant standard of compensation was,
as in ADC v Hungary, the Chorzów standard.188 The Tribunal based its valuation on the
amounts invested by claimant and included amounts spent after the date of expropriation.189
Had the Tribunal not considered these amounts, the investor would not have received com-
pensation for losses caused by the unlawful act. Naturally, the goal of wiping out all the con-
sequences would not have been fulfilled if the investor did not receive compensation for all of
the amounts invested. The case can thus be taken into account as a case that bases the valua-
tion on the date of the award. However, it is possible to view it as an example of a valuation
based on the date of expropriation coupled with an award of damages for incidental expenses.
This is mostly a theoretical point.

The arguably most expansive application of the Chorzów standard thus far is the case of Yu-
kos Owners v Russia. The case concerned Russia’s expropriation of the oil company Yukos.
The expropriation was deemed by the PCA Tribunal to be unlawful as it was not carried out
under due process of law and no compensation had been paid.190 Regarding the question of
the relevant valuation date, the Tribunal held that the claimant was entitled to choose between
the date of expropriation and the date of the award.191 In reaching this conclusion, the Tribu-

187 ADC v Hungary paragraphs 497-499, 514, 519
188 Vivendi v Argentina paragraphs 8.2.3-8.2.7
189 Ibid paragraphs 8.3.15-8.3.20
190 Yukos Owners v Russia paragraphs 1583-1585
191 Ibid paragraphs 1763 ; 1769


nal first looked at the Energy Charter Treaty Article 13 and the preparatory works and held
that these did not contain anything of relevance on the date of valuation for damages.192 In-
stead the Tribunal looked to Article 35 and Article 36 of the ILC Articles, and held that Arti-
cle 35 entails an “obligation of restitution that applies as of the date of when a decision is
rendered.”193 The Tribunal went on by explaining that restitution would demand the payment
of the higher value as of the date of the award and the date of the expropriation.194 This meant
that the state bore the risks of both any increase and any decrease in the real value of the in-
vestment from the date of the expropriation until the date of the award.

In reaching the value of the investment ‘but for’ the expropriation, the Tribunal relied on the
RTS Oil ; Gas Index and adjusted the submitted and corrected 2007 valuation backwards to
the date of the expropriation and the date of the award.195 This approach assumed that Yukos’
value would have mirrored the development of the index. However, it is rather unlikely that
Yukos actually would have done so. On the other hand, assuming that a company would have
performed on par with the index might in most circumstances be the most reasonable assump-
tion. Taking into account that there was no indication in the case that the investors actually
would have divested prior to the date of the award, it seems to have been implicitly assumed
that they would have retained the investment until the date of the award. Therefore, the com-
pensatory objective underlying the Chorzów standard necessitated an award of compensation
based on the value as of the date of the award. Had the Tribunal not done so, the award would
not have wiped out the consequences of Russia’s unlawful acts. The total difference in value
between the two dates, after a deducting for contributory fault, was more than USD 33 bil-
lion.196 The award clearly shows how important the choice of valuation date can be.

Two cases can be illustrative of the Chorzów standard applied but found non-significant for
the valuation date. In Rumeli Telekom v Kazakhstan the ICSID Tribunal surprisingly held that
the Chorzów standard only applies to other wrongful acts than unlawful expropriation.197 As
such, the case is not an example of the Chorzów standard; however, the Tribunal noted that in
the present case a choice of standard would not have mattered, as the correct approach in ei-

192 Ibid paragraph 1765
193 Ibid paragraph 1766 – The Tribunal seems to have overlooked that Article 35 adopts the narrow definition of
194 Ibid paragraphs 1767-1768
195 Ibid paragraph 1815
196 Ibid paragraph 1826
197 Rumeli Telekom v Kazakhstan paragraph 792


ther case was to award the value at the time of the expropriation.198 For this reason the case is
dealt with here.

The investment in the Rumeli case had appreciated since the date of expropriation and the
Tribunal would have had to consider this ex-post information if the Chorzów standard had
been applied. Interestingly, the application of this standard would probably have led to the
same result as applying the investment treaty standard, because the Tribunal viewed the posi-
tive developments as attributable to the new owner, which meant that the claimant would not
have been in this improved situation ‘but for’ the unlawful acts.199 Choosing a later valuation
date would not have been necessary in these circumstances for the achievement of the com-
pensatory objective behind the Chorzów standard, as an earlier date of valuation fulfilled this
objective. It can thus be argued that this case is in line with the previously discussed cases.

The second case is Kardassopoulos v Georgia, where the ICSID Tribunal held that Georgia
had violated the due process requirement and unlawfully expropriated the investment by ter-
minating the investor’s concession.200 The Tribunal seems to have implicitly applied the
Chorzów standard as it noted that the Tribunal had to award compensation because restitution
no longer was possible.201 The Tribunal recognized that the principle of full reparation might
require that damages are awarded as of the date of the award if “it is demonstrated that the
Claimants would, but for the taking, have retained their investment.”202

When applying the Chorzów standard to the specifics of the case, the Tribunal held that the
investor would have sold the investment ‘but for’ the unlawful acts of Georgia.203 The evi-
dence the Tribunal relied on for the likely hypothetical outcome is not available in the award,
however, it appears that the investor had been negotiating a sale in the months prior to the
expropriation,204 and that a sale of the shares would have been an “an entirely acceptable out-
come”.205 The Tribunal also relied on industry experts that testified that the consortium the
investment was a part of “would likely have sought to negotiate with the claimant and pur-
chase” the investment.206 As this finding suggests, retention of the investment is a prerequisite
for choosing a later valuation date. If the investment would have been sold around the date of

198 Ibid paragraph 793
199 Ibid paragraphs 807-808
200 Kardassopoulos v Georgia paragraphs 404-405
201 Ibid paragraph 512
202 Ibid paragraph 514
203 Ibid paragraph 515
204 Ibid paragraph 140
205 Ibid paragraph 515
206 Ibid paragraph 516


expropriation there would not have been any loss of value with regards to the investment be-
yond this date. Naturally, the compensatory objective underlying the Chorzów standard would
not demand a later valuation date in such cases.

The final case worth mentioning in this regard is Unglaube v Costa Rica. The case concerned
a lengthy attempt by Costa Rica to expropriate a real estate property. At the date of the award
Costa Rica still had not formally expropriated the investment. The ICSID Tribunal held that
since 22 July 2003 the state had taken actions that effectively deprived the claimant of the
normal rights of ownership.207 These actions were deemed unlawful, as adequate compensa-
tion had not been paid.208 In order to wipe out the consequences of the unlawful act the Tribu-
nal awarded the fair market value of the investment as of 1 January 2006. This was based on
an assumption that the investor would have sold the property six months prior to a market

There is no trace in the award of any proof that the investor actually would have sold the in-
vestment at that point. The only link between the loss and the unlawful act was that the state’s
actions prevented the investor from selling at an earlier point.210 The investor had claimed the
price as of the market peak but the Tribunal refused to award this sum as this would have
credited the investor with “perfect judgment regarding a highly changeable real estate market
as well as perfect market timing”.211 As this shows the Chorzów standard does not warrant
choosing the optimal date of valuation but rather the probable date. Naturally, “losses” that it
is unlikely that the investor would have suffered, should not be compensated. This is perfectly
in line with the previously discussed cases.

5.4.3 Creeping Expropriations – The Date of Expropriation?

Unless an indirect expropriation is the result of a single act tantamount to expropriation, it
often becomes difficult to determine the exact date of the expropriation. Creeping expropria-
tions are the result of a series of measures taken by the state that in sum constitutes an expro-
priation. Setting an exact date as the ‘moment of expropriation’ when the taking does not oc-
cur at one particular instant can prove to be a difficult challenge for tribunals.212

207 Unglaube v Costa Rica paragraph 223
208 Ibid paragraph 305
209 Ibid paragraph 318
210 Ibid paragraph 316
211 Ibid paragraph 317
212 Corey (2012) p. 997


For indirect expropriations the tendency in arbitration practice has been to focus on the date
on which the government measures lead to an “irreversible deprivation” as the date of expro-
priation.213 This moment is normally towards the end of the state’s actions and if the fair mar-
ket value of the investment is assessed as of that date it runs the risk of not compensating the
investor for some of the negative effects of the expropriation.214 Conversely, if the date of
expropriation is set early in the series of events, there is a risk that the state has to compensate
for reduced value caused by lawful measures.215

Reisman and Sloane were among the first to recognize this potential risk for deviation from
the “correct” measure of damages that equating the moment of expropriation with the moment
of valuation poses. They suggested that “the ‘moment of expropriation’ should be distin-
guished from the ‘moment of valuation'” in order to avoid over- or undercompensating the
investor.216 Their article was published in 2004 and it seems that their article was premised on
investment treaties containing the relevant standard of compensation also for unlawful expro-

As noted above indirect expropriations should generally be held to be unlawful acts for which
the state is liable to make reparation. Since the publication of their article, recent case law has
moved away from the view that the investment treaty standard of compensation also applies
for unlawful expropriations. Instead, tribunals now focus on awarding damages that wipes out
the consequences of the unlawful act. Even though tribunals should look away from loss
caused by the unlawful act in theory could do so without adjusting the date of valuation, a
practical way of avoiding under- or overcompensating the investor is to delink the ‘moment of
expropriation’ from the date of valuation in cases concerning creeping expropriation.

The case of Kardassopoulos v Georgia seems to be a good example of just that. The case
concerned an expropriation of the two investors’ rights in an oil pipeline. The government
expropriated the investment through two decrees. The expropriation was deemed to have been
effected by the second decree on 20 February 1996. The Tribunal separated the date of valua-
tion and the date of expropriation, and held that 10 November 1995 – the day before the first
decree – was the relevant date of valuation.218 The Tribunal explained that the reason behind
this choice of an earlier date was “to ensure full reparation and to avoid any diminution of

213 Marboe (2009) p. 135
214 Corey (2012) p. 998
215 Ibid p. 997
216 Reisman (2004) p. 148
217 Ibid p. 133
218 Kardassopoulos v Costa Rica paragraph 517


value attributable to the State’s conduct leading up to the expropriation.”219 This approach is
in line with the Chorzów standard because it ensures that all of the investor’s loss is wiped

The previously mentioned case of Unglaube v Costa Rica is also illustrative of the new ap-
proach to valuation dates. Costa Rica had from July 2003 until the award tried and failed to
expropriate the investor’s real estate property through several resolutions. During that period
the real estate prices in the region had risen sharply until 2006, and then stabilized before
dropping substantially.220 The Tribunal considered that “the determination of a ‘date of ex-
propriation’ (and its use as ‘the date of valuation’ of the 75-Meter Strip) presents a compli-
cated and unsatisfactory set of choices.” The Tribunal saw “no rational basis for selecting” a
date of expropriation.221 Instead, the tribunal assumed that the investor would have sold the
investment just prior to the peak.222

As these two cases show the potential risk posed by linking the date of valuation with the date
of expropriation should not be an issue any longer if tribunals apply the Chorzów standard.
This allows tribunals to choose an appropriate date of valuation ensures that all of the loss
caused by the unlawful act is wiped out. Determining the valuation date will necessarily de-
pend on the concrete factual circumstances of the particular case as has been shown by these

6 Final Remarks

As was shown through this thesis calculating damages and compensation for expropriation
can be a challenging exercise that requires the arbitrators to make numerous assumptions on
future events and comparable companies, transactions and assets. Even the choice of an ap-
propriate valuation method can be a difficult exercise. The complexity associated with calcu-
lating damages and compensation can necessitate the use of valuation experts for the tribu-
nals. As party submission often vary, it is necessary that arbitrators have an understanding of
the underlying basics of the valuation methods, and seek guidance when complex questions

Through the ‘new wind’ of awarding an amount higher than the fair market value of the ex-
propriated property, the rules on lawful expropriation might become more effective as states

219 l.c.
220 Unglaube v Costa Rica paragraph 313
221 Ibid paragraph 316 – with reference to Reisman (2004) p. 130-133
222 Ibid paragraph 318


now bear both the upside risk and downside risk. In the cases where the Chorzów standard
was applied significantly, the subsequent events did in fact develop favorably, which meant
that the investor’s loss was higher than the fair market value as of the date of valuation and as
such the results seems justified. As was shown in the discussion on Yukos v Russia, the appli-
cation of the Chorzów standard can be a challenging process when the investment is a busi-
ness that is destroyed. Even if it is clear that the investor’s loss would have been greater than
the sum as of the date of expropriation, measuring the loss of something that no longer exists
can for large assets over longer periods of time lead tribunals to make assumptions that the
investment would have followed the market.

As the differences between the sum awarded as compensation or damages can be substantial it
can be expected that the question of legality will receive much attention in future arbitral pro-


7 References
7.1 Bibliography

Abdala (2007) Abdala, Manuela A., Pablo T. Spiller and Sebastian Zuccon,
“Chorzów’s Compensation Standard As Applied In ADC v.
Hungary”, Transnational Dispute Management 3 (2007),
http://www.transnational-dispute- Accessed 20 April

Abdala (2008) Abdala, Manuela A., and Pablo T. Spiller, “Chorzów’s Standard
Rejuvenated – Assessing Damages in Investment Treaty Arbitra-
tions”, Journal of International Arbitration, Volume 25 issue 1
(2008), p. 103-120 (Sitert fra HeinOnline)

Commission (2007) Commission, Jeffery P., “Precedent in Investment Treaty Arbi-
tration – A Citation Analysis of a Developing Jurisprudence,
Journal of International Arbitration, Volume 24 Issue 2 (2007),
p. 129-158.
Accessed 20 April 2015

Corey (2012) Corey, Shain, “But Is Tt Just? The Inability for Current Adjudi-
catory Standards to Provide “Just Compensation” for Creeping
Expropriations”, Fordham Law Review, Volume 81 No. 2
(2012), p. 973-1012.
ber.pdf Accessed 20 April 2015

Damodaran (2006a) Damodaran, Aswath, “Chapter 7 – Relative Valuation: First
Principles.” I: Damodaran on Valuation: Security Analysis for
Investment and Corporate Finance, 2nd ed., 2006.
7.pdf) Accessed 20 April 2015

Damodaran (2006b) Damodaran, Aswath, “Chapter 15 – The Value of Synergy.” I:
Damodaran on Valuation: Security Analysis for Investment and
Corporate Finance, 2nd ed., 2006.
15.pdf) Accessed 20 April 2015

Damodaran (2008) Damodaran, Aswath, “What is the riskfree rate? A Search for
the Basic Building Block”, Stern School of Business working
paper (2008).
Accessed 20 April 2015.

Damodaran (2012) Damodaran, Aswath, “The value of control: implications for
control premiums, minority discounts and voting share differen-


tials”, NYU Journal of Law & Business (2012), Vol. 8(2), p 487-
content/uploads/JLBv8n2_5_Damodaran.pdf Accessed 20
April 2015

Dolzer (2012) Dolzer, Rudolf, and Christoph Schreuer, Principles of Interna-
tional Investment Law, 2nd ed., 2012

Dow (2014) Dow, James, Expert Report of Professor James Dow, 8 Novem-
ber 2014, (
documents/italaw4155.pdf) Accessed 20 April 2015

Fortier (2004) Fortier, L. Yves, and Stephen L. Drymer, “Indirect Expropria-
tion in the Law of International Investment: I know It When I
See It, Or Caveat Investor”, ICSID Review (2004) Volume
19(2), p. 293-327
Accessed 20 April 2015

ILC (2001) International Law Commission, Draft Articles on Responsibility
of States for Internationally Wrongful Acts, with commentaries,
2001, (A/56/10).
_6_2001.pdf) Accessed 20 April 2015

Kantor (2008) Kantor, Mark, Valuation for Arbitration: Compensation Stand-
ards, Valuation Methods and Expert Evidence, 1st ed., 2008

Marboe (2007) Marboe, Irmgard, “Compensation and Damages in International
Law – The Limits of ‘Fair Market Value'”, The Journal of
World Investment & Trade (2006), Vol. 7(5), p. 723-759.
http://www.transnational-dispute- Accessed 20 April

Marboe (2009) Marboe, Irmgard, Calculation of Compensation and Damages in
International Investment Law, 2009

Newcombe (2009) Newcombe, Andrew, and Lluis Paradell, Law and Practice of
Investment Treaties, 2009.

Reinisch (2008) Reinisch, August, “Legality of Expropriations”, I: Standards of
Investment Protection, August Reinisch (ed.), 2008, p. 171-204.
df Accessed 20 April 2015

Reisman (2004) Reisman, Michael W., and Robert D. Sloane, “Indirect Expro-
priation and its Valuation on the BIT Generation”, Faculty
Scholarship Series (2004), paper 1002.

47 Accessed
20 April 2015

Ripinsky (2008) Ripinsky, Sergey, and Kevin Williams, Damages in Interna-
tional Investment Law, 2008

Sabahi (2011) Sabahi, Borzu, Compensation and Restitution in Investor-State
Arbitration, 1. Utg, 2011. (Sitert fra Oxford Scholarship Online)

Sornarajah (2010) Sornarajah, Muthucumaraswamy, The International Law on
Foreign Investment, 3rd ed. 2010.

Stern (2008) Stern, Brigitte, “In Search of the Frontiers of Indirect Expropria-
tion”, I: Contemporary Issues in International Arbitration and
Mediation: The Fordham Papers 2007, Arthur W. Rovine (red.),
2008, p. 29-52 (Sitert fra Brill Online)

UNCTAD (2007) UNCTAD, Bilateral Investment Treaties 1995-2006: Trends in
Investment Rulemaking, (UNCTAD/ITE/IIA/2006/5), 31 Janu-
ary 2007. ( Ac-
cessed 20 April 2015

UNCTAD (2011) UNCTAD, Expropriation: A Sequel,
(UNCTAD/DIAE/IA/2011/7), 29 November 2011.
( Ac-
cessed 20 April 2015

UNCTAD (2014) UNCTAD, World Investment Report 2014,
(UNCTAD/WIR/2014 )24 June 2014.
( Ac-
cessed 20 April 2015

Wells (2003) Wells, Lou, “Double Dipping in Arbitration Awards? An Econ-
omist Questions Damages Awarded to Karaha Bodas Company
in Indonesia”, 19 Arbitration Interational (2003), p. 471-482.
(Sitert fra HeinOnline)

World Bank (1992) World Bank, Legal Framework for the Treatment of Foreign
Investment, Report No. 11415 Volume II: Guidelines, World
Bank Group, 25 September 1992, (http://www-


pdf), Accessed 20 April 2015

Wälde (2007) Wälde, Thomas W., and Borzu Sabahi, “Compensation, Damag-
es and Valuation in International Investment Law”, Transna-
tional Dispute Management (2007), Vol. 4(6), p. 1-64.
http://www.transnational-dispute- Accessed 20 April


7.2 Table of Cases

All cases are available from unless specified otherwise.

ADC v Hungary ADC Affiliate Limited and ADC & ADMC Management
Limited v. The Republic of Hungary, ICSID Case No.
ARB/03/16, Award of the Tribunal of 2 October 2006
Asian Agricultural Products v
Sri Lanka
Asian Agricultural Products LTD v Republic of Sri
Lanka, ICSID Case No. ARB/87/3, Final Award of 27
June 1990
Ascom v Kazakhstan Anatolie Stati, Gabriel Stati, Ascom Group S.A., Terra
Raf Trans Traiding LTD. v Republic of Kazakhstan,
Stockholm Chamber of Commerce, Case No. 1:14-cv-
00175-ABJ, Award of 19 December 2013
Azurix v Argentina Azurix Corp. v The Argentina Republic, ICSID Case No.
ARB/01/12, Award of 14 July 2006
Biwater Gauff v Tanzania Biwater Gauff LTD. V United Republic of Tanzania, IC-
SID Case No. ARB/05/22, Award of 24 July 2008
CME v Czech Republic Partial
CME Czech Republic B.V. v The Czech Republic, UN-
CITRAL Arbitration Proceedings, Partial award of 13
September 2001
CME v Czech Republic CME Czech Republic B.V. v The Czech Republic, UN-
CITRAL Arbitration Proceedings, Final award of 14
March 2003
ConocoPhillips v Venezuela ConocoPhillips Petrozuata B.V., ConocoPhillips Hamaca
B.V., ConocoPhillips Gulf of Paria B.V., and ConocoPhil-
lips v Bolivarian Republic of Venezuela, ICSID Case No.
ARB/07/30, Decision on Jurisdiction and Merits of 3 Sep-
tember 2013
Enron v Argentina Enron Corporation and Ponderosa Assets, L.P., ICSID
Case No. ARB/01/3, Award of 22 May 2007
Funnekotter v Zimbabwe Bernardus Henricus Funnekotter and Others v Republic of
Zimbabwe, ICSID Case No. ARB/05/6, Award of 22 Ap-
ril 2009
Gold Reserve Inc v Venezuela Gold Reserve Inc v Bolivarian Republic of Venezuela,
ICSID Case No. ARB(AF)/09/1, Award of 22 September
Kardassopoulos v Greece Ioannis Kardassopoulos v The Republic of Georgia, IC-


SID Case No. ARB/05/18 Ron Fuchs v The Republic of
Georgia, ICSID Case No. ARB/07/15, Award of 3 March
Metalclad v Mexico Metalclad Corporation v The United Mexican States, IC-
SID Case No. ARB(AF)/97/1, Award of 30 August 2000
Mobil v Venezuela Mobil Corporation, Venezuela Holdings, B.V., Mobil
Cerro Negro Holding, Ltd., Mobil Venezolana de Petróle-
os Holdings, Inc., Mobil Cerro Negro, Ltd., and Mobil
Venezolana de Petróleos, Inc. v. Bolivarian Republic of
Venezuela, ICSID Case No. ARB/07/28, AWard of 9 Oc-
tober 2014
Rumeli Telekom v Kazakhstan Rumeli Telekom A.S. and Telsim Mobil Telekomikasyon
Hizmetleri A.S. v Republic of Kazaksthan, ICSID Case
No. ARB/05/16, Award of 29 July 2008
Saipem v Bangladesh Saipem S. p. A. v The Peoples Republic of Bangladesh,
ICSID Case No. ARB/05/7, Award of 30 June 2009
Santa Elena v Costa Rica Compañía del Desarrollo de Santa Elena, S.A. v The Re-
public of Costa Rica, ICSID Case No. ARB/96/1, Final
Award of 17 February 2000
Siemens v Argentina Siemens A.G. v. The Argentine Republic, ICSID Case
No. ARB/02/8, Award of 6 February 2007
SPP v Egypt Southern Pacific Properties (Middle East) Limited v Arab
Republic of Egypt, ICSID Case No. ARB/84/3, Award of
20 May 1992
Starrett Housing v Iran Starrett Housing Corporations, Starrett Systems Inc, and
Starrett Housing International Inc v The Governement of
the Islamic Republic of Iran, 16 Iran-US CTR (1987) 112,
Award No. 314-24-1, Final Award of 14 August 1987.
This award was only mention as it was referenced by Ri-
pinsky and is not available at
Tecmed v Mexico Técnicas Medioambientales Tecmed, S.A. v. The United
Mexican States, ICSID Case No. ARB (AF)/00/2, Award
of 30 August 2000
Tidewater v Venezuela Tidewater Investment SRL, Tidewater Caribe, C.A., v.
The Bolivarian Republic of Venezuela, ICSID Case No.
ARB/10/5, Award of 13 March 2015
Unglaube v Costa Rica Marion Unglaube v Republic of Costa Rica, ICSID Case
No. ARB/08/01, Reinhard Unglaube v Republic of Costa
Rica, ICSID Case No. ARB/09/20, Award of 16 May


Vivendi v Argentina Compañía de Aguas del Aconquija S.A. and Vivendi Uni-
versal S.A. v Argentine Republic, ICSID Case No.
ARB/97/3, Award of 20 August 2007
Yukos owners v Russia Hulley Enterprises Limited v The Russian Federation,
PCA Case No. AA 226, Yukos Universal Limited v The
Russian Federation, PCA Case No. AA 227, Veteran Pe-
troleum v The Russian Federation, PCA Case No. AA
228, Final award of 18 July 2014

7.3 Table of Treaties

Investment treaties:

Australia-Hong Kong BIT (1993)
Greece-South Africa BIT (2001)
Italy-Bosnia and Herzegovina BIT (2000)
Italy-Tanzania BIT (2003)
NAFTA Convention (1994)
Norway-Russia BIT (1995)

Other treaties:

ICSID Convention (1965)
New York Convention (1956)