Abaclat and Others v. Argentina
Dissenting Opinion, Georges Abi-Saab
Table of Contents Paras.
I - Introductory Remarks 1-19
II - The Fourth Preliminary Objection 20-33
III - The Ninth Preliminary Objection 34-117
A - Investment under the ICSID Convention in General 38-61
B - Applying the BIT and the Convention to the Alleged Investment 62-72
C - The Territorial Link 73-119
- 1) The Legal Criteria 78-87
- 2) The Material Criteria for Determining the Situs of the 88-117
“Investment”
IV - The First Preliminary Objection : Mass Claims, Collective Actions
and the Consent of the Respondent 120-273
A - A Plea to Jurisdiction or to Admissibility? 122-127
B - The Legal Nature and Specificities of Mass Claims Actions and their
Bearing on the “Consent to Arbitrate” 128-145
C - Judicial Innovation, Collective Proceedings and “Consent to
Arbitrate” 146-153
D - The Interpretation of Silence 154-175
E - Mass Claims Processes in International Law 176-189
F - The Limits of Procedural Improvisation 194-262
1) On Filling Gaps 210-220
2) The Scope of the Proposed “Adaptations” and their Incidence
on the Rights of the Parties 221-244
3) Balancing of Interests Redux 245-262
G - Epilogue on Policy Considerations 263-274
V – Conclusion 275-277
1
Abaclat and Others
(Case formerly known as Giovanna A Beccara and Others)
(Claimants)
and
The Argentine Republic
(Respondent)
Decision on Jurisdiction and Admissibility
Dissenting Opinion
Professor Georges Abi-Saab
2
I - Introductory Remarks
1 - In spite of my great respect and esteem for my co-arbitrators, I am compelled to part ways
with their majority award for imperative reasons of principle, pertaining to substance and
method.
2 - Substantively, I consider that this Tribunal faces two glaringly insuperable obstacles that
prevent it from taking jurisdiction over the collective mass claims action submitted to it,
namely:
a - that the sovereign debt instruments (whether we call them “bonds”, “obligaciones”,
“security entitlements” or otherwise) that are at the basis of these claims, do not constitute a
“protected investment” under the ICSID Convention and the Argentina-Italy Treaty (see
Section III below), and
b - that this ad hoc ICSID International Arbitral Tribunal has no jurisdiction under the
ICSID Convention and the BIT over the present collective mass claims action, absent
Argentina’s consent, nor does it have the power to elaborate new procedures to handle such
an action (See Section IV below).
3 - Saying that does not mean that I accept the rest of the majority award. Indeed, one of the
basic reasons of my unease with this excessively long award, is its style of turning around the
3
main issues and drowning them into an ocean of minutia and elaborated details, rather than
confronting them frontally and treating them thoroughly. This is why I shall not, in this
opinion, enumerate and analyze all the “objects” of my objections, but - while reserving my
position on the other preliminary objections of Argentina - shall limit my remarks and
analysis mainly to the two fundamental points mentioned above, which overshadow all the
others; not, however, before addressing my second level of disagreement.
4 - The second ground of my disagreement with the majority award, is situated at a further
remove, at the level of legal method and reasoning, as well as the understanding of some
basic principles, rules and concepts of international law, such as jurisdiction and
admissibility, and the principles and methods of interpretation ; that which I shall illustrate by
using the example of the handling by the majority award of the fourth preliminary objection,
based on the requirement of prior consultations and resort to national courts for 18 months
before going to arbitration. (See Section II below).
5 - However, before turning to these objections, I have to articulate some basic premises and
concepts from which my analysis proceeds.
6 - i) This is technically an international ad hoc arbitral Tribunal. It is “ad hoc” because
specially established to hear one specific case, suit or action. It is “international” because it is
rooted in two layers of international treaties : the ICSID Convention and the Bilateral
Investment Treaty between Italy and Argentina. As such it functions under, and is governed
by international law, and has to be clearly distinguished from “international” commercial
4
arbitration tribunals, such as those established within the framework of the International
Chamber of Commerce, which function under national law and ultimate national judicial
control. It is necessary to keep this distinction in mind, as solutions may vary at these two
distinct levels; and what is sometimes permissible in municipal law, is not necessarily
acceptable in international law.
7 - ii) In international law, all tribunals - not only arbitral, but even judicial - are tribunals of
attributed, hence limited jurisdiction (juridictions d’attribution). There is no tribunal or
system of tribunals of plenary or general jurisdiction (juridiction de droit commun) that
covers all cases and subjects, barring exceptions falling under - i.e. attributed to - the
jurisdiction of a specialized tribunal. This is because, in the absence of a centralized power
on the international level that exercises the judicial function through a judicial system
empowered from above (or rather incarnating the judicial power as part of the centralized
power), all international adjudicatory bodies are empowered from below, being based on the
consent and agreement of the subjects (i.e. the litigants, les justiciables) themselves (with the
very limited exception of tribunals created by international organizations in the exercise of
their powers under their constitutive treaties, which are also ultimately based on the consent
of the subjects that concluded or adhered to these constitutive treaties).
8 - iii) This is the reason why, the fundamental principle and basic rule in international
adjudication, is that of the consensual basis of jurisdiction. It also explains the prominent
place of questions of jurisdiction both in the jurisprudence and in the writings on
international adjudication. It explains as well the widely shared perception that the first task
5
of an international tribunal is to ascertain its jurisdiction; and the great care international
tribunals take in establishing from the outset, the existence and limits of the consent of the
parties before them, on which their jurisdiction is founded.
9 - iv) Jurisdiction as a concept in international law, partakes of its generic meaning in the
general theory of law; but is further specified and particularized in function of the
fundamental principle of the consensual basis of international adjudication.
10 - The term “jurisdiction” (from the latin “jurisdictio”, literally to “pronounce” or
“enunciate the law”, dire le droit), when used in the judicial or adjudicative context, denotes
“the legal power of an organ to decide cases (in general) or a case (in particular) by
application of law”. In other words, it is the empowerment to exercise the judicial function.
Thus, jurisdiction refers first and foremost to a legal power to exercise a certain type of
activity, the judicial function1
.
1
See different definitions of the term (putting emphasis on power), in Black’s Law
Dictionary (6th ed., 1990) : Jurisdiction : A term of comprehensive import embracing every kind
of judicial action...It is the power of the court to decide a matter in controversy and presupposes
the existence of a duly constituted court with control over the subject matter and the parties...
Jurisdiction defines the powers of courts to inquire into facts, apply the law, make decisions, and
declare judgment... The legal right by which judges exercise their authority...Power and
authority of a court to hear and determine a judicial proceeding; and a power to render particular
judgment in question...” (emphasis added).
11 - In English, however, the term “jurisdiction” is also used to designate the “ambit” or
“sphere” within or over which this power is exercised, though in other languages, such as
French and Spanish, another term is used to designate the “ambit”, which is “competence”.
Some writers and arbitral decisions seem to be oblivious of the primary and more
6
fundamental meaning of jurisdiction as power, limiting the term to the ambit of this power.
Jurisdiction as ambit is defined as a four-dimensional sphere, by its extent and limits in time
(ratione temporis) and space (ratione loci), and over subjects (ratione personae) and objects
or subjects-matters (ratione materiae).
12 - v) In international law, because of its consensual basis, jurisdiction as an ambit is
analyzed and scrutinized at two different levels, where adjudication is not intended for one
case only, but takes place within an institutional setting, either of a standing organ (such as
the ICJ) or a framework within which ad hoc tribunals are established (such as the PCA and
the ICSID) :
a) - “general jurisdiction” which defines the objective range and outer limits of the
ambit for all cases, according to the constitutive instrument of the organ (e.g. the ICJ Statute),
or the framework convention (e.g. the ICSID Convention;.
b) - “special jurisdiction” which defines the subjective range and limits of the ambit of
jurisdiction of the organ in a particular case, according to the specific jurisdictional title
bearing the consent of the parties, on the basis of which the case is brought before the organ.
13 - The consent of the parties cannot go beyond the objective limits of the general
jurisdiction of the organ, but can restrict its jurisdiction further within these objective limits,
by the parties subjecting their consent to additional limits, conditions or reservations. These
can relate to any of the four dimensions of the ambit. But they can also relate to the powers
of the organ as such (the primary meaning of jurisdiction), as long as this does not touch the
hard core requirements of the judicial function, for example by excluding from its powers the
7
indication of provisional measures.
14 - In case of ICSID arbitration, the jurisdictional title for general jurisdiction is the
Convention itself. For “special jurisdiction”, the title is usually two-tiered : the BIT between
the two States and the consent in writing of the two parties to the dispute.
15 - Thus, for an ICSID ad hoc tribunal, a triple-layered consent is usually required. For such
a Tribunal to take jurisdiction over a case, it has to ascertain :
a - that the case falls within the jurisdiction of the ICSID as defined by the ICSID
Convention and related instruments (Rules of Procedure, etc.);
b - that it falls within the bounds defined by the BIT between the State party to the
ICSID arbitration and the national State of the investor (which usually - but not necessarily also
carries the consent of the two states to arbitrate);
c - and that the case is covered by the specific written consent of the parties to the
dispute (which can be for the State party, the BIT or another unilateral act bearing its
consent. But it can also be a compromissory clause in a contract between the parties; and for
the investor it can be the request of arbitration itself).
The Tribunal has to ascertain the existence and scope of the consent of the parties within the
limits prescribed by them in each of the above-mentioned instruments.
16 - vii) The requirement to ascertain the existence and scope of consent, while strict and
exacting in international law, does not mean the restrictive interpretation of the jurisdictional
title (the old theory of interpretation in favour of sovereignty, as far as the State party is
8
concerned). But it does not mean either its extensive interpretation beyond “the horizon of
foreseeability”; i.e. extending jurisdiction to what the party or parties could not have
foreseen at the time the treaty was concluded or consent was given. Interpretation is limited
here to the establishment of the reality and extent of the consent of the party or parties, at the
time it was given.
17 - viii) As with jurisdiction, the concept of admissibility in international law partakes of its
generic meaning in the general theory of law, but is further particularized in function of the
specificities of international adjudication, including its consensual basis.
18 - Generically, the admissibility conditions relate to the claim, and whether it is ripe and
capable of being examined judicially, as well as to the claimant, and whether he or she is
legally empowered to bring the claim to court.
19 - It is true that these conditions were adjusted to the specificities of international law, and
some additional ones were developed, by way of custom. But none of these conditions has
anything to do with the determination of the scope of consent whether to the general or the
special jurisdiction of tribunals, contrary to what the majority award asserts (para. 496); these
questions being jurisdictional by essence.
II - The Fourth Preliminary Objection
9
20 - This objection is based on the non-fulfilment by the Claimants of the requirements of
prior consultations (or negotiations) and, in case they fail to settle the dispute, resort to local
courts for 18 months before going to arbitration. These two conditions are stipulated in
paragraphs 1 to 3 of article 8 of the BIT dealing with the settlement of disputes, which
constitutes the second layer of the jurisdictional title in the present case (para. 15 above), and
the preliminary objection based on them is presented by the Respondent as a plea to
jurisdiction.
21 - The majority award rejects this legal characterization on the basis of an
incomprehensible distinction between “conditioning its [a State party’s] consent to ICSID
jurisdiction to the fulfilment of a pre-condition” - which, I presume, would constitute a
jurisdictional limitation according to the majority award - and “conditioning the effective
implementation of such consent, i.e. the possibility to resort to ICSID arbitration, to the
fulfilment of such a pre-condition” (para. 494), which can lead “only - if at all - to a lack of
admissibility of the claim” (para. 496).
22 - Abstruseness and the absence of any suggested criterion to effect this distinction apart, I
can see no legal or logical objective reason to distinguish between the different conditions,
reservations or limitations that parties attach to their consent in the jurisdictional title, be it
one or a multi-layered one, as in the present case, and be these reservations or conditions
related to the general or the special jurisdiction of the tribunal.
23 - It is true that, under general international law, the two requirements in question are
10
considered as conditions of admissibility. But when such conditions are included in the
jurisdictional title, they condition, like any other reservation inserted in the jurisdictional
title, the consent of the party or parties making them, to submit to the jurisdiction of the
judicial or arbitral organ, and limit by that much the exercise by the organ of its jurisdiction.
In other words, in this case these conditions become conventionally jurisdictional, in addition
to being admissibility conditions by their legal nature.
24 - But the fatal blow to the majority opinion’s distinction comes from much closer quarters.
For the non-adepts at legal analysis, it suffices to read the second sentence of Article 26 of
the ICSID Convention (“A contracting State may require the exhaustion of local
administrative or judicial remedies as a condition of its consent to arbitration under this
Convention”), to discover that such a stipulated requirement conditions the consent to
arbitrate and not merely “the effective implementation of such consent”, whatever that
means. As such, it is a limit to jurisdiction, and the plea based on it is a plea to jurisdiction
and not to admissibility.
25 - Be that as it may, and regardless of the classification of the objection as a plea to
jurisdiction or to admissibility, the result of the non-fulfilment of the requirements should
have been the same, the dismissal of the case. But what is even more striking in the treatment
by the majority award of these two conditions (under admissibility), is the short shrift it gives
them and the cavalier way it disposes of them.
26 - Thus, the requirement of prior consultations, in spite of the peremptory “shall” of article
11
8/1 of the BIT2
, and the clear text of this paragraph, the majority award practically strikes it
out of the treaty, together with the paragraph that stipulates it, on the ground, inter alia, that:
“In the view of the Tribunal, the consultation requirement set forth in Article 8(1) BIT
is not to be considered of a mandatory nature but as the expression of the good will of
the Parties to try firstly to settle any dispute in an amicable way” (para. 564).
And this is because the provision contains the proviso “to the extent possible”. But this
proviso is inherent to all “obligations of means” or of “best efforts” (obligations de moyen),
which would be, according to the majority award’s reasoning, devoid of any legal effect and
can be simply ignored, rendering the provisions stipulating them totally useless (with no effet
utile)!
2
Article 8 :
1. Any dispute relating to investments that may arise between an investor from one of the
Contracting Parties and the other Party, with respect to matters regulated by this Agreement,
shall be, insofar as possible, resolved through amicable consultations between the parties to the
dispute” (emphasis added).
27 - More aggravating still, is the manner in which the majority award treated the second of
these requirements, that of resorting to Argentinean tribunals for 18 months before initiating
arbitration. Here, the majority award admits that this requirement should have been, but was
not complied with by the Claimants. But it immediately adds, at para. 579 : “At the same
time, the wording of Article 8 BIT itself does not suffice to draw specific conclusions with
regard to the consequence of non compliance with the order established by Article 8", and, at
para. 580, that “Claimants’ disregard of the...requirement is in itself not yet sufficient to
preclude Claimants from resorting to arbitration”.
12
28 - These last two sentences are very odd indeed. For no instrument, laying down
jurisdictional limits or admissibility conditions, specifies the legal consequences of non
observance of these limits or non fulfilment of these conditions. These consequences are
embedded in the very legal classification of these as jurisdictional limits or admissibility
conditions. According to the general rules of law and rules of general international law, non
compliance begets the inevitable legal sanction of dismissing the case, as falling outside the
jurisdiction of the tribunal or as inadmissible. Only if the parties want to waive or vary this
sanction do they address the effect of non compliance in the instrument. Otherwise, in case
of silence, it is the rules of general international law that apply.
29 - Instead, and in spite of its admission that the requirement is obligatory and that it was
not fulfiled, the majority award decides to proceed to a “weighing of the interests of the
Parties” (para. 582), it concludes that “it would be unfair to deprive the investor of its right to
resort to arbitration based on the mere disregard of the 18 months litigation requirement”
(para. 583); adding : “[T]his conclusion derives more from a weighing of the specific
interests at stake than from the application of the general principle of futility” (para. 584);
before rejecting the preliminary objection.
30 - Here again, the majority award takes the liberty of striking out a clear conventional
requirement, on the basis of its purely subjective judgment; that which calls for three
remarks.
13
31 - First, the “balancing of interests”, that the majority award arrogates to itself the right to
go behind the text (or across the mirror) in order to operate, has already been done, at the
appropriate legislative level, by the parties themselves, introducing the requirement in the
Treaty, but limiting it to 18 months. And it was reflected or registered in the clear text. It is
not open to the Tribunal, which is now called upon to apply it, to put it into question in the
name of a rebalancing of interests more to its liking, at the expense of the one agreed by the
parties.
32 - Secondly, both the language (“it would be unfair”) and the stance of the argument, are
those of a tribunal judging ex aequo et bono, which is prohibited by the rules of general
international law and the ICSID Convention, except by agreement of the parties (Article
42/3); absent which, the judgment is ultra vires.
33 - Thirdly, the presumed unfairness derives, according to the majority award, from the fact
that “the deprivation of the investor’s right to resort to arbitration would, in effect, deprive
him of an important and efficient dispute settlement means” (para. 583). Elsewhere in the
award, it is even contended that the dismissal of the case would deprive the securities holders
of any effective means of dispute settlement. But I beg to differ. For, apart from the
Argentinean courts that the Claimants had not even tried, they could have gone to the chosen
fora in the security entitlement instruments, which are neutral fora of major financial centers;
or to Italian courts, against the banks that sold them the security entitlements in the bonds,
and whose ambiguous role, when later grouped as Task Force Argentina, is strongly
contested by the Respondent as having diverted the bondholders from those normal fora for
14
this type of disputes, towards the extraordinary forum of ICSID arbitration, in order to
circumvent their own responsibility.
III - The Ninth Preliminary Objection
34 - This objection is based on the contention that the security entitlements in the
Argentinean bonds held by the Claimants, do not constitute a covered or protected
“investment” under the ICSID Convention and the Bilateral Investment Treaty between Italy
and Argentina; and that, in consequence, the Tribunal has no jurisdiction ratione materiae
over the case.
35 - This is, to my knowledge, the first ICSID case that involves a sovereign debt bond (or a
security entitlement therein), totally unrelated to a specific project or economic operation or
enterprise in the borrowing State. It raises a major issue as to the jurisdiction of ICSID
tribunals over a vast new field, with incalculable economic and political ramifications.
36 - The assessment of this preliminary objection requires defining what is a protected or
covered “investment” under the ICSID Convention and the BIT, and whether the financial
security entitlements in question fall within this definition.
37 - In what follows, I shall proceed to address these questions by successive
approximations, indicating, where appropriate, the points of my divergence from the majority
15
award.
Investment under the ICSID Convention in General
38 - Starting with the ICSID Convention, the task is to identify the purport of the term
“investment” in paragraph 1 of article 25, which defines the scope of jurisdiction of the
Centre as extending “to any legal dispute arising directly out of an investment...”. It is to be
noted, however, that beyond this mention, the Convention does not provide expressly a
definition of what is meant by “investment” for its purposes.
39 - This led to one opinion according to which, since the ICSID Convention does not define
“investment”, this task is totally abdicated or left to the BIT (or another jurisdictional title, on
the basis of which the case is brought before an ICSID arbitral tribunal). In consequence,
according to this opinion, there is no need for a “double-barreled” test the claim has to pass,
under both the ICSID Convention and the BIT, in order to qualify as an investment. It
suffices to qualify only under the BIT (or another jurisdictional title reflecting the consent of
the Parties).
40 - This opinion does not withstand scrutiny. That the ICSID Convention does not provide
an express definition of investment does not automatically imply that the definition is totally
left to the BITs. This is because words have an intrinsic meaning, hence a limited and
limiting one, however large and vague it may be (although there is always a penumbra
around the limits which provides the margin of interpretation). Without limits, words would
16
be meaningless, because undistinguishable from one another. The intrinsic meaning of a
word, which is its “ordinary” meaning, is further specified by the way it is used and the
context in which it is used; and if it figures in a treaty, by the object and purpose of the treaty.
41 - A clear distinction has to be made between the use of the term “investment” in the
financial context and in the ICSID context. In financial markets, “investment” covers the
acquisition of any kind of assets such as deposit accounts, debt and equity securities, credit
default swaps and derivatives.
42 - This over-wide financial concept of investment is fundamentally different from the
concept of investment envisaged when drafting the ICSID Convention and followed in its
context since. And in the words of the Annulment Committee in Mitchell v. Democratic
Republic of Congo :
“It is obvious that the special and privileged arrangements established by the
Washington Convention can be applied only to the type of investment which the
Contracting States to the Convention envisaged”3
.
43 - The Travaux Préparatoires of the ICSID Convention reveal the limiting or restricting
intent behind the introduction and further qualification of the term “investment” in the
provision that became article 25. Thus, in response to concerns about the over-broad
jurisdiction ratione materiae in the Working Paper which simply referred to “disputes”, the
3
Decision on the Application for Annulment of the Award of 9 February 2004 (1 Nov.
2006). ICSID case No ARB/99/7, para. 25.
17
Preliminary Draft introduced the term “investment” as a qualifier before “disputes”, in
defining the ambit of jurisdiction (“any existing or future investment dispute of a legal
nature”)4
.
4
History of the ICSID Convention, vol. I, pp. 110-112.
18
44 - This was further narrowed in the First Draft : “all legal disputes...arising out of or in
connection with any investment”5
. This formula was in turn strongly attacked as too wide,
particularly the clause “in connection with”, which was struck out by a vote of 26 to 8, while
adding the further qualifier “directly” into the Second (revised) Draft that became the actual
formula : “any dispute arising directly out of an investment”6
.
45 - The purpose for using the term “investment” in article 25/1 was thus to set objective
outer-limits to the types of disputes that can be treated within the ICSID7
; and that is enough
to refute the opinion referred to in paragraph 38 above. It is true that these outer-limits bound
a vast ambit, to the point of not being clearly visible to some. But they exist all the same.
5
Ibid. p. 116.
6
Ibid. p. 118.
7
The exclusion of certain types of transactions, starting with ordinary commercial
transactions, i.e. the purchase of goods and services in the normal course of business, was clearly
on the mind of the drafters of the Convention, particularly its main architect, A. Broches (ibid.
Vol II, p. 83). It was frequently reiterated by ICSID Tribunals, e.g. SGS v. Pakistan, (Decision
on Jurisdiction, 6 August 2003, para. 133) : “...investment...embod[ies] certain core meaning
which distinguishes it from ‘an ordinary commercial transaction’ such as a simple, stand alone,
sale of goods or services”; Phoenix Action v. Czech Republic (Award 15 April 2009, para. 96) :
“There is nothing like a total discretion, even if the definition [of investment] developed by
ICSID case law is quite broad and encompassing. There are indeed some basic criteria and
parties are not free to decide in BITS that anything - like a sale of goods or a dowry for example
- is an investment”.
19
46 - Differently put, the term “investment” in article 25/1 of the ICSID Convention, whilst
flexible enough, is not infinitely elastic8
. It leaves much latitude and a wide margin of
interpretation and further specification to States in their BITs; but not to the point of
rendering it totally vacuous, without any legal effect. In other words, the term has a hard-core
that cannot be waived even by agreement of States parties to a BIT9
.
47 - To identify this hard-core, one has to look further into the general context of the ICSID
Convention and the circumstances surrounding its elaboration as well as to its object and
purpose.
48 - It is most significant that the ICSID Convention was elaborated and the Centre
established on the initiative and within the framework of the International Bank for
Reconstruction and Development; an institution which concentrated its activities since the
early sixties almost exclusively to the second facet of its mandate according to its title, i.e.
the “development” of the less developed countries.
49 - The Preamble of the Convention clearly reveals its “developmental” object and purpose,
8
See Michael Waibel, “Opening Pandora’s Box : Sovereign Bonds in International
Arbitration” 101 AJIL (2007, No. 4), p. 711 at p. 722.
9
See Phoenix Action v. Czech Republic, supra note 7. The Tribunal adds : “...BITs which
are bilateral arrangements between two States parties, cannot contradict the definition of the
ICSID Convention. In other words, they can confirm the ICSID notion or restrict it, but they
cannot expand it in order to have access to ICSID. A definition included in a BIT being based on
a text agreed between two States cannot set aside the definition of the ICSID Convention, which
is a multilateral agreement”.
20
as a means of encouraging “international cooperation for development, and the role of private
international investment therein”; and this by making available “facilities for international
conciliation or arbitration”, besides the national courts of host States, to settle potential
disputes arising from such investments between private foreign investors and those States.
50 - The investment that the Convention seeks to encourage by providing it with an
international procedural guarantee is that which contributes to the economic development of
the host country, i.e. to the expansion of its productive capacity, a contribution that
presupposes a commitment to this task not only of economic resources, but also in terms of
duration in time and the taking of risk, with the expectation of reaping profits and/or
revenue in return.
51 - The different attempts by ICSID tribunals to capture the essence of this hard-core
concept of investment in article 25 and formulate it in terms of criteria or conditions particularly
the classical Salini criteria10
; but also in Phoenix Action v. Czech Republic11
; as
well as in the recent Romak S.A. v. Uzbekistan case12
- all turn around the same idea, with
minor variations.
10
Salini Construttori v. Morocco, Decision on Jurisdiction, 23 July 2001, ICSID case No.
ARB/09/4, para. 55-56).
11
Supra, note 7, at para. 82.
12
PCA case No. AA 280, Award of 26 Nov. 2009, para. 180, 207, quoted in the majority
award, para. 370.
21
52 - The fact that the Salini criteria or the other similar formulations are not expressly laid
down in the ICSID Convention does not mean that they do not articulate, perhaps
imperfectly, an obligatory requirement of article 25, or that this requirement has no
constraining effect if States parties to the ICSID Convention chose to ignore it in their BITs,
as the muddled analysis of this issue in the majority award seems at one point to suggest
(para. 364, also 351). But the award hedges the issue by contending that in any case, the
investment in question satisfies the requirement, at least as formulated in the Romak v.
Uzbekistan case (para.371); a contention subject to verification below.
53 - What type of investment corresponds to this requirement ? It is worth recalling that the
main object and purpose of the Convention was to encourage the flow of private foreign
investment into developing countries, by making available an additional international
procedural facility or guarantee that counter-balances the host State’s regulatory authority
over investment and economic activities in its territory; in other words, by providing a neutral
international forum in case of dispute, as an alternative to submitting to the jurisdiction of the
tribunals of the host State.
54 - Such facility is basically needed by private foreign direct investment, for it is this type of
investment that once it is carried out in the host country, for example by building factories or
establishing enterprises, falls under the imperium of the host State in terms of legislation and
adjudication.
55 - Direct foreign investment is then the “ideal type” of investment (in the Weberian sense
22
of the term) for ICSID purposes. But does it exhaust the ambit of ICSID jurisdiction ratione
materiae ? And if not, how far can an alleged investment depart from the ideal type and still
be covered by the Convention, i.e. and still be considered as falling within the objective
outer- limits set by article 25 ?
56 - This question arises particularly in relation to “portfolio investments” and other financial
negotiable products traded in the financial markets, which cover a wide spectrum ranging
from standardized instruments such as shares, bonds and loans to structured and derivative
products, such as hedges (of currencies, oil, etc.) and credit default swaps.
57 - Such widely dispersed off-the shelf financial products, with their high velocity of
circulation and their remoteness, the same as their holders, from the State in whose territory
the investment is supposed to take place (being traded within seconds at the touch of a button
in capital markets, with no involvement or knowledge of the borrowing country, nor passage
through the territory or the legal system of that State), seem at first blush to be worlds apart
from the direct foreign investment model, which is usually long negotiated and extensively
embedded in the legal environment of the host State.
58 - This raises acutely the question of the conformity of these financial products with the
requirements of article 25 and evokes a kind of informal presumption that, because of their
intrinsic characteristics described above, they are excluded per se, i.e. automatically, from the
protected or covered investments under the Convention.
23
59 - All the same, the empirical evidence reveals that whilst the vast majority of transactions
considered by ICSID tribunals to constitute covered or protected investments under the
Convention consisted of direct foreign investment, some of these tribunals have also
considered negotiable financial products, in certain circumstances, as covered investments.
Some BITs also include financial products, or some of them, in their illustrative list of
investment.
60 - This practice goes against the presumption that financial products are disqualified per se
and automatically as protected investment under the Convention; that which led a learned
commentator to observe that “[p]ortfolio investment(s)...are not excluded as a rule”13
.
61 - “Not [being] excluded as a rule” simply means that they are not excluded automatically,
per se. But it also means that they can be excluded in certain cases. In other words, it means
that the exclusion or inclusion cannot be operated automatically, on the basis of the mere
classification of the instruments per se, but has to be done on a case by case basis, depending
on other factors and criteria that lie beyond the mere categorization of the instrument in
question; factors and criteria that I examine below in relation to the alleged investment in the
present case.
Applying the BIT and the Convention to the Alleged Investment
13
G. Sacerdoti “Bilateral Treaties and Multilateral Instruments on Investment
Protection”, Recueil des Cours, Hague Academy of International Law, vol. 269 (1997), p. 251, at
p. 307.
24
62 - Article 1 (“Definitions”) of the BIT between Italy and Argentina provides in its first
paragraph (in the unofficial English translation of Argentina) :
“For the purposes of this Agreement :
1. The term ‘investment’ shall mean, in conformity with the legal system of the
receiving Country and independently from the legal form adopted or from any other
connected legal system, any contribution or asset invested or reinvested by physical or
juridical persons of one Contracting Party in the territory of the other, in accordance
with the laws and regulations of the latter.
Within this general context, the following are specifically, but not exclusively,
considered to be investments :
.....
c. debt obligations, public or private securities or any other right to benefits or
services with an economic value, as well as capitalized income;”
63 - The Claimants’ English translation of lit. c. is somewhat different :
“c. bonds, private or public financial instruments or any other right to performances or
services having economic value, including capitalized revenues;”
64 - The Tribunal’s English translation is as follows :
“c. obligations, private or public titles or any other right to performances of services
having economic value, including capitalized revenues;”
25
65 - Having examined the wording of this provision and the nature of the alleged investment
in casu, the majority award concludes that : “Claimants purchase of security entitlements in
Argentinean bonds constitute a contribution which qualifies as ‘investment’ per se under
Article 1(1) of the BIT” (para. 371).
66 - Before reaching this conclusion, the majority award dispenses with the requirements of
Article 25 of the ICSID Convention, by examining three alternative theories or views (which
it paradoxically counts as two) (para.368). The first is to apply the double-barreled test; but if
the alleged investment or contribution satisfies the BIT test, but not the Salini test, the former
test should prevail, i.e. discarding the requirements of Article 25, according to Salini (paras.
363-368). The second view is simply to discard the double-barreled test and require only a
BIT test, discarding formally and totally any requirements by the ICSID Convention in this
regard! (para. 369). The third view, pays lip service to Article 25 of the Convention by
maintaining that, if need be, the term “investment” per se in Article 1/1 of the BIT can be
analyzed in the same manner as in Article 25 of the Convention, as interpreted in the Romak
v.Uzbekistan case (para. 371). From which statement the majority award presumably expects
the reader to infer and assume that the alleged investment satisfies the criteria of Romak,
without any further proof or demonstration on the part or the award, which jumps directly to
the conclusion that “Claimants’ purchase of security entitlements in Argentinean bonds
constitute a contribution which qualifies as investment per se under Article 1(1) of the BIT”
(para. 371).
67 - Be that as it may, this conclusion or finding of the majority award calls for verification
26
on several levels, namely, (by ascending order of importance) :
i - Do the securities in question really correspond to Article 1/1/c of the BIT ?
ii - Could they be considered as protected investments in Argentina in view of their
legal remoteness, the same as their present holders, from Argentina, and the bonds it initially
issued ?
iii - Assuming that both preceding questions are answered in the affirmative, do these
securities fulfil the substantive conditions of the Convention and the BIT relating to the
contribution, particularly the territorial link ?
68 - (i) Concerning the first question : whether the alleged investment is covered by Article
1/1/c of the BIT or not, there is no need to revisit here the controversy over the correct
English translation of “obligaciones”, and whether it is “bonds” (Claimants), or “debt
obligations” (Respondent), or merely “obligations” (Tribunal). On this question I agree with
the majority analysis that Article 1/1/c covers financial instruments, and that its language is
large enough to encompass the security entitlements in the Argentinean bonds.
69 - (ii) The problem raised by the second question is much more complicated than transpires
from the simplistic argument by which the majority award brushes it asides, simply affirming
that the bonds and the security entitlements “are part of one and the same economic
operation” (para. 359).
70 - The award fails clearly to distinguish between purchases on the primary market,
involving the issuer (Argentina) and the first buyers of the issue (the underwriters), and the
27
secondary market, where previously issued securities are traded, without any involvement of
the sovereign debtor. An ICSID Tribunal cannot look only at the economics of a transaction,
without taking into consideration its legal framework and structure, in order to determine
whether it qualifies as a protected “investment” or not.
71 - Even from a purely economic point of view (not to mention the legal perspective), the
passage from the primary to the secondary market is neither automatic nor certain. The
underwriters of the bonds bear the risk of not attracting enough demand, which is one reason
why they receive an underwriting spread. Moreover, they may want to keep bonds as part of
their portfolio. Similarly, and also from an exclusively economic point of view, the position
of Argentina in these two markets is totally different. In the primary market, Argentina
received the proceeds of the initial issuance of the bonds from the underwriters. By contrast,
the flow of funds triggered by transactions in the secondary market is exclusively between
the buyer and seller of the security entitlements, its volume depending on the conditions
prevailing in that market, and bearing no visible relation to the lump-sum received by
Argentina from the underwriters at issuance14
.
14
See, Waibel, “Opening Pandora’s Box”, supra note 8. p.727.
72 - The Tribunal is thus bound to look at the circumstances of the individual purchases of
security entitlements, and their traceability to - i.e. the strength or tenuousness of their legal
nexus with - Argentina, before it can decide whether the dispute over each of them “aris[es]
directly out of an investment”; in other words whether they satisfy the requirements of a
28
covered or protected “investment” under the Convention and the BIT, on which hinges its
jurisdiction ratione materiae.
The Territorial Link
73 - Assuming, arguendo, that the answer to the first two questions calling for verification is
in the affirmative (an assumption that I refute regarding the second question, as just
explained above), there remains the third, and more problematic question, namely : do these
security entitlements satisfy the other, substantive, conditions of the Convention and the BIT,
particularly that of a territorial nexus or link with Argentina ?
74 - A territorial link or nexus is inherent in the concept of “investment” in article 25 of the
ICSID Convention. The whole idea behind the Convention was to encourage the flow of
private foreign investment to developing countries by offering an international guarantee in
the form of an alternative neutral adjudication of disputes arising out of such investment in
the territory of the host States, typically subject to its laws and courts.
75 - In addition, the Argentina-Italy BIT unambiguously requires a territorial link between
the alleged investment and the host country. The Preamble of the Argentina-Italy BIT
expresses the State parties’ intentions to “create favourable conditions for investments by
nationals and companies of either State in the territory of the other” - a standard formulation
found in many BITs. The definition of investment in Article 1 refers to “any contribution or
asset invested or reinvested by physical or juridical persons of one Contracting Party in the
territory of the other”, and Article 1/4 carefully delimits the BIT’s territorial ambit.
29
76 - The BIT’s substantive provisions are also conditioned on a territorial link, including :
- Article 2 (protection and promotion of investments) calls on each contracting party
to “encourage the making of investments in its territory by investors of the other Contracting
Party”;
- Article 2/2 (full protection and security) obliges each party to refrain from “adopting
unjustified or discriminatory measures that impair the management, maintenance, enjoyment,
transformation, cessation and liquidation of investments made in its territory by investors of
the other Contracting Party”;
- Article 3 (national and most favoured treatment) likewise contains the qualifier
“within its own territory”;
- Article 4 (compensation for damage or loss) refers to “the Contracting Party in
whose territory the investment was made”.
77 - This requisite territorial link is clearly absent in the present case. The alleged investment,
the security entitlements, are not located in Argentina by applying either the legal or the
material criteria for determining such location
78 - 1) The legal criteria : the financial securities instruments that constitute the alleged
investment, i.e. the security entitlements in Argentinean bonds, have been sold in
international financial markets, outside Argentina, with choice of law and forum selection
clauses subjecting them to laws and fora foreign to Argentina. In fact, they were
intentionally situated outside Argentina and out of reach of its laws and tribunals. There is no
way then to say (and no legal basis for saying) that they are legally located in Argentina.
30
79 - The majority award contests this clear conclusion on two grounds :
a) The first is that the above conclusion “would mean that forum selection clauses
determine the place where contractual performance is supposed to take place”, whilst they
are merely “of a procedural nature aiming to determine the place of settlement of a dispute
relating to contractual performance”, but “have nothing to do with the place where a party is
supposed to perform its obligations” (para. 379).
80 - The majority award appears thus to suggest that the determinative factor is the place of
performance. If this criterion is applied, there can be no doubt that the place of performance
under the securities instruments at issue is invariably outside Argentina, given the use of
fiscal agents, paying agents, depositories and places of payment all situated outside
Argentina. Factors other than the place of performance also point to the location of the
securities in question outside Argentina.
81 - Among these other factors, I consider a clause selecting courts external to the host State,
as a prominent one in determining where the security instruments, and the underlying right to
repayment of the debt, are located15
.
15
Bayview v. Mexico, Preliminary Objections, Award of 19 June 2007, paras 101-102.
31
82 - To answer the question whether the securities in question are located in Argentinean
territory, this Tribunal needs to determine the situs of the debt - i.e. the alleged investment using
a systematic approach consistent with well founded and established precedents and
drawing on private international law rules. The foreign governing law and foreign forum,
while not determinative by themselves, are important factors in determining situs. Other
factors include the currency of payment, the place of payment and the residence of the
intermediaries. On any of these criteria, the transactions at issue here were deliberately
structured so as to have their situs outside Argentina16
.
83 - b) - The other argument by which the majority award tries to counter the obvious
conclusion that the alleged investment is legally located outside Argentina, is that forum
selection clauses (ignoring the other numerous connecting factors and criteria mentioned
above) are contractual stipulations and as such they are irrelevant for locating the
“investment” for the purposes of a “treaty claim”, which is based on other rights and
obligations derived from the BIT (para. 379).
16
See, Zachary Douglas, The International Law of Investment Claims, (Cambridge
University Press, 2009), p. 171; Michael Waibel, Sovereign Defaults before International Courts
and Tribunals (Cambridge University Press, 2011), p. 239.
84 - But this facile escape route (échappatoire) is to no avail in the instant case. A treaty
claim is necessarily based on a right that has been allegedly violated; here, the debt that was
not repaid. If this right is created by contract, it is the contract that governs its legal existence
and the modalities of this existence, including the location of this right (and its reciprocal
32
obligation). And the right in the present case has been purposefully located outside
Argentina.
85 - The treaty claim cannot by-pass or circumvent this right, or change its modalities of
existence, including its situs according to its legal title - the contract and the applicable law
under which it exists - as this right, in its fixed legal configuration, serves as the legal basis
underlying the treaty claim.
86 - A treaty claim cannot allege a violation of a right, while ignoring the specific legal
configuration of this right. The treaty may define the kinds of violation, but the underlying
right subject to these violations, is defined only by its legal title, and the applicable law that
governs its existence.
87 - In conclusion, the alleged investment, i.e. the financial securities and their underlying
debt, by all legally recognized connecting factors and criteria, have their situs outside
Argentina; a treaty claim based on the nonpayment of this debt, or other violations of the
right it represents, cannot change the legal configuration of this right, including its situs
outside Argentina. In consequence, the alleged investment does not constitute an “investment
in the territory of Argentina”, and thus falls outside the ambit of jurisdiction of the Tribunal
ratione materiae.
88 - 2) The material criteria for determining the situs of the “investment” : As was just
explained, the majority award discards the contractual clauses and arrangements stipulated in
33
the securities instruments as relevant factors for the location of the situs of the investment
under the BIT and the ICSID Convention, whilst proposing other material or economic
criteria. It is noteworthy in this respect that the section under which this question is addressed
in the majority award is entitled “Made in Argentina” (i.e. investment “made in Argentina”)
and not “Made in the territory of Argentina”, following the language of article 1/1 of the
BIT, an omission symptomatic of the result-oriented style of the whole award.
89 - The majority award starts by an affirmation that “the determination of the place of the
investment... depends on the nature of such investment” (para. 374); to reach the conclusion
that an “investment of purely financial nature” need not be “linked to a specific economic
enterprise or operation taking place in the territory of the host State” (para. 375), basing itself
on arguments it draws from a) Article 1/1/c of the BIT, and b) the nature of the investment.
These arguments are examined in what follows.
90 - a) Arguments drawn from article 1/1/c of the BIT : The majority award states that :
i - Article 1/1/c “designates financial instruments as an express kind of investment
covered by the BIT”, i.e. as a covered investment per se, or as such; in consequence,
ii - “it would be contrary to the BIT’s wording and aim to attach a further condition to
the protection of financial investment instruments” (para. 375)
91 - First of all, it is necessary to explain what the qualification “ per se” (frequently used by
the award) exactly means in this context. As was already explained (paras. 56-61 above),
financial instruments, because of the intrinsic characteristics that appear to be incompatible
34
with the type of investment envisaged in article 25 of the ICSID Convention, suffered from a
kind of informal presumption that they are disqualified per se (or as such) from constituting a
covered investment under the ICSID Convention (para. 58 above). But once a BIT includes
these financial instruments in its enumeration of investments covered by the treaty, this
presumption is obviously rebutted. However, if these financial instruments are not excluded
per se (or automatically) that does not mean that they are included per se, in the sense of
being automatically considered as legally self-sufficient to constitute a protected investment.
We have here to recall the distinction (dating back to Roman law) between the instrumentum
and the negotium, the instrument that registers and vehicles a legal act or transaction and the
legal transaction itself17
. Article 1/1/c recognizes the financial securities per se as valid
instrumenta for investment. But the investment itself, as a legal act or transaction that is
attested and vehicled by these instruments, has to satisfy the substantive requirements for
investment, in article 25 of the ICSID Convention as well as in the BIT, if any.
92 - Leaving aside the requirements of article 25 of the Convention discussed above (para.
47-52), does the BIT impose any other substantive conditions ? And is linking the investment
“to a specific economic enterprise or operation taking place in the territory of the host State”
one of them?
17
This distinction between the instrumentum and the negotium was clearly on the mind
of the drafters of Article 1/1 of the BIT. Thus, in their enumeration, where they considered that
confusion between the two may arise, they expressly made the distinction e.g., under (a)
“...including, to the extent they may be used as investments, security interests on the property of
third Party; under (d) “credits directly related to an investment...” (emphasis added).
93 - Here, the majority award interjects its second argument drawn from article 1/1/c, namely
35
that, given the designation, in this provision, of financial instruments as covered investments,
“it would be contrary to the BIT’s wording and aim to attach a further condition to the
protection of financial investment instruments” (para. 375).
94 -This argument disregards a major fact. Of course, one cannot introduce a “further
condition” into the Treaty. However, two such conditions are already writ large therein. It is
true that they do not figure in Article 1/1/c itself. But all one has to do is to raise his sight a
few lines above this provision to the chapeau of Article 1/1, which applies to all the types and
vehicles of investment enumerated under it, including (c), and which defines investment inter
alia as that made by nationals “of one Contracting Party in the territory of the other [Party],
in accordance with the laws and regulations of the latter” (emphasis added).
95 - The ordinary meaning of the emphasized words could not be clearer. And how can the
fact that the investment has been made or realized in the territory of the host country be
proved or demonstrated, except by tracing it to a specific project, enterprise or activity in that
territory that corresponds to the economic meaning of investment in article 25 of the ICSID
Convention (i.e. that it contributes to the expansion of the country’s productive capacity)?
96 - b) Arguments drawn from the nature of the investment : Against this clear “territorial
imperative” of the Convention and the BIT, the majority award develops a second line of
argument or defense, consisting of four mutually supporting affirmations, namely :
i - “The determination of the place of investment depends.... on the nature of such
investment”; and for “an investment of a purely financial nature, the relevant criteria cannot
36
be the same as those applying to an investment consisting of business operations and/or
involving manpower and property”;
ii - For “investments of a purely financial nature the relevant criteria should be where
and/or for the benefit of whom the funds are ultimately used, and not the place where the
funds were paid out or transferred”;
iii - “This is also the view taken by other arbitral tribunals”;
iv - “Thus, the relevant question is where the invested funds ultimately made available
to the Host State [sic], and did they support the latter’s economic development” (para. 374).
97 - None of these affirmations withstands scrutiny.
i - There is absolutely no basis in the ICSID Convention for drawing such a distinction
between “investments of a purely financial nature” and other types of investment in this or
other respects. Investment under ICSID is a unified category corresponding to the economic
meaning of investment. The question that is raised about financial instruments under ICSID
concern their high velocity of circulation which distends their link, as well as that of their
last holders, with the ultimate economic enterprise or activity that materializes the investment
in the economic, i.e. ICSID, sense of the term, within the territory of the host State. The
question is not about the necessity of locating this enterprise or activity in the territory of the
host State.
98 - Indeed, the most persuasive legal justification for admitting financial instruments as
protected investments, in spite of their stark contrast to the ideal type of investment under
ICSID, the direct foreign investment (see para 55 above), is that these investments finance,
37
be it at several removes, specific economic projects, enterprises or activities which, had they
been undertaken directly by these foreign financial investors, would have constituted foreign
direct investment; which of course can only take place within the territory of the host State.
99 - ii - The governing texts, particularly article 1/1 of the BIT, are clear as to the relevant
criterion for locating the situs of the investment. It is “where” the investment is made or
realized, and not “for the benefit of whom”.
100 - It is ironic that the majority award cites the SGS v. Republic of the Philippines decision
in support of its position; an award enunciating that : “In accordance with normal principles
of treaty interpretation, investments made outside the territory of the Respondent State,
however beneficial to it, would not be covered by the BIT”18
; a dictum repudiating in no
uncertain terms the criterion of “for the benefit of whom” proffered by the majority award.
18
SGS Société Générale de Surveillance SA v. Republic of the Philippines, ICSID Case
No. ARB/02/6 , Decision on Jurisdiction and Admissibility, 24 January 2004, para. 99; see also
Canadian Cattleman for Fair Trade v. United States, NAFTA, Award on Jurisdiction, 28
January 2008, para. 127.
38
101 - iii - Indeed, the cases cited by the majority award do not lend it as much support as it
contends and are all quite distinguishable from the present case. The majority opinion cites
three awards19
. Only one of them involves financial instruments, namely Fedax v. Venezuela.
And it is the only one of the three (as well, to my knowledge, of all other ICSID arbitral
decisions) that carries a dictum suggesting the over-loose, ultimate beneficiary test which is
followed by the majority award in the present case (but Fedax is distinguishable from the
present case on facts as explained below).
102 - The Fedax decision was widely criticized both by other tribunals and in the literature,
including by Professor Dolzer, one of the leading legal experts of the Claimants in the
present case, who commented in an article that “the Fedax decision is not without ambiguity
in its construction of ‘investment’”; and in relation to the “territorial dimension of foreign
investments”, that :
“...the Fedax decision (paragraph 41)...assumed that the absence of a physical transfer
of funds will not stand in the way of the existence of an ‘investment’. Without
explaining the rationale for this view in any detail, the Fedax tribunal considered
apparently that it is sufficient that the funds made available by the investor are utilized
by the host country as the beneficiary of the transaction so as to finance its various
19
In note 147 to paragraph 374. The three awards are : Fedax N.V. v. Republic of
Venezuela (ICSID Case No. ARB/96/3), Decision of the Tribunal on Objections to Jurisdiction
of 11 July 1997, para. 41; SGS Société Générale de Surveillance S.A. v. Islamic Republic of
Pakistan (ICSID Case No. ARB/01/13), Decision of the Tribunal on Objections to Jurisdiction
of 6 August 2003, paras. 136-140; SGS v. Republic of the Philippines, para. 111.
39
governmental needs”20
.
20
Rudolf Dolzer, “The Notion of Investment in Recent Practice”, in Charmovitz et al.
(eds) Law in the Service of Human Dignity : Essays in Honour of Florentino Feliciano (CUP,
2005), p. 261 at pp. 269, 272.
40
103 - Similarly, the SGS v. The Republic of the Philippines21
decision - that the majority
award cites in support of its position! - apart from its flat repudiation of the Fedax test in the
dictum reproduced above (para. 100), spared no effort to distance itself from that case, not
only by distinguishing it, but also by pointing to certain factors that severely limit its general
relevance (if any). Thus, after noting that “[t]he Tribunal in the Fedax case gave a very broad
definition of territoriality”, it observes in a footnote to that sentence that :
“The territorial requirement in the BIT [Venezuela-Netherlands] was ...less
categorical. It referred to the protection in its territory of investments of nationals of
the other Contracting Party...and this only in a clause dealing with entry, not in a
general clause defining the scope of the Treaty as a whole”.
104 - Just after the general remark about the “very broad definition of territoriality”,the
Tribunal proceeds to distinguish Fedax :
“...but the focus of the decision was whether the endorsee of a promissory note issued
with respect to an investment had itself made an investment, and whether the dispute
over non-payment of the note arise ‘directly’ out of an investment within the meaning
of Article 25(1) of the ICSID Convention”22
.
21
Cited in note 18 above.
22
Ibid. at para. 110 and footnote 41 in that decision. The Tribunal hints that it would
have gone further in the critical analysis (or rather criticism) of Fedax, were it not that “Counsel
for the Respondent declined to argue that Fedax was in this respect wrongly decided, but at any
rate the circumstances of the Fedax case were very different from the present” (Ibid.).
41
105 - Fedax is an isolated case. It is an outlier. But I need not expand further on whether it
was correctly decided or not, as it is clearly distinguishable in this respect (of territorial link)
from the present case on facts; primarily by two significant facts. The promissory notes at
issue in Fedax were governed by the host State’s law. They were not free-standing or
unhinged from any specific project or economic activity in the host country, as they were
initially given in exchange for the provision of specific services in Venezuela. The issue in
that case was whether an endorsement of six promissory notes outside Venezuela severed
their link with the underlying transaction, and not about the necessity of the existence of such
underlying transaction23
.
106 - The other two cases cited by the majority award, SGS v. Pakistan and SGS v. the
Republic of the Philippines, have nothing to support the majority award’s position, either in
their reasoning and dicta or in their facts. Indeed, one can say the opposite as far as SGS v.
the Republic of the Philippines, given its critical and distant position vis à vis Fedax, as
exposed above. In both SGS cases the issue was whether services provided by SGS and
attendant expenses largely undertaken outside the territory of the host State, could be
considered as an investment “in the territory” of the host State. And in both cases the
Tribunals found that a reasonable part of the service and the expenses, though perhaps not the
major part, took place in the territory of the host State; and that the service was not
constituted of two operations : one outside and the other inside the territory of the host State,
but one overall service straddling the frontier of that State, and as such well anchored in its
23
See Waibel, Sovereign Defaults, supra note 16, p. 225.
42
territory.
107 - In none of these cases, including Fedax, or others in which the question of territoriality
was raised, was the need to trace the alleged investment to an underlying specific project or
activity within the territory of the host State, questioned. The issue in all these cases was
whether what takes place outside the territory of the host State, such as the payment of funds
or the endorsement of the promissory notes, distends this link to the point of legally severing
it.
108 - The situation in the present case is totally different. The security entitlements in
question are free-standing, and totally unhinged. They do not form part of an economic
project, operation or activity in Argentina. Nor are they issued in support of a public project
or a commercial undertaking there. In other words, they have no specific economic
anchorage in Argentina, allowing them to be seen and considered as an investment “in the
territory of Argentina”.
109 - The inescapable conclusion of this rapid survey of cases invoked by the majority award
in support of its position, is that, unlike these cases, there is a missing link in the present case
that prevents the alleged investment from being recognized as covered or protected
investment under the BIT and article 25 of the Convention, namely the traceability of the
alleged investment (or its link, be it tenuous) to an underlying specific economic project or
operation taking place in the territory of Argentina.
110 - iv - In order to surmount this tough obstacle, the majority award resorts to an
43
astounding exercise of logical gymnastics, operating a double logical somersault that ends up
standing the sequence of legal reasoning on its head, by way of two unverifiable (and for one
totally untenable) assumptions,
111 - a) As concerns the first of these assumptions, the majority award asserts that : “There
is no doubt that the funds generated through the bonds issuance process were ultimately made
available to Argentina [and here comes the first logical jump dissimulated by the first
assumption] and served to finance Argentina’s economic development” (para. 378, emphasis
added). Thus, according to the majority award, any loan or funds made available to a
government (and regardless of the way it uses them, which is “irrelevant”) must be deemed,
by virtue of an unrebuttable presumption, to be contributing to the development of the
country of that government, and thus corresponding to the economic concept of investment
of article 25 of the ICSID Convention (i.e. contributing to the expansion of the productive
capacities of the country).
112 - But this presumption is rebuttable on two grounds. First, not every loan can constitute
an investment in the sense of the Convention. A simple loan in itself is merely an “ordinary
commercial transaction”. For it to become an investment, it has to fulfil the conditions
discussed above. This is why, for example, in the enumeration of the various types of
protected investment in article 1/1 of the BIT, under (d), “credits”, which are a species of
loans, are qualified by the clause “which are directly linked to an investment”.
113 - The other ground on which this first presumption is rebuttable is the mere fact that not
44
all funds made available to governments are necessarily used as “investment” in projects or
activities contributing to the expansion of the productive capacities of the country. Such
funds can be used to finance wars, even wars of aggression, or oppressive measures against
restive populations, or even be diverted through corruption to private ends. This is why, for
such loans to constitute investments under the ICSID Convention, they have to be concretely
traced, even at several removes, to a particular productive project or activity in the territory
of the host country; and not merely by postulating a stop-gap abstract assumption that does
not hold its ground.
114 - b) The second assumption is even more remarkable than the first, and totally
dependent on it. In fact, it is undetachable from it; thus one has to formulate them together, in
something like the following : Because the funds were made available to Argentina and used
by it (whatever the use which is “irrelevant” and which is not traceable or proven, but
assumed), these funds “as such, must be considered to have contributed to Argentina’s
economic development [first assumption] and thus [and here comes the second assumption
dissimulating the second logical jump] to have been made in Argentina” (para. 378, emphasis
added). Put in straight forward language, without the ellipsis, the proposition would be : As
the funds made available to Argentina must be considered (i.e. assumed) to have contributed
to Argentina’s economic development, they also, and as a necessary consequence, must be
considered (or assumed) to have been made in Argentina.
115 - Thus, by assuming a contribution to the economic development of Argentina, one
necessarily assumes also its location in Argentina, without having to demonstrate or prove
45
either. The falsity of such reasoning is too evident to need further elaboration.
116 - Moreover, the presumed logical necessity of drawing or deducing the second
presumption from the first is also false. It suffices to recall in this regard the dictum of SGS v.
the Republic of the Philippines that “investment made outside the territory of the Respondent
State, however beneficial to it, would not be covered by the BIT”, and the example it gives of
“the construction of an embassy in a third State or the provision of security services to such
an embassy” which “would not involve investments in the territory of the State whose
embassy it was, and would not be protected by the BIT”24
.
117 - Here again, the logical legal sequence is stood on its head. Rather than demonstrating
the contribution by tracing the alleged investment to (or proving its link with) a productive
project or activity in the territory of the host State, the majority award deduces a situs from a
presumed contribution hanging in the air.
24
Case cited in note 16 above, at para. 99.
118 - In conclusion, the present case is, to my knowledge, the first one to come before an
ICSID tribunal in which the alleged investment is totally free-standing and unhinged, without
any anchorage, however remote, into an underlying economic project, enterprise or activity
in the territory of the host State. None of the logical short-cuts put forward by the majority
award to palliate this absence, holds water.
46
119 - As both the ICSID Convention and the BIT require that the investment be made in the
territory of the host State for the investment to be covered by the treaty and fall within the
jurisdictional ambit of ICSID, and as this territorial link is lacking in the present case, I
conclude that in the absence of a “protected investment”, the case has to be dismissed, as
falling outside the jurisdiction ratione materiae of the Tribunal.
IV - The First Preliminary Objection : Collective Mass Claims Actions and
the Consent of the Respondent
120 - This preliminary objection is based on the contention of the Respondent that its consent
to arbitrate by acceding to the ICSID Convention and concluding the BIT does not cover
collective mass claims actions, including the present one; that handling mass claims in a
single arbitral case or proceeding, or as one arbitral suit or action, necessarily undermines
the due process rights of the Respondent and would be unmanageable; and that, in any case,
the Tribunal is not empowered, under the Convention and the Rules of Procedure, to effect
the procedural changes enabling it to handle such an action.
121 - The preliminary objection raises three issues :
1 - Does the consent of Argentina cover collective mass claims actions ? and is this a
question of jurisdiction or of admissibility as contended by the majority award ?
2 - What is the legal nature of collective mass claims actions or suits, and what are the
problems and questions they raise, particularly as to the due process rights of the parties and
47
to the interpretation of the scope of arbitral clauses - i.e. the jurisdictional title - both in
municipal and in international law.
3 - Finally, does the Tribunal have the power to effect the procedural changes, or
“adaptations” as it calls them, that it needs to be able to handle the case (para. 491).
A - A Plea to Jurisdiction or to Admissibility ?
122 - The majority award adopts, “for distinguishing issues of jurisdiction from issues of
admissibility”, the following criterion :
“If there was only one Claimant, what would be the requirements for ICSID’s
jurisdiction over its claim ? If the issue raised relates to another aspect of the
proceedings, which would not apply if there was just one Claimant, then it must be
considered a matter of admissibility and not of jurisdiction” (para. 249).
123 - Applying this criterion to the present case, the majority award “answers in the
negative” the question “whether a specific consent regarding the specific conditions in
which the present arbitration would be conducted is required”, on the basis of the following
argument :
“Assuming the Tribunal has jurisdiction over claims of several individual claimants
...it is difficult to conceive why and how the Tribunal could ‘loose’ jurisdiction where
the number of claimants outgrows a certain threshold...what is the relevant threshold?
And can the Tribunal really ‘loose’ jurisdiction it has when looking at Claimants
individually ?” (paras. 484-490).
48
124 - On that basis, the majority award dismisses Argentina’s objection :
“Issue 1(a) : Argentina’s consent to jurisdiction of the Centre includes claims
presented by multiple Claimants in a single proceeding” (para. 502).
125 - The majority award considers that :
“...the relevant question is not ‘has Argentina consented to the mass proceedings’, but
rather ‘can ICSID arbitration be conducted in the form of ‘mass proceedings’
considering that this would require an adaptation and/or modification by the Tribunal
of certain procedural rules provided for under the current ICSID framework”?
(para.491); “which is a question of admissibility” (para. 492).
126 - This legal recharacterization of Argentina’s plea by the majority award is, in my view,
conceptually wrong. It adopts an extremely narrow, in fact partial, concept of jurisdiction,
limiting it to the ambit within which jurisdiction is exercised. But, as explained above (para.
10 ff), jurisdiction is first and foremost a power, the legal power to exercise the judicial or
arbitral function. Any limits to this power, whether inherent or consensual, i.e. stipulated in
the jurisdictional title (consent within certain limits, or subject to reservations or conditions
relating to the powers of the organ) are jurisdictional by essence. They are no less
jurisdictional, in fact more so, than the limits relating to one of the four dimensions of the
ambit within which jurisdiction is exercised (para. 13 above).
127 - The preliminary objection of Argentina draws precisely upon this category of limits to
jurisdiction as a power, maintaining that Argentina’s consent cannot be interpreted to cover
49
the power of the Tribunal to hear collective mass claims actions requiring resort to atypical or
abnormal procedures. If that is the purport of the preliminary objection, then finding that the
Tribunal has jurisdiction ratione materiae over the subject-matter of one claim or even the
60.000 claims (a finding that I already refuted, for absence of a protected investment in the
previous section), neither disposes of this preliminary objection, nor does it address it in
anyway.
B - The Legal Nature and Specificities of Collective Mass Claims Actions and their
Bearing on the “Consent to Arbitrate”
128 - To address this preliminary objection, one has first to examine the legal nature and
specific characteristics of mass claims actions, and the abnormal or atypical procedures their
handling requires, to be able to evaluate whether a mere consent to submit to arbitration can
be construed as extending to include them or not.
129 - In their pleadings, particularly during the oral phase, the Parties and their legal experts
paid special attention to the legal characterization of the present proceedings. The
Respondent considered it not a simple multi-party action, but a collective mass claims action;
one of the leading experts in the field called by the Respondent, Professor Richard Nagareda,
classified it as a class action. Counsel to the Claimants, by contrast, argued that it is a mere
multi-party action; as there is no limit to the number of parties in such an action and no
known threshold that separates a multi-party from a collective mass claims action. Later on,
Counsel to the Claimants spoke of a non-class aggregate proceeding.
50
130 - The majority award tries to skirt round the subject (in order not to tie its hands) by
using the neutral expression “ ‘mass proceedings’...as referring simply to the high number of
Claimants appearing together as one mass, and without any prejudgement on the procedural
classification of the present proceedings as a specific kind of ‘collective proceedings’
recognized under any specific legal order” (para. 480). But it does not quite succeed in
keeping this neutral stance; for after summarizing part of a recent article surveying different
types of collective proceedings25
, it ends up classifying the present proceedings as a hybrid,
partaking of two species. But this exercise in legal genetic engineering, like all genetic
engineering, risks producing a monster. In any case, it does not work here for reasons
expounded below.
131 - The survey of different types of collective proceedings, based on Strong’s article,
classifies them in two categories :
25
Stacy I. Strong, “From Class to Collective : The De-Americanization of Class
Arbitration”. Arbitration International, vol. 26, No 4(2010), pp. 493-548. Footnote 176 of the
Award, wrongly cites the author’s name as Stark, and not Strong.
51
i - Aggregate procedures : I use advisedly “procedures” rather than “proceedings”,
because the aggregation takes place before the “proceedings” properly speaking, i.e. the
adjudicative or judicial phase of examining the claims. It consists of aggregating or
centralizing the different individual claims arising “out of the same fact pattern”, as in the
present case, in different ways. In England, the Group Litigation Order establishes a judicial
registry of these claims, which are then assigned to the same judge for management purposes.
In the US, such related claims are “consolidated in a single federal venue to ensure
efficiencies of scale during the pre-trial period”, beyond which “the cases are separated and
heard individually regarding issues of liability and/or damages”26
.
132 - Both methods are legal techniques for rationalizing the distribution of individual
claims arising from the same events, upstream, at the pre-trial or pre-judicial stage. But once
the individual claims are before the tribunals for judgment, they are handled as such i.e. as
individual claims, according to the normal procedures, though the tribunals can use
procedural devices normally available to them, such as joinder and consolidation, for
purposes of procedural economy. But there is no change or alteration of the procedure
followed to handle these claims other than as normal individual claims. And it is on this last
point that the “aggregate procedures” model does not suit the purposes of the majority award.
26
Strong, ibid. p. 504.
133 - ii - Representative proceedings : This is the most important type of collective
proceedings, its prime example being “class actions” that emerged relatively recently in the
52
US and elsewhere. The other species of representative proceedings are in fact minor
variations on the theme of class actions.
134 - Class actions or representative proceedings can be brought either by a member of the
class, acting as a kind of negotiorum gestor on behalf of the other members of the class (who
can join in or take their distance from the action through the techniques of “opting in” or
“opting out”) or by an agent representing the whole class, such as a consumers association or
the Sierra Club.
135 - What makes class actions or representative proceedings (unlike aggregate claims
subject to aggregate procedures only at the pre-trial or pre-judicial stage), feasible or
amenable to be treated judicially or adjudicatively in one single proceedings or as one action,
is that in a class actions or representative proceedings there is, in the final analysis, only one
claim, albeit with many, or even a mass of claimants. The tribunal deals thus with one claim
and can examine every aspect of it specifically, through adversarial debate and scrutiny that
guarantees to the parties, particularly the respondent, all their due process rights.
136 - The first attempts to introduce class actions, which are the prime model of
representative proceedings, were related to issues of public interest such as the protection of
the environment or the consumers. In this, they recall (at least in their rationale, without
necessarily assimilating them to) an ancient ancestor, the institution of actio popularis,
whereby a citizen, in the name of the community, acts in defense of an indivisible public
good. And the remedies sought in class actions were geared to the nature of these types of
53
public interest claims, namely declaratory judgments and injunctive relief, i.e. orders to cease
and desist, that are the same for all claimants and which are formulated as a function of the
subject-matter of the claim, and not as a function of the personality, or the number of the
claimants, or their individual subjective grievances.
137 - It is true, as is stated in the majority award, that in some jurisdictions, the relief “can
take the form of individual damages or representative relief (e.g. declaratory or injunctive
relief)” (para. 483). But the majority award omits to reproduce from Strong’s article it
excerpts the following comment on the cumulation of the two types of relief :
“One approach allows class claimants to pursue individual damages as well as other
types of relief. This technique, which is common in the US, has met with a great deal
of criticism internally, since the provision of individual damages in a representative
suit is seen in many (but not all) civil law systems as violating the defendant’s right to
defend against individualized, as opposed to generalized claims, as well as the rights
of individualized members of the collective to conduct their own individualized
cases”27
.
27
Strong, ibid, at p. 504.
138 - After this brief survey of the two main types of “collective proceedings”, the majority
award proposes a legal characterization of the present case as “a sort of hybrid kind of
collective proceedings, in the sense that it starts as aggregate proceedings, but then continues
with features similar to representative proceedings due to the high number of claimants
54
involved” (para. 488).
139 - However, as mentioned above, this exercise of legal genetic engineering does not work.
First of all, there is no explanation whatever of the reason or cause nor of the trigger of such
a transmutation of the proceedings, starting as one thing that no sooner transforms itself into
quite another. It is understandable that the case starts as aggregate proceedings, in the sense
explained above and in the majority award, considering that it groups a multitude of
individual claims arising out of the same fact pattern. But these remain individualized claims.
Following the aggregate procedures model simply means that they would be collected and
managed during the pre-trial or strictly pre-judicial or pre-arbitral phase, in order to
rationalize their funneling to other tribunals that will examine them judicially as individual
claims. But they would not change nature or characteristics as a result of their prior
aggregation. They remain individualized claims.
140 - For the proceedings to change suddenly into a representative one, the individual claims
have to be cast into one claim or transformed into identical fungible claims, which comes
down to the same thing (one claim, multiplied by the number of claimants) to enable the
tribunal to treat them judicially as one. But this is not possible here, as the claims retain their
individualized character.
141 - To skirt round this fundamental problem, the majority award resorts to the slippage
technique, considering that “group examination of claims is acceptable where claims raised
by a multitude of claimants are to be considered identical or at least sufficiently
55
homogeneous” (para 540; emphasis added ).
142 - However, homogeneity is in the eyes of the beholder. One can always reach a sufficient
level of homogeneity, i.e. common denominators, by climbing up the ladder of abstraction
and/or by weeding out all the specificities of the claims that appear inconvenient. And that is
precisely what the majority award does with the claims under consideration.
143 - Indeed, the majority award sets aside all the specificities of the claims concerning the
security entitlements (price, date of purchase, place of purchase, in which currency,
applicable law, chosen forum, etc.) as characteristics relevant only to the contractual rights of
their holders, i.e. to contract claims; while what counts here, according to the majority award,
are the treaty claims, which are homogeneous (paras. 541-543). And the majority award
harps on the adjective “homogeneous”, by repeating it throughout this part in almost every
line. But no amount or repetition can exorcize the fundamental flaws in this line of argument.
Indeed, it suffices to ask, how can the Tribunal for example evaluate a treaty claim for
compensating damages caused to an asset, without knowing (or while making abstraction of)
the time the asset was acquired, the price paid for it and the currency of denomination ?
144 - If one examines what the majority award calls “the claims deriving from the BIT”
(para. 543), which it considers “homogeneous”, the result is clear. They are claims related to
the Argentinean economic crisis that led to Argentina’s cessation of payment. In other words,
this crisis constitutes “the same fact pattern” out of which the claims arose, and which is the
criterion used for resorting to “aggregate proceedings”, according to the majority award. But
56
as the award itself states, this aggregation takes place only at the pre-trial, or pre-judicial
phase. By contrast, during the judicial phase, the claims are treated individually. In other
words, the “level of homogeneity” resulting from the circumstance that the claims arose “out
of the same fact pattern”, is sufficient for aggregating and registering them for purposes of
pre-trial management and rationalizing their funneling towards the tribunals that will
ultimately examine them as individual claims; but not sufficient for what the majority award
calls “group examination”, i.e. their examination as if they are one claim (as in a class action
or a representative proceedings), which they are not, either in fact or in law.
145 - The efforts of the majority award at legal genetic engineering lead nowhere. The hybrid
proceedings, brain-child of the majority award’s legal imagination, lacks a theoretical basis
to stand on, let alone a solid legal basis. It also raises the question of the extent to which such
an imaginary “hybrid action” and, for that matter, the particularities of collective proceedings
in general, have a bearing on the interpretation of the “consent to arbitrate”.
C - Judicial Innovation, Collective Proceedings and “Consent to Arbitrate”
146 - In their pleadings, counsel and legal experts for the Claimants argued - to incite this
Tribunal to do the same - that in the US, courts foreshadowed class actions and devised their
procedures, before they were formally adopted by legislation and integrated in the codes of
civil procedure.
147 - Regardless of the veracity of this narrative and assuming it is true, it does not help
57
much supporting the conclusion sought to be drawn from it, i.e. that the same can be done by
this ad hoc international arbitral Tribunal. This is because courts in general are part of the
judiciary, the judicial branch or arm of the State. Their judicial power, i.e. jurisdiction, thus
stems from above, and not from the consent of the parties or litigants before them. As
repositories of the judicial power of the State, they have some leeway, particularly in
common law jurisdictions, to develop the law and engage, within reasonable limits, in legal
experimentation; unlike arbitral tribunals that draw their jurisdiction from the consent of the
parties and have to confine their judicial activities within the bounds of this consent.
148 - Even in the U.S, the national jurisdiction where it is contended that the courts
foreshadowed class action and devised procedures for them before they were enshrined in
legislation, the US Supreme Court, in two successive decisions in 2010 and 2011, has been
quite strict in interpreting consent and underscoring the need to respect its bounds, when it
came to arbitration, in response to exactly the same question that is put to this Tribunal, i.e.
“Does a regular arbitration clause cover class actions ?”
149 - I realize that the majority award, following in this the Claimants, refuse to consider the
present collective mass claims action as a class action, although its “hybrid model” ends up
being a “representative proceeding”, which is more or less the same thing. But the US
Supreme Court’s reasoning and answer turns on the “mass” character of the action rather
than on its characterization as a class action or otherwise.
58
150 - In the first of these cases, Stolt-Nielsen S.A. v. Animal Feeds International Corp28
, the
Court held :
“An implicit agreement to authorize class-action arbitration... is not a term that the
arbitrator may infer solely from the fact of the parties’ agreement to arbitrate. This is
so because class action arbitration changes the nature of arbitration to such a degree
that it cannot be presumed the parties consented to it by simply agreeing to submit
their disputes to an arbitrator”.29
151 - The Court also noted that the “changes brought about by the shift from bilateral
arbitration to class-action arbitration [are] fundamental”; and concluded that consent to
class-action arbitration cannot be “presumed” from the parties “mere silence”, given the
fundamental differences between class-action arbitration and bilateral arbitration30
.
28
130 S.Ct.1758 (2010).
29
Ibid. at 1774.
30
Slip Opinion, at 22-23.
59
152 - In the second case AT&T Mobility LLC v. Concepcion ET UX (2011)31
, the Court,
while referring to Stolt-Nielsen, and building on it, goes even further, in highlighting the
difference between the two types of arbitration :
“Class arbitration...interferes with fundamental attributes of arbitration...class
arbitration greatly increases risks to defendants. The absence of multilayered review
makes it more likely that errors will go uncorrected. That risk of error may become
unacceptable when damages allegedly owed to thousands of claimants are aggregated
and decided at once. Arbitration is poorly suited to these higher stakes”32
.
153 - The powerful reasoning of these two judgments and the strong arguments they put
forward - the fundamental differences between the two modes of arbitration, the regular
bilateral one and the class action or representative proceedings arbitration; differences that
“change the nature of arbitration”; and the great risks to which the later mode exposes
defendants - all converge to highlight the necessity of a special consent of the parties in order
to resort to class action or representative proceedings arbitration; consent that cannot be
merely deduced or assumed from a simple consent to go to arbitration.
D - The Interpretation of Silence
154 - The majority award totally ignores this widely known and commented upon
31
563 US (2011), (decision of 27 April 2011).
32
Slip Opinion, at 2.
60
jurisprudence and arguments. Instead, it goes into a cursory “interpretation of the silence of
the ICSID Framework”.
155 - The reasoning of the majority award can be summarized as follows : given the “object”
of the BIT and the “spirit” of the ICSID Convention, which are presumably, according to the
majority award, reduced solely to affording “effective protection” to investments and
investors, “where the BIT covers investments which are susceptible to involve a high number
of investors...it would be contrary to the purpose of the BIT and the spirit of ICSID to require
in addition to the consent to ICSID arbitration in general a supplementary express consent to
the form of such arbitration” (para. 518); i.e. “it would be contrary to the purpose of the BIT
and to the spirit of ICSID to interpret this silence as a ‘qualified silence’ categorically
prohibiting collective proceedings, just because it was not mentioned in the ICSID
Convention” (para. 519). And if consent to ICSID arbitration covers collective proceedings,
it has also - presumably by necessity or necessary implication - to cover the power of the
Tribunal to devise the adequate procedures to deal with them, in derogation of “the standard
procedure” (para. 519).
156 - This wobbly line of argument by affirmation, assumption and reasoning by necessity,
does not withstand scrutiny. In the first place, it is premissed on the assumption that the
Tribunal has jurisdiction ratione materiae over the subject-matter of the mass claims; an
assumption that I have refuted in the previous section.
157 - Secondly, the reasoning is also premissed on and proceeds from a purely subjective,
truncated and partial representation of the object and purpose of the ICSID Convention and
61
the BIT, which is a recurrent theme throughout the award. The reader cannot but be struck by
the repetition expressis verbis in two successive paragraphs (para. 518 and 519), at a few
lines interval, of the reference to the “spirit” of the ICSID Convention and the “purpose” of
the BIT. Spiritism apart, the object and purpose of these two treaties - the ICSID Convention
and the BIT - are described as being exclusively to afford maximum protection to foreign
investment and foreign investors; as if these treaties were “unilateral contracts” creating
rights for the benefit of one party only. In consequence, according to this vision, all the
provisions of these treaties have to be interpreted exclusively with this aim in mind.
158 - Viewed from this perspective, all the limitations to the jurisdiction of ICSID tribunals,
whether inherent or patiently and carefully negotiated and stipulated in the treaty to protect
the interests of the State party (which are after all, the collective interests of its population)
are seen as obstacles in the way of achieving the “purpose” of the treaties, which have to be
overcome at any price and by whatever argument.
159 - This unilateral vision is in stark contrast to the “object and purpose” of the ICSID
Convention, as expressed in the Report of the Executive Directors in the following terms :
“While the broad objective of the Convention is to encourage a larger flow of private
international investment, the provisions of the Convention maintain a careful balance
between the interests of investors and those of host States”.33
33
Report of the Executive Directors on the Convention, para. 13.
62
160 - Proceeding from these premisses, the majority award simply affirms that “it would be
contrary to the purpose of the BIT and to the spirit ICSID to interpret this silence [of the
ICSID Convention or Framework as it calls it] as a ‘qualified silence’ categorically
prohibiting collective proceedings, just because it was not mentioned in the ICSID
Convention” (para. 519).
161 - The majority award does not explain at any length what it exactly means here by
“qualified silence”, nor why it considers that this concept does not apply in the present case
or the consequences of its application or non application. This results in confusion at two
levels, the first of which relates to the meaning and relevance of the term in casu. “Qualified
silence” is not a term of art; for example, it does not appear in law dictionaries. Occasionally,
however, one encounters, in writings on American constitutional law, such expressions as
“the qualified silence of the law-maker” (or “the intentional imperfection of the law”). They
describe, in a piece of legislation dealing with a particular subject, its silence over certain
substantive issues falling within the subject treated, that were left intentionally open (for
future legislative treatment, for lack of consensus, etc.). In consequence, analogy cannot be
used to extend the legal regulation to these issues by those who subsequently interpret and
apply the law, as this would go against the intent of the law-maker. But this is a completely
different context, poles apart from that of establishing the existence and scope of the
individual consent of a State to submit to arbitration, which is what is at issue here.
162 - This brings me to the second level of confusion in the majority award. Indeed, the
award, in its analysis, confuses here two layers of consent (see above para.15) :
63
a) the consent or acceptance to be bound by the ICSID Convention as such, by
ratifying or acceding to it, which by itself does not contain any consent or obligation to
arbitrate; and
b) the specific acceptance by a State to submit a dispute or certain disputes to
arbitration within ICSID, whether by way of a BIT or otherwise (a compromissory clause,
etc.), which constitutes the specific jurisdictional title.
The two do not coincide in time nor necessarily in their scope (para. 13 above).
163 - Argentina’s jurisdictional plea is directed more specifically to the second layer of
consent, though it covers both layers; whilst the majority award addresses only the first
layer, ignoring almost totally the second layer of consent, which is the crux of Argentina’s
plea.
164 - If what is at issue here is Argentina’s specific consent to arbitrate, then what counts
most in the interpretation of the jurisdictional title is the establishment of the validity and
scope of that consent at the time it was given. It is thus paradoxical, that in order to support
its affirmation that the ICSID Convention’s silence over “collective proceedings”, is not a
“‘qualified silence’ categorically prohibiting [such] proceedings”, the majority award
advances the argument that “at the time of conclusion of the ICSID Convention, collective
proceedings were quasi inexistent, and although some discussions seem to have taken place
with regard to multi-party arbitrations, these discussions were not conclusive on the intention
to either accept or refuse multi-party arbitrations, and even less so with regard to
admissibility of collective proceedings” (para. 519).
64
165 - It is paradoxical because this argument supports exactly the opposite proposition to that
which the award is attempting to prove. The fact that class actions or representative
proceedings were almost unknown in national jurisdictions, and more so on the international
level, at the time of the conclusion of the ICSID Convention in the mid-sixties of the last
century, proves that these representative proceedings were way beyond the “horizon of
foreseeability” of the drafters of the ICSID Convention. Those drafters could not have
envisaged such proceedings; nor is there any basis to assume that they would have included
them, had they envisaged them, given the fundamental differences between these proceedings
and the arbitration model familiar to them (as discussed below). Those drafters were simply
creating a framework for ad hoc international arbitration, within the parameters of ad hoc
international arbitration as they knew them at that time, particularly its specific consensual
basis for every case, as with all international adjudication. They were not establishing an
open-ended standing court of general jurisdiction (juridiction de droit commun) covering all
possible present and future disputes. This is clearly reflected in the last paragraph of the
Preamble of the ICSID Convention :
“Declaring that no Contracting State shall by the mere fact of its ratification,
acceptance or approval of this Convention and without its consent be deemed to be
under any obligation to submit any particular dispute to conciliation or arbitration”
166 - Thus, the above argument of the majority award based on the unforeseeability of
collective actions at the time of drafting the Convention, paradoxically bears in favour of
interpreting the silence of the ICSID Convention as not extending to cover such an extra-
65
ordinary type of judicial activity as representative proceedings or class actions. However,
even if we assume arguendo that the ICSID Convention can be over-stretched by subsequent
interpretation, pushing back its outer-limits to accommodate collective proceedings, this
would not apply to the second layer of consent, the specific consent to arbitrate within the
ICSID framework.
167 - Here, the interpretation of the jurisdictional title has to establish the reality, validity and
scope of consent, as it was given, at the time it was given, by each party. And it is here that
the “interpretation of silence” of the jurisdictional title becomes crucial : Does the mere
“consent to arbitrate” within the ICSID framework cover mass claims actions and
proceedings (whatever we call them) ?
168 - As a rule, silence does not have a meaning in itself. It cannot be made to speak one way
or another; unless it is surrounded by circumstances that impregnate it with a certain
meaning; what is called in civil law and in French “silence circonstancié”, a more expressive
term than “qualified silence” which is also used, on rare occasions, in international law in the
same sense34
.
34
See for example the WTO Panel Report Guatemala - Definitive Anti-Dumping
Measures on Grey Portland Cement from Mexico, Report of the Panel of 24 October 2000, at
para. 8.23 : “We note that ‘acquiescence’ amounts to ‘qualified silence’, whereby silence in the
face of events that call for a reaction of some sort may be interpreted as a presumed consent”.
See also, in the same sense, Mark Villiger, Customary International Law and Treaties (The
Hague, Kluwer, 1977), p. 39.
169 - The circumstances that surround and impart meaning to silence, can be factual or legal.
Both these types of circumstances exist and can easily be identified in casu.
66
170 - The elements of fact or factual circumstances that qualify the silence of a jurisdictional
title by which a State consents to ICSID arbitration, in the sense of limiting it to normal bilateral
(or limited multiparty) arbitration, thus excluding collective mass claims actions or
representative proceedings, are simply the great differences between these two types of
proceedings.
171 - The sheer difference in volume and number of claims and claimants between the
bilateral or even the limited multiparty arbitration (14 Claimants in the largest case in
ICSID)35
, which permits an individual, detailed and adversarial examination of every claim
and claimant, and the 60.000 claimants in the instant case, is staggering. It constitutes what is
called in physics a “quantum leap” (saut qualitatif) which transforms quantitative change
beyond a certain point into qualitative change36
. The change here is from the micro-analysis
and treatment of individual phenomena (i.e. claims and claimants) individually, into the
35
Bernardus Henricus Funnekotter and others v. Zimbabwe. ICSID Case No. ARB/05/6,
Award of 22 April 2009.
36
The argument raised at one point by the Claimants, but not taken up by the majority
award (though occasionally it seems to imply it), that collective mass claims actions are mere
multi-party actions as there is no recognized threshold that separates them, is extremely tenuous,
verging on the absurd. International law knows several such situations. For example, up to now
there is no agreed or generally recognized threshold that separates air space, which is part of the
territory of the State, from outer-space which is a res communis. Similarly, until the Montego
Bay UN Convention on the Law of the Sea of 1982, there was no general agreement on the width
of the territorial sea, i.e. on the threshold that separates that sea, which is part of the territory of
the State, from the high seas. But here again, the absence of an agreed threshold did not mean
that these two areas did not have, or prevented them from having two radically different legal
status. In any case, whatever the threshold that separates, or the criterion for distinguishing,
multi-party actions from collective mass claims actions, it is obvious that 60.000 claims fall on
the side of collective mass claims actions.
67
macro-analysis and treatment of large masses of individual phenomena as an impersonal
whole, that cannot take into consideration the individuality of the components.
172 - It is precisely this fundamental difference that the US Supreme Court highlighted in the
Stolt Nielsen judgement when it found that representative proceedings or “class action
arbitration changes the nature of arbitration to such a degree that it cannot be presumed the
parties consented to it by simply agreeing to submit their disputes to an arbitrator”37
.
173 - The other type of circumstances that can surround silence and impart it with meaning
and significance, are of a legal nature, pertaining to the legal context of the interpreted
instrument, such as the concepts of “course of performance”, “course of dealing” and “trade
usages”, in interpreting contracts, or other legal concepts and principles that form part of the
legal environment of the interpreted instrument. One such legal concept or rule which is well
known in the arbitration environment but over which the majority award passes in silence, is
described in a very recent article by the same author quoted by the majority award, S.I.
Strong38
, as follows :
37
Supra note 29, and accompanying quotation.
38
S.I. Strong, “Does Class Arbitration ‘Change the Nature’ of Arbitration ? Stolt Nielsen,
AT&T and Return to First Principles”, 17 Harvard Negotiation Law Review , (forthcoming
2012), available electronically at http://ssm/abstract=179128.
68
“[T]he first question raised in every arbitration is whether the parties have agreed to
arbitrate this particular dispute. This issue ... can be termed ‘primary consent’...
Instead, class arbitration focuses for the most part on what can be called ‘secondary
consent’, meaning consent to this particular type of procedure”. She adds : “This
concept is by no means unique to class disputes, since traditional multiparty
arbitrations are also required to establish secondary consent in cases where the
arbitration agreements are silent or ambiguous as to multiparty treatment”39
.
174 - According to this concept or rule, the need to establish secondary consent applies to
both multi-party arbitration and a fortiori to collective mass claims actions or representative
proceedings.
39
Ibid, pp. 55-56, emphasis added, footnotes omitted. In footnote 255, Strong cites Lew
et al, Comparative International Commercial Arbitration (2003), on the need to establish
secondary consent in both class and traditional multiparty arbitration; and in note 256 she cites
numerous cases on the requirement of secondary consent in multiparty arbitrations.
69
175 - This applies also in the ICSID context. That we have no precedents or writings about
the subject is simply because the issue did not arise until now before an ICSID tribunal as
concerns collective mass claims actions. Concerning multi-party arbitrations, as the majority
award states, the issue was raised during the drafting of the ICSID Convention, but the
question was left open. It was debated during the latest revision of the Rules, but again was
not expressly addressed in the revised Rules of 2006. But the absence of written regulation
does not mean absence of rules. Indeed, the few cases of multi-party arbitrations that took
place within the ICSID framework (and also NAFTA), were always either with the clear
agreement of the parties or with no objection from the Respondent, which amounts to an
implied consent. Thus, the rule of “secondary consent” was consistently upheld in multiparty
arbitration40
. And if this is the case in multi-party arbitrations, a fortiori it has also to be
the case for collective mass claims actions.
E - Mass Claims Processes in International Law
176 - Most of the analysis up to now has been of the “consent to arbitrate” and its limits in
national legal systems, including international commercial arbitration which is, in the final
analysis, subject to national law and national judicial control. But of course ICSID arbitration
is situated on the international level, i.e. in international law, which is, as described above
(paras. 8,16) much more strict and exacting as regards the requirement of consent and its
40
Schreuer, The ICSID Convention. A Commentary (2d edition, 2009), p. 163, para. 279,
cites the Klöckner v. Cameroon case, where Cameroon initially raised the objection that the use
of the word “national” in singular in Article 25(1) of the ICSID Convention barred an arbitration
with multiple parties on the Claimants side, but subsequently dropped the objection.
70
ascertainment as basis of jurisdiction.
177 - The foregoing analysis thus applies a fortiori and with greater legal force when it
comes to ad hoc international arbitral tribunals, such as the ICSID tribunals, and the
interpretation of the “consent to arbitrate” before them. The analysis leads inexorably to the
conclusion that a mere “consent to arbitrate” does not cover the fundamentally different and
atypical proceedings of collective mass claims actions, such as the present one (whether we
qualify them as collective, class, representative, aggregate or otherwise), and that a
“secondary consent” is required for such proceedings to be covered.
178 - In their written and oral pleadings, counsel and legal experts for the Claimants argue
that several mass claims processing programs took place on the international level in recent
time, with a view to convincing the Tribunal a) that such procedures are possible and
admissible on the international level; and b) that the Tribunal, by analogy, has jurisdiction
and is empowered to do the same, which also implies, according to them, the Tribunal’s
power to devise the appropriate procedures to handle such mass claims.
179 - Although the majority award chose to ignore this aspect, I consider it important and
highly relevant to examine the other experiences of processing mass claims in international
law, in order to establish what can or cannot be done in the present case.
180 - The analogy that counsel and legal experts for the Claimants sought to draw between
the mass claims processing programs that existed up to now on the international level and the
71
present collective mass claims action, is a false analogy. It glosses over fundamental
differences between these programs and the present case.
181 - First and foremost among these differences is the legal basis on which these programs
were established, and which is the crux of the matter here under discussion, namely whether
the Respondent, by merely consenting to arbitration within the ICSID framework, has also
consented to arbitrate the present collective mass claims action.
182 - If we examine all the examples of international mass claims programs that have been
mentioned by counsel and legal experts for the Claimants, and which are discussed in one of
the Claimants legal annexes to which recurrent reference was made41
, covering indeed all the
known instances, we note the following common major features that distinguish them from
the present case :
183 - a) Each one of them, without exception, was specifically established to process a
particular set of mass claims. None of them was set in motion by an application to a standing
Tribunal, or on the basis of a prior compromissory clause (or another prior jurisdictional title)
as in the present case.
184 - b) In all these cases, with one exception, the program, as a specific response to a
specific crisis (“the same set of facts” giving rise to the claims), was established by an
41
CLA 185. In fact, it is an excerpt of a book edited by Judge Howard Holtzman and
Edda Kristjansdottir, International Mass Claims Processes : Legal and Practical Perspectives
(OUP 2007), prepared under the auspices of the PCA Steering Committee on International Mass
Claims.
72
agreement between the parties, i.e. on the basis of the consent of all parties concerned. The
one exception is the United Nations Compensation Commission (the UNCC), which was
established by the Security Council, using its exceptional powers (mandatory on all member
States) under Chapter VII of the UN Charter.
185 - The fact that all known cases, save for the UNCC, which is of no relevance to the
present case, are based on a special agreement between the parties, is of particular
significance to the present case, and to the point under discussion here, namely whether a
mere consent to arbitrate within the ICSID framework covers subsequent collective mass
claims actions. If in all known international cases (barring the fiat of the Security Council) a
special agreement is needed, then, in the absence of such agreement, but within an already
existing arbitral framework, at least a “secondary consent” is needed.
186 - c) All these mass claims processing programs on the international level, with one
notable exception (and another minor one, see below note 42) took the form of administrative
compensation commissions, quasi-judicial bodies rather than fully judicial or arbitral
tribunals. Of course, they are subject to certain due process requirements, but this is not the
same as judicial or arbitral due process. Thus, the procedure is normally non-adversarial, and
the remedies can be in terms of flat-rate compensation, etc. The treatment of claims is not
necessarily individual. Although, in the UNCC for example, for State claims and large
individual claims, the examination was individual. This type of procedure is not relevant to
the present case, where the requirements of the judicial or arbitral function and the judicial
or arbitral due process have to be completely and strictly observed.
73
187 - The one notable exception is the Iran-US Claims Tribunal. And this exception proves
the rule. Claims are dealt with individually, though occasionally related claims were dealt
with in one case (as is done in all ordinary tribunals). The Iran-US Claims Tribunal was
established by special agreement to deal with a specific set of claims. And it functions not as
an ad hoc tribunal dealing with all these claims at one go, in one suit or action, but as a
standing regular Tribunal examining claims individually and adversarially42
, the same as the
old Mixed-Arbitral Tribunals of the 19th and early 20th century.
188 - d) None of these mass claims processing programs, whether called Tribunal,
Commission or otherwise, took on itself or was authorized to invent its own procedures from
scratch. The main parameters of these procedures, particularly the ones that deviate from the
normal patterns and are thus more controversial, were set out in the agreement or the legal
instrument establishing the body (or their drafting and revision were kept in the hands of the
Parties or the organ that established the program).
189 - On all these crucial points, the international precedents and experiences make it
abundantly clear the necessity of the agreement of all parties in order to establish a
mechanism with jurisdiction to process a particular set of mass claims - or collaterally, to
endow an already existing body or framework with such jurisdiction - and to devise the
procedures for so doing; in stark contradiction with the findings of the majority award in the
42
The other minor exception is the first phase of The Claims Resolution Tribunal for
Dormant Accounts in Switzerland (CVRT-I), which also functioned along similar lines.
74
present case.
190 - My conclusion on the first preliminary objection of Argentina is that, in conformity
with what is followed both on the national law level, including in international commercial
arbitration, and on the international law level, which is the applicable law in the present case,
a mere acceptance to arbitrate does not cover collective mass claims actions (regardless of
the denomination) and that a special or secondary consent is needed for such collective
actions.
191 - In consequence, Argentina’s consent to the ICSID Convention and the BIT does not
extends to collective mass claims actions, and this Tribunal lacks jurisdiction to hear the
present case.
192 - And if the Tribunal has no jurisdiction to hear the case, the question of whether it has
the power to devise, revise or adapt the rules of procedure to handle it, becomes moot.
193 - But as the majority award found that the Tribunal has jurisdiction over the present
collective mass claims action and that, in consequence - reasoning by necessity - it also has to
have the power to devise or “adapt” the rules of procedure to handle it, I feel duty bound to
address this last question.
F - The Limits of Procedural Improvisation
75
194 - Even if arguendo the Tribunal had jurisdiction to hear the present collective mass
claims action (which I already refuted), it still faces a fundamental obstacle that prevents it
from proceeding as the majority award proposes to do. This insuperable obstacle is that the
Tribunal is not authorized, nor has the power, under the ICSID Convention and the
Arbitration Rules, to adopt proprio motu (on its own motion or initiative), a whole set of
rules of procedure, replacing the normal “Arbitration Rules” or large segments thereof,
without the agreement of the Parties or the approval of the Administrative Council, even if it
is only for the purposes of handling the present case.
195 - Two points of clarification are here in order. First, although this question was raised as
part of Argentina’s first preliminary objection, it is based on different grounds, as it assumes
arguendo that Argentina’s double consent to the ICSID Convention and the BIT covers
collective mass claims actions, which is precisely what Argentina contests in its first
preliminary objection. If Argentina’s consent is also mentioned here, it is as a third type or
layer of consent envisaged in Article 44 of the ICSID Convention, when it refers to rules of
procedure “agreed by the parties”, in the context of elaborating rules of procedure for a
specific case; on the assumption that the Tribunal has already established its jurisdiction over
the case. The distinction between these three layers of consent (to Convention, BIT and
within Article 44), should always be kept in mind, which is not the case of the majority
award.
196 - The second point of clarification concerns the legal nature of this preliminary point.
What is in question here is the inherent and institutional limitations on the power of the
76
Tribunal to devise, revise or “adapt” the rules of procedure in a specific case, i.e. certain
limits on its powers in exercising its judicial or arbitral function. As such, it is essentially a
question of jurisdiction - jurisdiction in its first and foremost sense of the legal power to
exercise the judicial or arbitral function (see above para.10) - and not a question of
admissibility as maintained by the majority opinion.
197 - Be that as it may, the results of the examination of this issue must be the same. For if it
leads to a finding that the Tribunal is not empowered to undertake the “revisions” or
“adaptations” set out in the majority award, the award would be ultra vires, regardless of the
legal characterization by the award of this preliminary issue.
198 - The majority award, having determined that the silence of the “ICSID framework” over
collective mass claims actions constitutes a “gap” - i.e. an unintended omission - and not a
“qualified silence” - i.e. a deliberate exclusion or limitation (while glossing over the limits of
Argentina’s consent to arbitrate), seeks to rest its power to fill this so-called “gap” on Article
44 of the ICSID Convention and Article 19 of the Arbitration Rules, describing them as “the
mere expression of the inherent power of any tribunal to resolve procedural questions in the
event of lacunae” (para. 521).
199 - Article 19 of the Arbitration Rules, entitled “Procedural Orders”, provides :
“The Tribunal shall make the orders required for the conduct of the proceedings”.
This Article has nothing to do with filling gaps. It merely states the obvious, that the Tribunal
controls and directs the unfolding of the proceedings through procedural orders, frequently
77
reflecting also the agreement of the parties. It thus decides such matters as fixing the timelimits,
the language or languages used, whether the case will have one or two phases, etc., as
well as deciding on the different requests of the parties and the incidents of procedure.
200 - Article 44 of the ICSID Convention is more to the point here under discussion.
Enunciating the applicable procedural law, it provides, in its second sentence :
“If any question of procedure arises which is not covered by this Section or the
Arbitration Rules or any rules agreed by the parties, the Tribunal shall decide the
question”.
201 - It is clear from this text that the situation envisaged in that provision is that of an issue
arising in the course of the proceedings which is not addressed by any specific rule of the
applicable procedural law. In other words, it is a situation where no rule exists at all; and not
one where a rule exists, but may not provide the best or most suitable solution to the issue at
hand, which raises the question of revising or replacing the existing rule rather than filling a
gap.
202 - It is also clear from the text that what is envisaged in terms of “filling a gap” is the odd
missing rule (by inadvertence or oversight) or a small missing element or cog of a rule
necessary for its implementation, what is called in the general theory of law a “technical
gap”; and not whole sets or chapters of rules that cover complete segments of procedure
(such as the administration of proof or the role and due process rights of the parties in the
proceedings). Such major subjects require a “legislative jurisdiction” (the jurisdiction of the
78
parties or the organ that adopted the initial legal regulation), going far beyond the limited
power of filling gaps that can be encompassed within the judicial or arbitral function.
203 - Indeed, all the examples that can be found of the exercise of such a power, whether in
the ICSID framework or by other international tribunals, never go beyond such limited gaps
calling for one specific norm or rule to fill it. Thus, in the ICSID context, the examples that
can be found of decisions taken in the exercise of the power granted by Article 44 are limited
to admitting an amicus curia submission (in Suez and AWG v. Argentina); or staying
proceedings pending the determination, by some other competent forum, of an issue relevant
to the tribunal’s own decision (SGS v. the Republic of the Philippines)43
.
204 - The majority award is conscious of these limitations. Thus, after affirming its power to
fill gaps under article 44, it adds : “...this does not mean that the scope of this power is
unlimited...the Tribunal is bound by the limits set forth by Article 44" (para. 520). It then
proceeds to articulate these limits as follows :
43
See Schreuer, The ICSID Convention. A Commentary (2nd ed. 2009), Art. 44, paras 57,
58).
“[T]his power is limited to filling gaps left by the ICSID Convention and the
Arbitration Rules...[i]n contrast [to] a modification of existing rules [which] can only
be effected subject to the parties’ agreement..”(para. 522, emphasis added). It is
“further limited to the filling of gaps... in the specific proceedings at hand... [i.e.]
limited to the design of specific rules to deal with specific problems arising in the
proceedings at hand” (para. 523). Thus, “the Tribunal cannot i) modify the current
79
arbitration rules [presumably in a specific case] without the Parties consent; ii) adopt a
full set of procedure unless the Parties have agreed that the Arbitration Rules....
should not apply without substituting their own rules” (para. 524).
205 - Problems however arise when the majority award tries to demonstrate how what it
proposes to do, by designing specific rules to replace a good part of the “Arbitration Rules”,
falls within these limits. The award starts by posing the key question :
“Can the Tribunal fill the gap created by the collective aspect of the claim on an ad
hoc basis and through the design of specific rules, or would this require the creation
and/or modification of general rules which are under the competence of the
Administrative Council ?” (para. 526, emphasis added).
206 - Responding to this question by the simple application of the limits on the power of the
Tribunal under Article 44, as formulated by the majority award itself, leads logically, and
inescapably in my submission, to a clear negative answer.
207 - But the majority award tries to fudge this clear answer by considering that “[t]his
question cannot (and should not) be answered in the abstract...[but] in a concrete manner, i.e.
asking itself (i) what are the [needed] specific rules...and (ii) can [they] be considered to
fall within [the power of] Article 44 ?” (para. 527).
208 - The majority award then embarks on a tortuous argumentation , ending with the
Tribunal arrogating to itself the power to set aside, in large measure, the existing Rules of
80
Procedure, and replacing them by another set of rules of its own; acting as a legislator, be it
for one case.
209 - In order critically to examine the reasoning of the majority award in a comprehensible
manner, rather than analyzing it statement by statement, I propose to divide the arguments
into three categories, corresponding to the three lines of the shifting strategy of justification
followed by the award in defense of its proposed course of action :
i) that it is merely filling a gap;
ii) that in any case it is introducing innocuous changes that do not affect the
substantive rights of the parties and only minimally their procedural rights;
iii) and to the limited extent to which these latter rights are affected, their limitation is
justified by a balancing of the interests of the parties.
a) - On Filling Gaps :
210 - Having declared that the power of the Tribunal under Article 44 of the Convention “is
limited to filling gaps” and does not extend to “a modification of existing rules” (para. 522),
the majority award proceeds to describe the method it proposes to fill the gap, consisting in
“the design of the specific rules... that would be necessary in order to be able to conduct the
present proceedings” (para. 526-527); “...the adaptations required relat[ing] strictly to the
manner of conducting the present proceedings” (para. 534). In other words, “it is the manner
in which the Tribunal will conduct such examination [of the claims, i.e. the controverted
rights and obligations of the Parties] which may diverge from the usual ICSID proceedings”
(para. 534).
81
211 - But if the new specific rules cover “the manner in which the Tribunal will conduct [the]
examination of the [claims]” or “the conduct of the present proceedings”, this obviously
covers most of what is usually covered by the rules of procedure, including those in the
Convention and the Arbitration Rules. And these new rules “diverge from the usual ICSID
proceedings”, i.e. Rules of Procedure (emphasis added). In other words, they set them aside
and replace them in casu.
212 - Obviously, this has nothing to do with filling gaps. First we are not speaking here of a
gap, in the current sense in rules of procedure of a “technical gap” - the odd, inadvertently
unprovided for contingency, or the small missing detail (like a cog or a screw) necessary for
implementing a rule - which can be filled by a tribunal or another implementing organ. We
are speaking here of whole chapters, segments or sections of regulation. Technically, this is
an “unauthentic” or “legislative gap”, calling not for filling a small gap, but for a complement
of the legal regulation, which needs a “legislative jurisdiction”, belonging to the organ or the
parties who adopted the original instrument to be complemented, but which lies beyond the
“adjudicative jurisdiction” of the implementing organ.
213 - Secondly, what the majority award proposes to do is not “filling a gap”, but modifying
(“diverging from”) the existing rules of procedure. A gap or lacuna is a void, an empty legal
space not covered by any rule. But here we have a complete legal regulation, covering almost
every instance of the proceedings. What the majority award proposes is to replace most of it
by another set of rules, which cannot be seen, by any stretch of imagination, as filling a void.
82
It is clearly a modification or revision of the existing rules, though limited to the current
case.
214 - The majority award tries to defend its position as follows :
“The Tribunal...can and ought to fill gaps left where the application of existing rules
are not adapted to the specific dispute submitted to ICSID arbitration. In such a case,
the filling of the gap does not consist of an amendment of the written rule itself, but
rather of an adaptation of its application in a specific case” (para. 525, emphasis
added).
215 - This statement is extraordinary in at least two ways : first, it insists on using the
language of “filling gaps”, in the presence of norms dealing with the matter, i.e. where no
gap exists, while realizing full well that the suitability of the existing rules is another matter
that raises the question of their amendment, and not of filling gaps.
216 - Whence the second extraordinary feature of this argument, the distinction between “an
amendment of the written rule itself” and “an adaptation of its application in a specific case”
by “deviating” from its prescriptions, i.e. replacing it by another “specific rule”, as
abundantly described in the majority award. This baffling distinction is reminiscent of that
other mystifying one between “conditioning... consent to ICSID jurisdiction to the fulfillment
of a condition” and “conditioning the effective implementation of such consent... to the
fulfillment of such a precondition” (discussed above, paras. 21-24).
83
217 - But no amount of sophistry and playing on words or newspeak can conceal the fact that
the proposed adaptations “diverge from the usual ICSID proceeding” (para.533), i.e. from the
existing Rules of Procedure. Nor can they transform the legal nature of this act or operation,
dubbed here as “adaptation” which does not consist of providing for an inadvertently
overlooked minor contingency or a small missing detail in an existing rule needed for its
implementation that would correspond to the meaning of “filling a gap”; but of changing or
substituting the existing rules by other “specific rules”, which clearly constitute an
amendment, modification or revision of the existing rules; an operation which lies beyond the
limited power granted to the Tribunal under Article 44 of the Convention.
218 - Nor does the fact that this so-called “adaptation” is effected only for the purposes of the
current case, change the legal nature of the act or operation. After all, the jurisdiction of the
Tribunal is limited to the case before it and its powers and even existence, do not extend
beyond that. But in the specific case before it, the Tribunal will not follow the normative
prescriptions of the written rule, but will substitute for them other normative prescriptions
that “diverge” from them (rather than merely complement something that is missing in the
normative prescriptions, which would correspond to the description of “filling a gap”). This
means that in this specific case, the existing rule or rules have been “amended”, “modified”,
“revised” or more radically discarded and totally replaced. In other words, the arbitrators
would have exercised a “legislative jurisdiction” or “power”, that lies beyond what is
provided for in Article 44 of the Convention.
219 - And if in every case or whenever the arbitrators in a specific case consider that there
84
are more suitable or appealing rules of procedure, according to their will of whim, these
arbitrators can brush aside the written rules of procedure in deciding the case before them,
what would remain of the “written rule” and its binding normative effect ? It would be
reduced to a mere recommendatory directive at the discretion of the Tribunal. It is true that
the “Arbitration Rules” are largely jus dispositivum. But they can be replaced in a specific
case only by the agreement of the parties, and not by the fiat of the Tribunal over the
objection of a party.
220 - The artificial and tenuous character of this argument of the majority award, concerning
the legal nature of the operation or act of “adaptation” (and whether it is “filling a gap” or an
“amendment” of the Rules of Procedure), explains the slippage of the award’s reasoning to a
second line of defense, namely that, in any case, the “adaptations” envisaged are innocuous,
with no or negligible consequence on the rights of the parties.
b) The Scope of the Proposed “Adaptations” and their Incidence on the Rights of the
Parties
221 - In its second line of defense or justification, the majority award starts by trying to
minimize the magnitude of the “adaptations” it proposes to undertake by such statements as:
“...the adaptations...relate strictly to the manner of conducting the present proceedings, and in
particular how to collect and weigh evidence” (para. 534; see also paras. 530,531,533).
222 - But once the rules governing the manner in which the proceedings are conducted
85
throughout a case (particularly the presentation and adversarial examination of the
controverted claims and their supporting evidence by the parties and the assessment of these
claims and the weighing of the evidence by the Tribunal) are covered by the “specific rules”
designed by the Tribunal for the particular case under consideration, how much else is left to
be governed by the regular rules of procedure ? In fact, the proposed “adaptations” cover
large expanses, if not most of what is included in any rules of procedure.
223 - Another argument in the same vein of minimizing the import of the proposed changes
or “adaptations” of the rules of procedure, is the insistence of the majority award that these
“adaptations” do not affect the merits, i.e. the controverted claims, but only the manner in
which they will be examined and the way the Tribunal will conduct the proceedings. In other
terms, the majority award contends that these “adaptations” will not affect the substantive
rights and obligations of the parties (para. 530-534). This implies, however, that by contrast
they may touch their procedural rights (which the award acknowledges, and tries to justify
later on, as discussed below). And after briefly describing the proposed “adaptations” in
terms of simplification of the marshalling and weighing of evidence and the examination of
the claims44
, the award states that “such simplification of the examination process is to be
distinguished from the failure to proceed with such examination” (para. 531).
44
[I]t is undeniable that the Tribunal will not be in a position to examine all elements and
related documents in the same way as if there were only a handful of Claimants.... [but will
resort to] a simplified verification of evidentiary material concern[ing] either the depth of
examination of a ducument... or the number of documents to be examined, and if so their
selection process...” (Para. 531).
86
224 - These affirmations and statements are properly baffling, for several reasons : first,
Article 44, which is here under discussion, deals exclusively with rules of procedure. It goes
without saying that the Tribunal has no power under this article or in general to touch the
controverted substantive rights and obligations of the parties, that constitute the “object” or
“subject-matter” of the claims before the Tribunal. The Tribunal is here to judge, not to
retouch, these rights and obligations. If what the Tribunal does on the procedural level affects
these substantive rights and obligations, such action of the Tribunal is clearly both illegal
and ultra vires.
225 - Secondly, the implication that as long as the “adaptations” by the Tribunal are confined
to “set[ting] strict boundaries to certain of the Parties’ procedural rights” (para. 538), in fact
abridging or curtailing these rights, without touching directly the Parties’ substantive rights,
such adaptation would be justified, is based on a clear fallacy. This fallacy consists of the
radical distinction between substantive and procedural rights, and the treatment of the later as
a second class category of rights at the disposal and discretion of the Tribunal (a distinction
reminiscent of the manner in which the majority award distinguishes between jurisdiction
and admissibility).
226 - This radical distinction is false because if a substantive right, or a right tout court, is
defined in the general theory of law as “an interest (or a value) protected by law”, then the
procedural rights or guarantees that back it up are an essential part of this legal protection;
their curtailment, abridgement, or violation in any way, automatically puts in serious
jeopardy the substantive rights they protect. It is a sheer aberration to try radically to separate
87
the two the way proposed by the majority award, pretending that the limitation of one (the
procedural rights) does not affect the other (the substantive rights).
227 - In the third place, the statement that the “simplification” of the examination of the
claims and the evidence should be distinguished from failure to operate such an examination
at all, is off the mark. Obviously, if the Tribunal does not examine the claims and the
evidence, it would have totally failed to exercise its judicial or arbitral function. The question
is not whether it has undertaken such an examination or not, but whether the so-called
“simplified examination” it proposes to do, meets the judicial and arbitral standards of due
process. The answer to this question depends on the extent to which the simplification affects
the procedural rights of the parties, particularly the defense rights of the Respondent.
228 - The majority award tries to minimize the impact of its proposed “simplification” of the
procedure (linguistic slippage again from the more alarming “adaptations” to the rather
virtuous “simplification”), to hide the real legal nature of the operation, i.e. “modification”.
It does so by describing briefly the effects of the “adaptation” on the respective rights of the
two parties.
229 - Starting with the Claimants, which is the easier case, the majority award states that the
adaptation would concern only “the way [they] are represented” (para. 530), acknowledging
“that TFA has been provided with powers which may go beyond the power granted to a
normal agent under Rule 18 ICSID Arbitration Rules” (para.532). But the award immediately
adds : “Admitting the present collective proceedings would thus also mean accepting TFA’s
88
role as due representative of Claimants” (para. 532).
230 - This last sentence is yet another extraordinary statement. By which leap of faith, does
the taking of jurisdiction by the Tribunal over the case, even in the form of a collective
action, (assuming that the Tribunal has jurisdiction, an assumption I refuted above), justify
and render legal and acceptable to the Tribunal the exorbitant status and powers of TFA ?
Again reasoning by necessity that is not evident.
231 - Later on, the majority award justifies the exorbitant character of “the TFA Mandate
Package [which] has the effect to depriving [sic] Claimants of a substantial part of their
procedural rights”, by the fact that it “has been consciously accepted by Claimants” (para.
546). This can be a valid justification, subject to public policy limitations, if this consent
were fully informed and not tainted by fraud or forgery; that which is ascertained by the
majority award, but which is contradicted by certain recent criminal condemnations for
forgery (of the signatures of the mandates given to TFA) by Italian courts, that the majority
award refused to take into consideration.
232 - Be that as it may, what counts most here is the effects of the new “specific rules” or
“adaptations” on the rights of the Respondent, who has contested throughout the power of the
Tribunal to introduce these “adaptations”; unlike the Claimants who have defended the
power of the Tribunal to do so, and thus can be legitimately presumed to have accepted the
incidence of these adaptations on their procedural rights, in their own self-interest.
89
233 - The “[e]ffects on Argentina’s defense rights” of the proposed “adaptations” are
assessed by the majority award as follows :
“It appears that the effect of such examination method and procedure on Argentina’s
defense rights is limited and relative. Whilst it is true that Argentina may not be able
to enter into full length and detail into the individual circumstances of each Claimant,
it is not certain that such approach is at all necessary to protect Argentina’s
procedural rights in the light of the homogeneity of Claimants’ claims”“ (para. 545,
emphasis added; see also para. 536).
234 - This minimalist representation of the effects of the “adaptations” on the rights of the
Respondent - which contrasts with their more constraining description elsewhere in the
majority award (e.g. para. 538) - calls for the following critical comments :
235 - i) Justifying the abridgement and curtailment of the procedural rights of the
Respondent, particularly as concerns the individual examination of claims (and claimants) by
the argument that the mass claims in question are sufficiently homogeneous, does not stand
scrutiny (see above, paras. 139-142).
236 - This is because it is an absolute due process right of a respondent in a judicial or
arbitral proceeding, to have every element of the claim or claims presented against him,
examined by the tribunal, through adversarial debate that affords him full opportunity to
contest and refute these elements one by one, if he can..
90
237 - This procedural right of respondents does not necessarily preempt any and all “group
examination of claims” (para. 540). For as long as the mass claims are cast into one claim
(with a mass of claimants) or are identical claims (which comes to the same thing as the
tribunal, by examining one, would have examined them all) - which is the case for example
of “class actions” - the tribunal can examine adversarially all the aspects and components of
this one claim, in spite of the multitude of the claimants, totally safeguarding the due process
rights of the respondent45
.
45
Even in class actions, as noted by the US Supreme Court in the AT&T Mobility LLC v.
Concepcion ET UX case, “class arbitration greatly increases risks to defendants” (see above para.
152). These risks are exponentially greater when the mass is composed of differentiated claims.
238 - But in the present case, the Claimants vigourously refute the characterization of their
collective mass-claims action as class action. And although these mass-claims can be
considered as arising “out of the same fact pattern”, and thus share some common features
(whence a degree of homogeneity), they are not identical and preserve some individualized
features that distinguish them from one another. To the extent that the individual claims in
the mass differ from each other, it is the absolute due process right of the defense, and the
obligation of the Tribunal, to have them examined individually and adversarially by the
Tribunal. The Tribunal has also to address and pass judgement on each of them and state the
reasons on which it bases its judgement (Article 48/3 of the ICSID Convention). Failure to
meet these requirements is a cause of annulment under the ICSID Convention (Art. 52/d).
239 - A “group examination” of mass claims that feature certain individual differences
91
among themselves, can at best deliver rough or approximate justice. It may satisfy the due
process requirements of an emergency mass claims program or an administrative
compensation commission. But it definitely falls well below the stringent due process
standard of judicial or arbitral proceedings.
240 - ii) One cannot but note the hesitant and guarded language of the above-quoted
assessment by the majority award of the effect of the “adaptations” on Argentina’s defense
rights : - “it appears” that the effect “is limited and relative”; “it is not certain that such
approach” of individual examination of differentiated claims “is at all necessary to protect
Argentina’s defense rights” (above, para. 234).
241 - This language of “appearance” (of an effect being “limited and relative”) and of
“uncertainty” (whether a due process guarantee is “at all necessary” to protect a right),
reflects an evident unease of the majority award with its own argument, and its awareness of
its frail and risky character. For good reason. For here, the majority award is treading rather
heavily on the highly sensitive due process grounds.
242 - In the first place, the individual adversarial examination of differentiated (or nonidentical)
claims is an essential part of the very “object” or “subject-matter” of the
Respondent’s procedural defense rights, and is not a mere “approach”, as the majority award
calls it, which it has decided to dispense with, because it considers that “it is uncertain” that
such an examination “is at all necessary to protect Argentina’s procedural rights”. In other
words, according to the majority award, “it is not certain” that the subject-matter of a right -
92
what it prescribes and legally guarantees for its holder - “is at all necessary” for the
protection of that same right!
243 - Apart from logical confusion, this statement of the majority award reveals something
even more serious, namely, that the Tribunal has substituted itself for the Respondent and
decided in his place the question whether it is necessary for him to exercise some of his
procedural defense rights or not. Moreover, it has answered the question on the basis of
impressions and “appearance”, almost by guesswork - “it is uncertain” - in order to reach the
result it wanted to reach.
244 - Procedural rights, the same as substantive rights, are not at the disposal and the
discretion of the Tribunal. If they are ascertained, they have to be upheld and enforced by the
Tribunal; not confiscated, scuttled, abridged or curtailed by it. But this is exactly what the
majority award has done in the present case. By imposing the “adaptation” or “simplified
procedures”, that deny Argentina an individual adversarial examination of the differentiated
mass claims, the majority award has unlawfully curtailed the Respondent’s defense rights
and thus flagrantly violated the due process arbitral standards, apart from being manifestly
ultra vires.
c) Balancing of Interests Redux
245 - The majority award tries to evade this inexorable conclusion by shifting its argument,
through yet another slippage, to a third line of defense or justification of the “adaptations”, to
wit, the balancing of the interests of the Parties. A balancing that the majority award operates
93
by comparing the consequences or effects of the “adaptations” on the rights of the Parties
“with the consequences of rejecting the claims for lack of admissibility” (para. 537), in order
to answer the question “whether it would be justified to set strict boundaries to certain of the
Parties’ procedural rights, so as to give actual effective protection to the investment” (para.
538).
246 - According to the majority award, the consequences of admitting the “adaptations”
(“special rules, “simplified examination”, etc.) on the procedural rights of the Parties are
minimal, and the outcome beneficial (“the effective protection of the investment”). The
“adaptations” pose no problem for the Claimants, who have requested and defended them all
along, and accept their consequences in their own self-interest of protecting their investment.
As far as the Respondent is concerned, according to the majority award (discussed above,
para.232ff.), “[i]t appears that the effect of such [simplified] examination... on Argentina’s
defense rights is limited and relative”, and “it is not certain” that the due process guarantees
curtailed by the “adaptations” are “at all necessary to protect Argentina’s procedural rights”
(para. 545).
247 - By contrast, the consequences of dismissing the case on grounds of inadmissibility
would be serious and prejudicial, according to the majority award. The “only alternative” left
to Claimants would be for each of them “to file an individual ICSID claim” which would be
“cost prohibitive to many claimants...[and] may equal a denial of justice” (para. 537). It
would also “be a much bigger challenge to Argentina’s [the Respondent’s] effective defense
rights...to conduct 60.000 separate proceedings... than [their] mere limitation” by the
94
“adaptations” (para. 545).
248 - On the basis of this exercise of “balancing the interests” of the Parties’ at stake,
according to its own subjective representation and evaluation of these stakes, the majority
award finds that the “adaptations”, i.e. “the procedure necessary to deal with Claimants’
claims in a collective way”, are “admissible and acceptable under Article 44 ICSID
Convention, Rule 19 Arbitration Rules, as well as under the more general spirit, object and
aim of the ICSID Convention” (para. 547).
249 - This exercise of balancing of interests calls for critical examination. It is intrinsically
wrong, lies beyond the powers of the Tribunal and is based on wrong premises, as shown in
what follows.
250 - The “balancing of interests” as operated by the majority award is intrinsically wrong
because, by making the admissibility of the “adaptations” depend on the comparison of the
consequences of their admission on the procedural rights of the Parties “with the
consequences of rejecting the claims for lack of admissibility” (para. 537), the majority
award is treating these rights as “variables” (or rather as dependant variables), whose
recognition, scope and enforceability by the Tribunal are to be determined as a function of
the Tribunal’s subjective evaluation of the right balance of interests between the parties,
which thus becomes, according to this logic, the parametric gauge of these rights. This is
standing the legal logic on its head. A tribunal is duty bound to apply the law, i.e. to enforce
rights not to put them in question, according to its own evaluation of extra-legal (or
opportunity) considerations, be they its subjective representation of the interests of the
95
Parties.
251 - Such “balancing of interests” lies beyond the powers of the Tribunal. For, as explained
earlier (para. 31), the Tribunal thus acts as if the rules do not exist, and are not themselves the
result or outcome of a balancing of interests by those who adopted them. It is at that level, the
legislative or conventional level, that the balance of interests takes place, the level of
establishing the rules, not of applying them, which is that of tribunals. It is not open to the
Tribunal to arrogate to itself the legislative jurisdiction or power of re-examining the rules in
order to revise of refashion them, in the name of a rebalancing of interests of its own,
according to its will or whim. In other words, such an exercise of “balancing of interests” is
clearly ultra vires the powers of the Tribunal.
252 - Apart from the illegality and ultra vires character of the “balancing of interests”, as
represented and operated in the majority award, substantively the exercise is based on false
premisses of law and fact.
253 - i - To say that the only alternative open to the securities holders if the present
collective action is found inadmissible would be for each of them to file an individual ICSID
claim, is wrong in fact. For as already shown (above para 33), apart from the Argentinean
tribunals, that were not even tried, all the emphasis in the securities entitlement instruments
was put on the chosen forum therein for dispute settlement, i.e. the tribunals of some of the
major financial centres of the world as the judicial guarantee of the financial investment,
beyond the reach of Argentina’s law and tribunals. ICSID arbitration was never mentioned in
96
these instruments, nor in the advertisements and the road-shows for the sale of the
instruments (as abundantly shown in the pleadings). The one judicial guarantee or recourse
that stood out was the chosen forum.
254 - Moreover, one should not forget the Italian courts which were seized by some of the
securities entitlement holders who succeeded on numerous occasions in obtaining relief
against the banks that sold them the securities. This largely explains the creation by these
very banks of the TFA and their belated discovery of ICSID arbitration as a way of deflecting
responsibility from themselves.
255 - ii - To say that dismissing the case for inadmissibility of the adaptations, “may equal a
denial of justice” is triply wrong, because it is based on three false assumptions. In the first
place, to imply that this would deprive the claimants from any efficient dispute settlement
means, is wrong as just explained.
256 - Secondly, in order to speak of a denial of justice, a tribunal must first have jurisdiction
over the case. But this Tribunal lacks jurisdiction ratione materiae over the claims (Part III
above). It lacks jurisdiction to hear collective mass claims actions, as falling outside
Argentina’s consent to arbitrate (Part IV, A to E above), and it lacks the power to revise the
Arbitration Rules in the instant case, in the absence of both Parties consent, as just discussed.
Speaking of “denial of justice” in these circumstances is out of place.
257 - Thirdly, it is also out of place and even odd in international law to speak in general of
97
an absence of a competent forum (which is not the case here, as per para 253-254 above), in
terms of “denial of justice”. This is because international law, given its a-centralized
character, has no system of courts or tribunals of plenary or general jurisdiction (juridictions
de droit commun) covering all cases and litigants, barring a specific exception attributing
jurisdiction over a particular type of matter or litigant to another specialized organ, and where
every dispute or every claim can ultimately and necessarily find a competent forum to handle
it. By contrast, as explained earlier (para 7), all international courts and tribunals are tribunals
of attributed jurisdiction (juridictions d’attribution); a jurisdiction based on the consent of the
parties or litigants and confined within the limits of this consent.
258 - iii - In fact, the majority award tries to claim the same powers as a tribunal of plenary
or general jurisdiction through interpretation by calling on the “spirit”, “aim” and/or “object
and purpose” of the ICSID Convention and the BIT, “in accordance with the principles of
interpretation of treaties” (para. 528). But the principles of treaty interpretation, as codified in
articles 31 and 32 of the Vienna Convention on the Law of Treaties of 1969, prescribe
starting with the text in its context and be guided by the object and purpose of the treaty, if
need be, to clarify any subsisting ambiguity. These principles do not prescribe nor justify, the
invocation left and right by the Tribunal, of the object and purpose, as subjectively
represented by it, in order to deduce directly from them the solutions it seeks to reach,
jumping over text, context and general rules of international law applicable to the matter.
Yet, this is exactly what the majority award does all along, as already noted (see e.g. para. 24,
157-159 above).
98
259 - In the specific context of “balancing of interests” discussed here, the majority award
sollicites the object and purpose of the ICSID Convention and the BIT, according to its
subjective and partial interpretation of them as being exclusively the effective protection of
the investment, notwithstanding the legitimate interests of the host State, in order to answer
in the affirmative the rhetorical question it put at the outset. To recall, this question was
“whether...it would be justified to set strict boundaries to [i.e. abridge] certain of the Parties’
procedural rights [but in fact only to the Respondent’s, as the Claimants requested the
“adaptations” in their self-interest]... so as to give actual effective protection to the
investment”, i.e. the Claimants (para 538). And the answer, reached through the involved
reasoning discussed above, is the award’s finding that the “adaptations” are “admissible and
acceptable under article 44 ICSID Convention, Rule 19 Arbitration Rules, as well as under
the more general spirit, object and aim of the ICSID Convention (para. 547).
260 - Put in simple words, the answer is that the Tribunal can abridge the procedural rights of
one party, the Respondent, against its will, in order to give greater (“actual effective”)
protection to the investment, i.e. to the other Party, the Claimants’ investors. The onesidedness
of this alleged balancing of interests is too glaring to need any further elaboration.
261 - This is symptomatic of a general strategy followed throughout by the majority opinion,
consisting of three stages :
i - starting with a subjective, partial and truncated representation of the object and
purpose of the ICSID Convention and the BIT, as being exclusively the effective protection
of investment, all but totally disregarding the legitimate interests of the host State;
99
ii - considering, in consequence, that all limitations on the jurisdiction and powers of
the Tribunal are obstacles in the way of achieving this object and purpose, that should be
“down-graded”, (even with untenable arguments) from “jurisdictional” to “admissibility”
issues; wrongly assuming that admissibility is less important in its function and legal effects
than jurisdiction;
iii - assuming wrongly again that questions of admissibility, are at the discretion of the
Tribunal, which can dismiss them at will as a result of its own subjective “balancing of
interests” in the light of “the more general spirit, object and aim of the ICSID Convention”
(para. 547), as interpreted by the Tribunal.
Needless to say, this strategy has very little to do with rigourous legal reasoning and
deciding by law and all the characteristics of sheer policy.
262 - My conclusion on this sub-section (IV F) in general is that the “adaptations” of the
Arbitration Rules that the majority award prescribes for hearing the present case are
manifestly ultra vires the powers of the Tribunal under Article 44 of the Convention and
Rule 19 of the Arbitration Rules.
G - Epilogue on Policy Considerations
263 - Considering what has just been said, it is ironic that the majority award dismisses
offhand the “Respondent’s arguments regarding the appropriateness of ICSID proceedings in
the context of sovereign debt restructuring” (para. 548); adding that
100
“[P]olicy reasons are for States to take into account when negotiating BITs and
consenting to ICSID jurisdiction in general, not for the Tribunal to take into account
in order to repair an inappropriately negotiated or drafted BIT” (para. 550).
264 - The majority award should have heeded its own directive, rather than read, or in effect
write, into the Convention, the BIT and the Arbitration Rules, what is not there, but what it
wishes to see there.
265 - Indeed, the majority award accepts, in full and without reasoning, the Claimant’s
contention that Argentina’s policy arguments are “outdated and irrelevant” (para. 514),
calling them - in only slightly varied terms - “irrelevant and inaccurate” (para. 429(v). It also
seemingly subscribes to the Claimant’s policy argument that “opening the door to ICSID
arbitration would create a supplementary leverage against such rogue debtors and therefore
be beneficial to the efficiency of foreign debt restructuring” (ibid.). In order to create such
leverage over sovereign debtors, which the ICSID Convention did not foresee and which
financial markets did not contemplate, then or now, the majority award fashions an
unprecedented procedure for adjudicating the present claim of more than 60.000 holders of
Argentinean debt, out of the thin air of the tribunal’s gap-filling powers under article 44 of
the ICSID Convention. Driven in part by controversial policy considerations, hidden behind
references to the spirit of the ICSID Convention and the purpose of the BIT, the majority
award blows hot and cold at the same time, uncritically adopting the Claimant’s policy
arguments over the Respondent’s, to which it hardly gave any attention.
101
266 - Up to this point, the reasoning in this dissenting opinion has kept to strict law, and this
is how a judicial or arbitral decision should be. But a judge or an arbitrator, particularly an
international one, cannot be totally blind to the social, economic and political environment
which constitutes the larger context of the case. Policy considerations should never be
determinative of a judicial or arbitral decision. But, within the permissible margin of
interpretation, they can shed light on what makes sense or nonsense among possible
alternative solutions, when seen against the larger background.
267 - In this respect, the following three major policy issues or queries, collaterally raised
and necessarily affected by any decision taken in this case, should have been given more
attention:
268 - i - As noted earlier (para. 35), this is the first ICSID case that involves a sovereign
debt financial instrument (but I leave the sovereign debt aspect to the following point), that is
totally unhinged and detached from any specific economic activity or project in the host
country; in other words, completely detached from any form of investment in the economic
sense of the term, which is also that of ICSID Convention. Affirming the jurisdiction of
ICSID Tribunals over such instruments, would extend it over a vast new field. It would cover
virtually all capital market transactions, ranging from standardized financial instruments,
such as shares and bonds to structured and derivative products, such as hedges and credit
default swaps. It would thus open the way to converting them from specialized tribunals,
dealing with disputes arising out of a special type of investment, into commercial tribunals of
general jurisdiction, covering all manners of financial transactions, including the most
102
speculative varieties, which have nothing to do, in fact are light years away from the
economic investment for the encouragement of which the ICSID Convention was concluded.
Is this the direction for ICSID tribunals to follow?
269 - ii - This is also the first ICSID case, to my knowledge, that involves sovereign debt. It
raises prima facie a crucial question : do ICSID tribunals have jurisdiction over sovereign
debt instruments issued internationally, expressed in foreign currency and payable abroad,
governed by various external laws and subject to the jurisdiction of various external courts,
and traded as dematerialised security entitlements in global capital markets ?
270 - The majority award evades addressing this question frontally and on its merits, by
purportedly deciding the issue on another ground, i.e. finding that these are financial
instruments that are, as such, covered by the BIT (a finding refuted in Section III above). But
evading the question does not dispose of it. For it remains central not only for the
jurisdictional reach of ICSID arbitration in this particular case, but in relation to financial
market transactions in general, as discussed in the previous point. The answer also matters
greatly for the larger problem of sovereign debt crisis management, i.e. for the manner in
which the international community and countries borrowing abroad will resolve the present
and future sovereign debt crises and how the burden for such crises will be shared between
taxpayers and creditors; a perennial challenge that is now occupying the daily headlines and
confronting countries at all levels of development.
103
271 - Indeed, in view of the actual profound structural crisis of the international financial
system; the absence of agreed international procedures regulating State bankruptcy; and the
intense international discussions and efforts to improve the sovereign debt restructuring
process, the present case raises, in an accute manner, an international public policy issue
about the workability of future sovereign debt restructuring, should ICSID tribunals intervene
in sovereign debt disputes. It suffices to ponder the potential disrupting effect of different ad
hoc tribunals following separate ways or deciding at cross purposes with the desperate
international efforts to reconstruct a semblance of a coherent international financial
architecture. Such decisions can also potentially unravel patiently negotiated settlements
(through the effect of the most favoured creditor clause inserted in most settlements).
Ramifications that are depicted in a recent article as “opening Pandora’s box” for sovereign
debt46
.
Is this a proper role for ICSID’s tribunals ?
272 - iii - Finally, the third policy consideration is a cautionary one. It is to caution against
the tendency of certain ICSID tribunals to consider any limitation on their jurisdiction whether
inherent in the adjudicative function or carefully negotiated and stipulated in the
ICSID Convention or the BITs, to protect the legitimate interests of State parties - as an
obstacle in the way of achieving the object and purpose of these treaties, which they interpret
as being exclusively to afford maximum protection to investment, notwithstanding the
legitimate interests of the host State.
46
Michael Waibel, supra note 8.
273 - Viewed as such, these limitations have to be disposed of, according to the tribunals in
104
question, at any price, be it the proper interpretation of the law. This leads to tenuous and
untenable interpretations, particularly of jurisdictional titles, over-stretching the text beyond
the breaking point, in order to extend jurisdiction to where it does not exist, particularly
beyond its underlying consent.
274 - This misguided tendency, purporting to serve progress by expanding the rule of law
and the role of adjudication in international society, is the worst disservice that can be
rendered to the very positive movement that has characterized the last two decades towards
increasing resort to adjudication and the multiplication of adjudicative fora on the
international level. It is undermining the credibility not only of the ICSID system, but of the
very idea of objective international adjudication, by eroding the confidence of States, whose
consent remains the basis of jurisdiction, in the objectivity and good judgement of those to
whom they may entrust the ascertainment of their rights. The risk of a back-lash is already
pointing its head, it threatens to stop net, if not roll back the movement of progressive
judicialisation of international law.
What is the proper and prudential course for ICSID tribunals to take in this regard?
V - Conclusion
275 - In conclusion, and while reserving my position on the other preliminary objections
raised by Argentina, I consider and find that the first and ninth objections are valid and
105
should have been accepted and given effect by the Tribunal.
276 - In consequence, I consider and find
a) that this Tribunal has no jurisdiction ratione materiae over the sovereign debt
instruments (security entitlements in Argentinean bonds) that are at the basis of the mass
claims before the Tribunal, because they do not constitute a covered or protected investment
under the ICSID Convention and the BIT, in the absence of a territorial link with Argentina;
b) that this Tribunal has no jurisdiction under the ICSID Convention and the BIT over
the present collective mass claims action, absent the consent of Argentina, nor does it have
the power to devise new procedures to hear such an action.
277 - For these reasons, and with all due respect for my co-arbitrators, I cannot concur in the
majority award and enter this dissenting opinion.
[signed]
Georges Abi-Saab