Choice of Law and Interpretive Authority in Investor-State Arbitration
| Author | Joshua Karton |
| Position | Associate Professor and Associate Dean for Graduate Studies and Research, Faculty of Law, Queen's University |
| Pages | 217-268 |
217
(2017) 3(1) CJCCL
Choice of Law and Interpretive
Authority in Investor-State
Arbitration
Joshua Karton*
is article rejoins one of the core debates in investor-state arbitration, over the extent to
which arbitrators may refer to sources of international law beyond the investment treaty that
governs the dispute. is issue may appear esoteric, but the political backlash to investment
treaty arbitration is largely fueled by uncertainty over the content of the substantive rules
that bind states in their relations with foreign investors. Such uncertainty aords arbitrators
room to indulge what is alleged to be a pro-investor bias. It may chill regulatory initiatives,
even if in the end most states’ actions are vindicated. e problem at the heart of investment
arbitration is, therefore, a legal one, so there may be a legal response to the political backlash.
is article argues that arbitrators are obligated by the choice of law clauses contained in
most investment treaties to consider all potentially relevant sources of international law.
Arbitrators are akin to agents of the states that enter into investment treaties, and are bound
by choice of law provisions in those treaties. Since most of these refer simply to the text of
the treaty and “international law”, tribunals not only may but must refer to international
law beyond the treaty. Putting choice of law at the centre of determinations of tribunals’
interpretive authority refocuses arbitrators’ attention on states, which are, after all, the parties
to the arbitration agreements that empower investor-state tribunals. It gives proper weight
to the economic objectives of international investment law, but also provides arbitrators
with an appropriate basis on which to account for the public interest, via international law
doctrines of environmental protection, indigenous rights, and the like. Finally, it could help
stave o a continued backlash to investor-state arbitration, which would harm the global
investment climate and the global rule of law.
* Associate Professor and Associate Dean for Graduate Studies and Research,
Queen’s University Faculty of Law. BA, Yale (2001); JD, Columbia (2005); PhD,
Cambridge (2011). is paper is part of a larger research project generously
supported by the Social Sciences and Humanities Research Council of Canada and
the Queen’s University Senate Advisory Research Committee. Some of the concepts
incorporated into this paper were previously presented in various fora. I am
grateful for feedback received from participants at the 2015 ASIL-ILA Asia-Pacic
Research Forum, the 2015 Asian Law Institute Annual Conference, the 2014
Canadian Council on International Law Annual Conference, and the 2014 Junior
International Law Scholars Association Annual Meeting. Dilton Ribeiro provided
invaluable research assistance on matters of general international law.
218
Karton, Law and Interpretive Authority in Investor-State Arbitration
I. I: R “B” I A
II. U “B”: U G L
III. T S A A I A
S
IV. I G L I T A
V. I P A
A. States in the Driver’s Seat
B. In Most Cases, International Investment Agreements Should be
Interpreted Extensively
VI. C
I. Introduction: Reconceiving the “Backlash” to
Investment Arbitration
Arbitration of investment treaty disputes is in the news a lot these
days, and usually because someone new is denouncing it. Most
prominently, the proposed Trans-Pacic Partnership (“TPP”) and Trans-
Atlantic Trade and Investment Partnership (“TTIP”) are probably now
dead letters, attacked most bitterly in the places that pioneered – and
historically benetted the most from – investment treaty arbitration: the
USA and Western Europe. Recent political events, especially the Brexit
vote and the election of Donald Trump to the US presidency, indicate
that new investment treaties are less likely to be ratied, especially if
incorporated into multilateral trade conventions.
It is therefore all the more important that the existing system of
investor-state dispute settlement (ISDS) be made to work better for
its purported beneciaries: the people of the states that engage in it.1
Investment treaty arbitration, along with other means for peaceful
1. Cf. Ingo Venzke, “Investor-State Dispute Settlement in TTIP from
the Perspective of a Public Law eory of International Adjudication”
(2016) 7:3 Journal of World Investment & Trade 374 (arguing that “the
architects of TTIP as well as the critics of this edice seem to share a core
point – the demand, namely, that the law be spoken in the name of the
peoples and citizens” at 380).
219
(2017) 3(1) CJCCL
resolution of disputes between investors and the states that host their
investments, has the potential to stimulate foreign direct investment
(FDI) and promote the rule of law. Investment treaty arbitration is
not now living up to that potential. However, should states abandon it
because of its aws, they may also miss out on its benets.
is article advances a new way of thinking about the causes of
hostility to arbitration between investors and states, then proposes a new
way of thinking about how to improve the quality and consistency of the
justice provided by investor-state arbitration without making structural
changes to the ISDS system. e backlash to investor-state arbitration is
driven by political concerns, but there may be a legal response to it.
One of the core debates about investor-state arbitration concerns the
extent to which arbitrators may refer sources of international law outside
the text of the investment treaty that governs the dispute. is is essentially
a question of interpretation: to what sources may adjudicators refer when
clarifying ambiguities and lling gaps in the treaties, statutes, and contracts
that govern dierent aspects of an investor-state dispute, and how should
they resolve conicts between these sources. Some commentators argue
that investment arbitration tribunals should decide, to the extent possible,
within the text of the governing treaty and the investment contract –
“restrictive interpretation”.2 Others argue that general international law
is relevant both to interpret the relevant treaty and to introduce doctrines
not referred to in the treaty, including principles developed in treaties
and case law outside the investment context, in particular international
2. is view is perhaps most associated with Charles Brower. See e.g.
Charles N Brower & Sadie Blanchard, “From ‘Dealing in Virtue’ to
‘Proting from Injustice’: e Case Against Re-Statication of Investment
Dispute Settlement” (2014) 55:45 Harvard International Law Journal
Online, online:
Brower_Blanchard_to_Publish.pdf>; Charles N Brower & Shashank P
Kumar, “Investomercial Arbitration: Whence Cometh It? What Is It?
Whither Goeth It?” (2014) 30 ICSID Review 35; and Charles N Brower
& Sadie Blanchard, “What’s in a Meme? e Truth about Investor-State
Arbitration: Why It Need Not, and Must Not, Be Repossessed by States”
(2014) 52:3 Columbia Journal of Transnational Law 689.
220
Karton, Law and Interpretive Authority in Investor-State Arbitration
trade and human rights law – “extensive interpretation”.3 A third school
of thought emphasizes the public law nature of investment disputes,
arguing that tribunals determining whether a state has breached its
obligations to a foreign investor ought to take a public law approach
that emphasizes national rather than international law.4 ese threads in
the academic discourse are connected to more fundamental discussions
about the place of investment law within international law, which are in
turn related to broader debates over fragmentation and convergence in
international law.5
Such matters may appear to be esoteric, of concern only to academics
and policy wonks. In fact, the theoretical and political debates over
investment treaty arbitration are both largely fueled by the same basic
concern: uncertainty over the content of the substantive rules that bind
states in their relations with foreign investors. e substantive provisions
of most of the relevant instruments are incomplete and vaguely worded,
and it is hotly contested how arbitrators should ll these gaps and resolve
these ambiguities
e gaps and ambiguities create a zone of discretion for arbitrators that
many see as too broad – an issue that aects international law generally,
3. See e.g. Pierre-Marie Dupuy et al, eds, Human Rights in International
Investment Law and Arbitration (Oxford: Oxford University Press, 2009).
4. See generally Stephan W Schill, ed, International Investment Law and
Comparative Public Law (Oxford: Oxford University Press 2010); see
also Stephan W Schill, “Enhancing International Investment Law’s
Legitimacy: Conceptual and Methodological Foundations of a New
Public Law Approach” (2011) 52 Virginia Journal of International Law
57 at 68-71, 81; Benedict Kingsbury & Stephan W Schill, “Public Law
Concepts to Balance Investors’ Rights with State Regulatory Actions
in the Public Interest – e Concept of Proportionality” in Stephan W
Schill, ed, International Investment Law and Comparative Public Law
(Oxford: Oxford University Press, 2010) at 75.
5. See generally Freya Baetens, ed, Investment Law Within International
Law: Integrationist Perspectives, (Cambridge: Cambridge University Press,
2013).
221
(2017) 3(1) CJCCL
but is particularly acute in international investment law.6 e breadth of
arbitral discretion gives arbitrators space to indulge what is alleged to be
a pro-investor bias and permits tribunals to overrule reasonable attempts
by states to regulate commerce in the public interest.7 e uncertainty
that results may chill regulatory initiatives, even if states’ actions are
vindicated by arbitral awards in the end.8 e obscurity of the process
(and, secrecy, although less than in the past) add to the sense that some
kind of scam is being run.
e problem at the heart of ISDS is therefore a legal one, which means
that there may be a legal response to the political backlash. Investor-
state tribunals must recognize that their interpretive role in resolving
individual disputes implies a more fundamental role as guardians of the
coherence of investment law itself.9 To play this role properly, tribunals
must adopt a coherent interpretive approach. is article argues that,
when interpreting investment treaties, arbitrators are not only permitted
to consider sources of international law from outside investment treaties,
but are in most cases obligated to do by the choice of law provisions
in those treaties. Arbitral tribunals are akin to agents of the states that
enter into investment treaties, and must follow choice of law provisions
in investment treaties. In most cases, the law governing investor-state
6. Gleider Hernandez, “Interpretive Authority and the International
Judiciary” in Andrea Bianchi et al, eds, Interpretation in International Law
(Oxford: Oxford University Press, 2015) at 167.
7. Many of these charges are collected in a report published by the Corporate
Europe Observatory and the Transnational Institute. See Pia Eberhardt &
Cecilia Olivet, “Proting from Injustice: How law rms, arbitrators, and
nanciers are fuelling an investment arbitration boom” (November 2012),
online: .
8. Jürgen Kurtz, “Building Legitimacy rough Interpretation in Investor-
State Arbitration: On Consistency, Coherence, and the Identication
of Applicable Law” in Zachary Douglas et al, eds, e Foundations
of International Investment Law (Oxford: Oxford University Press,
2014) (“[e]ven with an outcome ledger that tilts generally in favour of
respondents, the deep variances in the jurisprudence make it almost
impossible for states to isolate when and why particular regulatory
initiatives might potentially engage investment treaty liability” at 270).
9. Hernandez, supra note 6 at 167-68.
222
Karton, Law and Interpretive Authority in Investor-State Arbitration
arbitrations is simply the text of the treaty and “international law”,10
which means that arbitral tribunals not only may but must refer to
international law outside the treaty.
is article is premised on the current ISDS system remaining
roughly in its current form. Here, I take no position on whether or
how the system ought to change structurally. Rather, my aim is to show
that a greater attention to the governing law would improve arbitral
decision-making without any structural changes – and moreover that the
approach I advocate is dictated by the structure of the existing system of
investor-state arbitrations. My line of argument is theoretical rather than
pragmatic, although I believe that it will also yield practical benets.
e proposed interpretive approach gives proper weight to the
economic objectives of international investment law, but also provides
arbitrators with an appropriate legal basis on which to account for the
public interest, in the form of international law doctrines of environmental
protection, indigenous rights, and the like. It also refocuses arbitrators’
attention on states, which are, after all, the parties to the arbitration
agreements that empower investor-state tribunals. Employment of a
coherent interpretive approach that pays attention to the choice of law
would help improve the quality and consistency of arbitral decision-
making and, in turn, promote buy-in from governments and the
populations they represent. It would also rearm investment law’s
place within international law and promote cross-fertilization between
investment law and other international legal disciplines.
Part II provides background, explaining how many of the criticisms
of investment treaty arbitration are rooted in uncertainty over how
the governing law is interpreted and applied. Part III presents the
core theoretical argument: that the structure of arbitral authority in
international investment law requires arbitrators to consult the choice
of law provision in the applicable investment agreement in order to
determine not only which laws to apply, but also how to interpret them.
Part IV applies those theoretical arguments to describe, in the abstract,
10. See infra notes 52-57 and accompanying text (this is something of an
oversimplication).
223
(2017) 3(1) CJCCL
how investment treaty tribunals ought to proceed. Finally, Part V explains
the main implications of the proposed approach, making prescriptions
about how states may act to shape arbitrators’ interpretive authority and
how arbitrators ought to interpret IIAs in the majority of cases where
the treaty contains an unqualied choice of “international law” as the
governing law.
Before proceeding, a brief note on terminology is needed, since
authors in this eld sometimes use the same terms to refer to dierent
things and dierent terms to refer to the same things. I will refer to the
overall system of resolving disputes between investors and the states
in which they invest as “investor-state dispute settlement” (ISDS)
and the heterogeneous body of rules relating to the international law
obligations of states to foreign investors as international investment
law (IIL). As the label for the main legal instruments of that system, I
use “international investment agreements” (IIAs), which describes any
agreement between states that has the purpose and eect of protecting
cross-border investments, whether those agreements are bilateral or
multilateral, whether they deal specically with investment or with trade
more generally, and whether they are memorialized in a treaty or in some
other form. Finally, to describe the primary means of resolving disputes
between investors and states that relate to IIAs, I will use “investor-state
arbitration” (ISA), although that term might conceivably also refer to
arbitrations between investors and states that do not arise from IIAs.
II. Understanding the “Backlash”: Uncertainty over
the Governing Law
It is now de rigueur to call political hostility to ISA the “backlash” against
investment arbitration. It is an appropriate term, since it captures the
reactive nature of much criticism of ISA, especially from politicians
and civil society. For many years ISA proceeded without opposition,
primarily because the general public and even most legislators had no
idea it existed. Since the term “backlash” was introduced to the literature
224
Karton, Law and Interpretive Authority in Investor-State Arbitration
in 2010,11 hostility to ISDS, and in particular to ISA, has only grown.
e components of this backlash are various. Some critics are
concerned about state sovereignty, some about environmental or human
rights protection, some about inconsistent outcomes, some about
democratic accountability, some about transparency, some about bias.
But many of the critiques of ISDS derive in part from one common
factor: the content of rules that will be applied to determine the merits
of investor-state disputes.
e problem of inconsistency is most obviously traceable to
uncertainty in the governing law, but other critiques can also be
characterized in terms of uncertainty. Given that state liability through
ISA is voluntary, it is not clear how much ISA could possibly “rob” states
of their sovereignty, but it is fair to argue that states have given up more
of their sovereignty than they realized when they ratied an IIA. Greater
certainty would at least make possible more informed decisions by states
on whether to enter into IIAs, how to draft them, whether to make
interpretive pronouncements after the entry into force of an IIA, and
how to pose arguments to a tribunal once a dispute arises.
Similarly, concerns about the ability of states to regulate in the public
interest despite their IIL obligations are really concerns about the content
of the governing law. How do the obligations created by IIAs relate to
countervailing principles of domestic public law? Are IIL obligations
supplemented or limited by substantive obligations created by other
areas of international law, in particular human rights law? Finally, how
do international law doctrines relating to the force of international
obligations, such as the law on state responsibility and doctrines like
proportionality and the margin of appreciation, aect states’ obligations
under IIL?
Of course, some of the critiques are not about the law being applied,
but rather about who applies it. One strain of critique emphasizes the
private character of arbitral tribunals, and argues that ISA tribunals in
particular are populated by business lawyers who are subjectively biased
11. Michael Waibel et al, eds, e Backlash Against Investment Arbitration
(Alphen aan den Rijn: Kluwer, 2010).
225
(2017) 3(1) CJCCL
in favour of investors and who have material incentives to take an
expansive attitude toward their own jurisdiction and the obligations of
states to investors. Regardless of the accuracy of these charges, much of
their force would be reduced if the content of IIL were more certain – the
more concrete and precise the applicable rules are, the less room there is
for adjudicator bias to aect outcomes.
Uncertainty over the governing law has a variety of causes, many
of them not resolvable without altering the nature of international law
or the structure of the ISDS system. Perhaps most importantly, IIL is
expressed in thousands of dierent IIAs concluded by dierent states
using dierent language; unlike other elds of international law, such as
the law of the sea or international trade law, IIL has no common treaty
or set of treaties that sets out the substantive obligations of states. e
closest thing to a canonical treaty is the Convention on the Settlement of
Investment Disputes Between States and Nationals of Other States12 (“ICSID
Convention”), which deals only with the means by which disputes over
IIL are to be resolved, and in any event applies only to a subset of investor-
state disputes. But even when one looks beyond the various IIAs, public
international law continues to have an inherent indeterminacy greater
than any national law, something that no one ISDS tribunal can resolve.
Even the status of international law as “law” continues to be contested
in some quarters (although much less so than in the past). Relatedly, the
youth of IIL as a distinct eld of law means that many of its details remain
to be worked out, simply because the issues have only arisen recently.
Moreover, the lack of any centralized legislative or judicial authority
with the power to pronounce on matters of IIL necessarily slows the
progressive development of the law, as is the fact that interpreting IIL is
largely left to ad hoc arbitral tribunals that recognize neither an adjudicative
hierarchy nor any doctrine of binding precedent. Sociologically, one
might add that the community of international investment lawyers – and
more specically of ISDS arbitrators, do not share a common professional
background or legal culture, and in fact may be divided between two
12. March 1965, 575UNTS 159 (entered into force 14 October 1966)
[ICSID Convention] (also called the “Washington Convention”).
226
Karton, Law and Interpretive Authority in Investor-State Arbitration
camps of commercial lawyers and public international lawyers. Finally,
the parties to ISAs – investors and capital-importing states – are often
repeat players who have irreconcilably opposed interests.
But regardless of the causes of legal uncertainty, many of them would
be resolved or deprived of their signicance by a more consistent approach
to the interpretation of the governing legal instruments. Writes Kurtz,
“Ultimately … it is the coherence and integrity of reasoning employed
by arbitral tribunals that is of greatest import to states parties (with
the highest potential to foster deeper commitment to the system)”.13
Unfortunately, ISA tribunals have not met this challenge. “[T]here is
a distinct and peculiar ‘moving target’ quality to the hermeneutics of
investment arbitration with arbitral tribunals often paying simple lip
service to the customary rules on treaty interpretation”.14
e decisions of ISA tribunals should not and will never be entirely
consistent, given the variety of dierently-worded IIAs that apply in
various ISAs, the dierent national laws that may apply to some aspects
of disputes, and the range of legitimate opinions on legal questions that
arise in disparate ISAs. However, the impossibility and undesirability of
consistent outcomes in ISDS should not make us give up on a consistent
interpretive approach. In the next section, I argue that ISA tribunals are
obligated by the structure of arbitral authority in ISDS to follow such a
consistent interpretive approach.
III. e Structure of Arbitral Authority in the
Investment Arbitration System
It is a common misconception that ISA tribunals have broad inherent
discretion with respect to the governing law, for example to interpret IIAs
in a restrictive or extensive manner, and to consider or reject principles
of international law developed outside the investment context. In fact,
while in a given case the tribunal may have such discretion, whether it
does or does not depends on the terms of the choice of law provision
in the relevant IIA. In other words, when states conclude an IIA, they
13. Kurtz, supra note 8 at 258.
14. Ibid at 275.
227
(2017) 3(1) CJCCL
have the power to determine not only the set of rules that tribunals must
apply, but also the way in which tribunals must interpret them.
e legal framework that supports ISA imposes very few constraints
on the way arbitrators are to interpret the governing law. Rather, arbitral
authority to interpret law, along with any restrictions on the exercise of
that authority, comes from agreement that empowers the tribunal. For
this reason, regardless of its public international law context, ISA has
an inherently contractarian character, an inheritance of the international
commercial arbitration models on which ISA jurisdictional and procedural
rules are based. In IIL, the agreement that empowers the tribunal is
usually contained in the applicable IIA. According to the “triangular”
nature of ISA, when states ratify an IIA, they make an “open oer” to
arbitrate. is oer may be accepted by any investor from another state
party simply by ling a request for arbitration, even if the investor lacks
a pre-existing legal relationship with the state.15 e terms of the oer
to arbitrate are specied in the IIA, and initiation of arbitration by the
investor constitutes the investor’s acceptance of those terms.
As will be seen, questions about the scope of the tribunal’s powers
and duties with respect to the decision on the merits cannot normally be
answered by reference to general law, but rather according to the terms of
the applicable IIA, and in particular the choice of law provision within it,
along with and any other agreements entered into between the host state
and the investor.
I will begin by examining the provisions of the treaties that form the
framework of the ISDS system, but do not themselves contain arbitration
agreements: the ICSID Convention16 and the United Nations Convention
on the Recognition and Enforcement of Foreign Arbitral Awards17 (“New
15. Julian Davis Mortenson, “Treaty Interpretation in International
Investment Law” in Michael Bowman & Dino Krisiotis, eds, Conceptual
and Contextual Perspectives on the Modern Law of Treaties (Cambridge:
Cambridge University Press, forthcoming 2017) at 2, online:
abstract=2757690/>.
16. ICSID Convention, supra note 12.
17. 10 June 1958, 330UNTS 38 (entered into force 7 June 1959) [New York
Convention].
228
Karton, Law and Interpretive Authority in Investor-State Arbitration
York Convention”). e former applies to all investor-state arbitrations
conducted under the auspices of ICSID,18 and the latter applies to nearly
all investor-state arbitrations conducted ad hoc (which are typically
governed by the UNCITRAL Arbitration Rules) or under the auspices of
other international arbitral institutions.19
e ICSID Convention and New York Convention do not directly
bind arbitrators or arbitrants; rather, they are directed at national courts,
which may be called upon to rule on the enforceability of arbitration
agreements and arbitration awards. By design, the two conventions say
next-to-nothing about the actual decisions made by arbitrators. For this
reason, and with a few exceptions that will be discussed below, they do
not regulate the decisions arbitrators make on the merits of disputes, or
even the rules of decision which arbitrators must apply, but only the way
in which arbitrators reach those decisions.
Under New York Convention Article V, which governs the enforcement
of arbitral awards, the grounds for refusal of enforcement are generally
jurisdictional and procedural. Only two provisions might conceivably be
engaged by an inapposite or inaccurate application of the governing law.
e rst is Article V(2)(b), the public policy exception, which permits
non-enforcement of an award only in narrow circumstances where
enforcement would violate the fundamental public policy of the state.20
To the author’s knowledge, no commercial or investor-state arbitral
award has ever been refused enforcement under Article V(2)(b) on the
ground that the tribunal misinterpreted the governing law.
18. See International Center for Settlement of Investment Disputes,
“Recognition and Enforcement – Additional Facility Arbitration”, online:
(AF-Arbitration).aspx>. e New York Convention also applies to the
recognition and enforcement of awards not subject to the ICSID
Convention that are administered by ICSID under the ICSID Additional
Facility Rules.
19. As a general matter, the New York Convention applies equally to
commercial and investor-state arbitration. Awards that arise from IIAs
involving non-parties to the New York Convention, such as Taiwan, will
not be subject to it.
20. New York Convention, supra note 17, art V(2)(b).
229
(2017) 3(1) CJCCL
e second is Article V(1)(d), which provides that awards may be
refused enforcement where the arbitral procedure was not in accordance
with the agreement of the parties. In extreme cases where the tribunal
blatantly applied a law dierent from the one chosen by the parties or
where the tribunal’s reasons display a total lack of legal reasoning, courts
have held that the tribunal’s actions constituted a procedural defect and
refused enforcement on that basis.21
To the extent that arbitrators make errors of law or reach a decision
other than by application of the governing law, courts will not normally
interfere unless those errors were so egregious as to constitute arbitrator
misconduct (harming a party’s due process rights or other otherwise
violating public policy), the award blatantly applies a dierent law than
the law chosen by the parties, or the award so disregards all legal rules that
the tribunal can be said to have arrogated to itself amiable composition
powers.22 If arbitrators apply the governing law incompetently, or if
they pay lip service to the law while actually deciding on some other
basis, the award is generally proof from challenge. Under New York
Convention Article V and most national laws, including those based on
the UNCITRAL Model Law on International Commercial Arbitration,
21. Only in a handful of commercial cases subject to the New York Convention
have awards been annulled or refused enforcement on this basis. e
case law is reviewed and discussed in James Hope & Mattias Rosengren,
“Arbitrators: a law unto themselves?” (3 December 2013), Commercial
Dispute Resolution, online:
arbitrators:-a-law-unto-themselves>.
22. Cf. Jan H Dalhuisen, “Legal Reasoning and Powers of International
Arbitrators” (2014), online:
id=2393705/> (“[t]here are minimum standards but they are few and only
geared to avoiding clear excess” at 22).
230
Karton, Law and Interpretive Authority in Investor-State Arbitration
there is no recourse against a legally incorrect award.23
For arbitrations subject to the ICSID Convention, which includes
most ISAs, the rules dier but the outcome is the same. e ICSID
Convention itself imposes only one duty directly upon tribunals relating
to how they should decide the merits of the dispute: the requirement in
Article 42(2) that tribunals may not bring a nding of non liquet on the
ground of silence or obscurity in the law. Arguably, this provision imposes
an obligation to decide legally, that is, in accordance with legal rules, but
it does no more than this. In any event, it says nothing about which rules
should be applied or how they should be interpreted. Moreover, Article
42(3) makes clear that the parties may empower the tribunal to decide
ex aequo et bono, so it is not even compulsory for ICSID tribunals decide
according to legal rules.24
Article 42(1) contains the ICSID Convention’s main rules as to the
governing law:
e Tribunal shall decide a dispute in accordance with such rules of law as
may be agreed by the parties. In the absence of such agreement, the Tribunal
shall apply the law of the Contracting State party to the dispute (including
its rules on the conict of laws) and such rules of international law as may be
applicable.25
According to the rst sentence of Article 42(1), the parties may make
a choice of law and, if they do so, the tribunal must apply that law.
In other words, the parties have absolute freedom to choose any rules
23. Courts frequently rearm this principle; see e.g. TCL Air Conditioner
(Zhongshan) Co Ltd v e Judges of the Federal Court of Australia [2013]
HCA 5. I leave aside whether parties may agree to confer upon courts (of
the seat or otherwise) the power to review awards for substantive errors
of law. State courts have divided on the validity of so-called heightened
judicial review agreements; see, e.g. the divergent decisions of the US and
German Supreme Courts: Hall St Assocs v Mattel, 552 US 576(2008);
BGH III ZB 07/06, 1 March 2007.
24. See generally Christoph Schreuer, “Decisions Ex Aequo et Bono Under
the ICSID Convention” (1996) 11:1 ICSID Review 37. Determination
ex aequo et bono permits adjudicators the greatest possible latitude to
consider justice and fairness without the need to resort to any rules of law;
it is quite rare in practice.
25. ICSID Convention, supra note 12, art 42(1).
231
(2017) 3(1) CJCCL
of law and their choice is binding upon the tribunal. If the parties do
not make a choice, the second sentence of Article 42(1) provides that
the tribunal must apply the law of the respondent state, together with
whatever rules of international law are applicable. is default provision
does require tribunals to apply the named laws, but it does nothing to
guide the tribunal’s interpretation of national or international law, beyond
specifying that a state’s law includes its rules on the conict of laws.
Most importantly for the purposes of this article, it implicitly delegates
to tribunals the determination of which rules of international law are
applicable and places no constraints whatsoever upon that determination.
Awards subject to the ICSID Convention are even more broadly
enforceable than those subject to the New York Convention. Under Article
54, contracting states “shall recognize [awards] rendered pursuant to” the
Convention, and must enforce “the pecuniary obligations imposed by
that award … as if it were a nal judgment of a court in that State”.26
e only exception given in the ICSID Convention itself is the statement
in Article 55 that national laws relating to sovereign immunity are not
aected by the Convention. States may also refuse to enforce ICSID
awards that violate the fundamental public policy of the state, but to
the author’s knowledge, as with New York Convention Article V(2)(b),
no ICSID award has been refused enforcement on public policy grounds
because the tribunal misinterpreted the law.
ICSID awards are also subject to annulment by a three-member
ad hoc annulment committee constituted for that specic purpose.27
Annulled awards have no force, so parties unwilling to treat the dispute
as ended must request the constitution of a new tribunal.28 e ICSID
Convention lists only ve grounds on which an award may be annulled, of
which only two are potentially relevant here: that the tribunal “manifestly
exceeded its powers”29 or that the award fails to “state the reasons on
which it is based”.30 An examination of the way these provisions have
26. Ibid, art 54(1).
27. Ibid, art 52.
28. Ibid, art 52(6).
29. Ibid, art 52(1)(b).
30. Ibid, art 52(1)(e).
232
Karton, Law and Interpretive Authority in Investor-State Arbitration
been applied by annulment committees shows that each does little or
nothing to restrain tribunals from interpreting the governing law in
idiosyncratic or even incorrect ways.
In some early annulment decisions – the so-called “rst generation”,
comprising Klöckner v Cameroon I31 and Amco v Indonesia I32 – the
annulment committees closely scrutinized the tribunals’ interpretation
of the governing law and application of that law to the facts of the
dispute, reasoning that a failure to accurately apply the governing law can
constitute either (or both) an excess of powers or a failure to state reasons.33
However, both annulment committees were heavily criticized for these
decisions.34 e modern annulment decisions, and the overwhelming
weight of commentary, hold that tribunals only manifestly exceed their
powers related to application of the governing law if they fail to apply
the chosen law altogether or blatantly apply a dierent law.35 If a tribunal
defectively or incompetently applies the chosen law, the award is proof
from annulment. Similarly, the modern position is that a tribunal has
failed to give reasons only where the annulment committee is unable to
follow the tribunal’s reasoning or see how it relates to the issues before
31. Klöckner Industrie-Anlagen GmbH and others v United Republic of
Cameroon and Société Camerounaise des Engrais II (1985), ICSID Case No
ARB/81/2, 2 ICSID Reports 95 [Klöckner I].
32. Amco Asia Co and others v Republic of Indonesia (1986), ICSID Case No
ARB/81/1, 1 ICSID Reports 509 [Amco I].
33. See Christoph Schreuer, “ree Generations of ICSID Annulment
Proceedings” in Emanuel Gaillard & Yas Banifatemi, eds, Annulment of
ICSID Awards, IAI International Arbitration Series No 1 (New York:
Juris, 2004) at 17.
34. Vladimír Balaš, “Review of Awards” in Peter Muchlinski et al, eds, Oxford
Handbook of International Investment Law (Oxford: Oxford University
Press, 2008) at 1148-49.
35. See e.g. Malaysian Historical Salvors Sdn BHD v Malaysia (2007) ICSID
Case No ARB/05/10. e decision of the annulment committee criticizes
the award’s failure to examine the express terms of the IIA as applicable
law, and instead to decide on the basis of default rules in the ICSID
Convention. e annulment committee held that such failure constituted
a manifest excess of powers requiring annulment under the ICSID
Convention. Ibid at para 80.
233
(2017) 3(1) CJCCL
the tribunal.36 Whether the reasons given by the tribunal are correct,
adequate, or convincing is irrelevant.37
In sum, there are only two obligations related to application of the
governing law that are directly imposed upon ISA tribunals by positive
law: that arbitrators may not refuse to apply the law on the grounds that
it is silent or obscure, and that arbitrators must give some reasons for
their decision. Neither mandates that the tribunal interpret the law in
any particular way. Even if one were to follow the now-discredited rst
generation of annulment decisions and nd that a failure to accurately
apply the governing law can constitute a manifest excess of powers, the
powers referred to are those granted by the arbitration agreement. We
must therefore look to the arbitration agreement to nd any constraints
on the tribunal’s power to apply the law.
What is meant by the arbitration agreement in this context is the
provision in the relevant IIA that expresses a contracting state’s consent
to arbitrate, if an investor from another contracting state initiates
arbitration. e term “parties” is not dened in the ICSID Convention,
but it is clear from the context that the term refers to the parties to the
dispute (i.e. the host state and the investor), rather than the parties to
the IIA (i.e. the two or more states that ratied it).38 roughout the
ICSID Convention, states involved in ISAs are referred to as “Contracting
36. e “second generation” of annulment decisions emphasized this point.
See e.g. Maritime International Nominees Establishment v Republic of
Guinea (1988), ICSID Case No ARB/84/4, Klöckner I, supra note 31 and
Amco I, supra note 32.
37. e “third generation” of annulment decisions repeatedly armed this
point. See e.g. Wena Hotels Ltd v Arab Republic of Egypt (2002), ICSID
Case No ARB/98/4; Empresas Lucchetti SA and Lucchetti of Peru SA v
e Republic of Peru (2007), ICSID Case No ARB/03/4; and CMS Gas
Transmission Co v e Republic of Argentina (2007), ICSID Case No
ARB/01/8.
38. See International Bank for Reconstruction and Development, Report of
the Executive Directors on the Convention on the Settlement of Investment
Disputes between States and Nationals of Other States, ICSID Convention,
Regulations and Rules, Doc No ICSID/15/Rev. 1 (2003), at paras 28-30,
online:
Reg%20Rules_EN_2003.pdf >.
234
Karton, Law and Interpretive Authority in Investor-State Arbitration
States”, whereas “parties” always refers to the parties to the dispute, rather
than the parties to the IIA.
It may therefore seem that investors, as parties to a dispute, have the
power to shape the governing law. Indeed they may, but only if the host
state agrees. A host state’s oer to arbitrate, as expressed in the dispute
resolution provisions of the IIA, is a conditional oer – conditional on
the investor’s acceptance of the terms of the IIA’s arbitration agreement.
Nothing prevents a state and investor from making a subsequent
agreement as to the choice of law; Article 42 of the ICSID Convention
does not require that the parties make a choice of law at any particular
time or in any particular form in order for that choice to bind the
tribunal. However, unless the host state and the investor agree to law
other than one stated in the IIA, the choice of law in the IIA binds the
tribunal. In practice, although other laws (in particular domestic laws
of the respondent states) are frequently applied to other aspects of an
investor-state dispute, the content of the obligations created by the IIA is
almost invariably determined according to the law specied in the choice
of law provision of the IIA, if there is one.
e contractarian structure of arbitral authority in ISA thus renders
fatuous arguments about whether investment arbitrators are agents of
the parties or trustees of the treaty regime. International relations theory
classically distinguishes between third party adjudicators who are agents
of contracting states versus those who are trustees of the underlying
regime.39 e distinction is not binary; agent and trustee are opposing
ends of a spectrum.40 Where a particular adjudicative body sits within
that spectrum depends primarily on the “zone of discretion” delegated
to the adjudicative body by states parties to the treaty empowering the
39. Kurtz, supra note 8 at 267 (citing Karen Alter, “Agent or Trustee:
International Courts in their Political Context” (2008) 14:1 European
Journal of International Relations 33l and Jean-Jacques Laont &
David Martimort, e eory of Incentives: e Principal – Agent Model
(Princeton: Princeton University Press, 2001)).
40. Anthea Roberts, “Power and Persuasion in Investment Treaty Arbitration:
the Dual Role of States” (2010) 104 American Journal of International
Law 179 at 187 [Roberts, “Power and Persuasion”].
235
(2017) 3(1) CJCCL
tribunal.41 e breadth of the zone of discretion is determined by the
sum of competences explicitly delegated to an adjudicator minus the
sum of control instruments available for use by principals to curb their
operations.42 Tribunals acting within a large zone of discretion act in a
“permissive strategic environment as trustees of the values that inhere in
the treaties that constituted them” and can “shape or control the evolution
of the [treaty] regime”.43 Tribunals acting as agents, by contrast, must
“align their adjudicatory activities far more closely with the immediate
preferences” of their principals.44
To determine where ISA tribunals sit along that spectrum, one must
examine the structural features of the system, in particular the “systems
of control” within ISA.45 Arbitral jurisdiction is limited to the specic
set of disputes described in the arbitration provisions of IIAs. Arbitrators
have no life tenure and are appointed ad hoc for each dispute; they are
therefore vulnerable to retaliation for their decisions. Arbitral awards may
be overturned, albeit on narrow grounds, and more generally may be
overridden by the renegotiation of IIAs. States therefore possess a number
of control powers that constrain the authority of investment arbitrators,
suggesting that states parties expect arbitrators to exercise their authority
closely in line with the states parties’ objectives.46
Indeed, the only important aspect of ISA tribunals consistent with a
41. Alec Stone Sweet & omas L Brunell, “Trustee Courts and the
Judicialization of International Regimes: e Politics of Majoritarian
Activism in the European Convention on Human Rights, the European
Union and the World Trade Organization” (2013) 1 Journal of Law and
Courts 61 at 65.
42. Ibid; For an alternative formulation, see Roberts, “Power and Persuasion,
supra note 40 at 185 (the scope of adjudicators’ zone of discretion is
dened by “the interpretive powers explicitly or implicitly delegated to
them minus the formal and informal powers retained by treaty parties to
inuence their interpretations, including through dialogue” at 185).
43. Kurtz, supra note 8 at 267.
44. Ibid at 268.
45. See generally W Michael Reisman, Systems of Control in International
Adjudication and Arbitration: Breakdown and Repair (Durham, North
Carolina: Duke University Press, 1992).
46. Kurtz, supra note 8 at 268-69.
236
Karton, Law and Interpretive Authority in Investor-State Arbitration
trustee type of role (beyond the simple fact that they may issue decisions
that are binding on states) is the degree of interpretive authority that
is typically delegated to arbitrators. Writes Roberts, investment treaties
involve:
a low level of precision, because the commitments themselves are broad and
vague (e.g. the promise to treat investors fairly and equitably). Although
imprecision is normally associated with state discretion, when it is coupled
with a high degree of obligation and delegation, the opposite is true: the
body charged with interpreting and applying the standard is aorded wide
discretion. 47
e net result is a signicant shift of interpretive power from the
treaty parties to ISA tribunals. e vague, standard-like substantive
obligations contained in IIAs represent a greater degree of delegation to
adjudicators than would be entailed by more precise, rule-like normative
prescriptions.48
Nevertheless, while such delegation may be broad in a given
arbitration, it is entirely contingent upon the language of the particular
IIA, whose substantive previsions may be vague or precise. Either way,
the tribunal’s authority is entirely circumscribed. Unlike many matters
relating to ISDS therefore, it is unarguable that arbitrators can only ever
be agents of the states parties to the IIA. Given the radical decentralization
of the ISDS system,49 there is not even a treaty regime for them to be
trustees of in the way that, for example the European Court of Human
Rights is charged with developing, maintaining and, furthering the goals
of the European Convention on Human Rights. Beyond treaty regimes,
ISA tribunals have no inherent obligation to “international law”, “civil
society”, or any other such abstraction. ey have only the power and
duty to resolve the individual disputes for which they are constituted by
interpreting and applying the law chosen by the parties, in the manner
and according to the procedures that the parties direct.
is is not to say that arbitrators are merely agents of the states
47. Roberts, “Power and Persuasion”, supra note 40 at 190 [citations omitted].
48. Kenneth W Abbott et al, “e Concept of Legalization” (2000) 54:3
International Organization 401 at 413.
49. Mortenson, supra note 15 at 11.
237
(2017) 3(1) CJCCL
party to the IIA and nothing else. At minimum, a tribunal’s authority
“derives from both the general grant of power by the treaty parties and
the specic invocation of that grant by an investor of one treaty party
(as claimant) against another treaty party (as respondent)”.50 is is the
“signal innovation” of modern IIAs, and their dening feature – that
they provide a direct remedy for individual investors against states,
without any intermediation.51 More concretely, the investor also has a
role in appointing the members of the tribunal, which means that states
“play a lesser role in determining the appointment and reappointment of
investment arbitrators compared to most international judges”. 52
However, the investor normally plays no role in shaping the
tribunal’s interpretive authority. Writes Roberts, “Investment tribunals
cannot be viewed only as agents of the disputing parties because the
disputing parties’ rights and the investment tribunal’s powers are dened
and delimited by the treaty’s grant of power”.53 Accordingly, in exercising
their power to interpret and apply the governing law – to decide the
merits of disputes – ISA tribunals should act as if they were agents of the
states parties alone.
IV. Interpretation of the Governing Law in
Investment Treaty Arbitrations
ISA tribunals derive neither their power to apply the law nor their duties
associated with the exercise of that power from the general law. Within
the set of public international law dispute resolution institutions, those
that are arbitral in character must be sharply distinguished from those that
derive their authority in other ways, in particular the “standing” courts
such as the ICJ and the World Trade Organization Appellate Body. No
treaty regime empowers arbitral tribunals; even IIAs only provide for the
establishment of tribunals if and when an investor makes a claim against
50. Roberts, “Power and Persuasion”, supra note 40 at 182.
51. Mortenson, supra note 15 at 2.
52. Anthea Roberts, “Clash of Paradigms: Actors and Analogies Shaping
the Investment Treaty System” (2013) 107:1 American Journal of
International Law 45 at 61 [Roberts, “Clash of Paradigms”].
53. Ibid.
238
Karton, Law and Interpretive Authority in Investor-State Arbitration
an individual state. Except to the extent that positive law imposes non-
derogable duties upon the tribunal (such as a duty to declare conicts of
interest, not to take bribes, or, more prosaically, to provide reasons for
their decision), tribunals owe only those duties that the parties impose
and owe performance of those duties only to the parties.
e structure of interpretive authority in ISA has not been
fundamentally altered by its transposition from the international
commercial arbitration context to public international law disputes and
“triangular” ISA. ere is an arbitration agreement (the dispute resolution
provisions of the IIA); there are parties to that arbitration agreement (the
states that enacted the IIA); there are chosen procedures (the ICSID
Convention and Rules, the New York Convention and UNCITRAL Rules,
or whatever other combination the parties select); nally, there is a choice
of the rules of law according to which the tribunal must decide the merits
of the dispute: the choice of law provision in the IIA, potentially modied
by other agreements between the host state and the investor. If there is
no choice of law, this should be taken as an implied choice of the default
law, such as that mandated by the second sentence of Article 42(1) of
the ICSID Convention, or alternatively as a delegation to the tribunal of
the power to choose the applicable law. One way or another, there will
be an identiable set of rules of law that the tribunal must apply, save
only the rare circumstance where the parties have chosen ex aequo et bono
determination.
To make these generalities more concrete, consider the question of
whether tribunals have a duty to decide consistently with prior tribunals.
e Saipem tribunal thought so; it found that it had a “duty to seek
to contribute to the harmonious development of investment law”.54 On
the other hand, the Romak tribunal held that, “Ultimately, the Arbitral
Tribunal has not been entrusted, by the Parties or otherwise, with a mission
54. Saipem SpA v e People’s Republic of Bangladesh (2009), ICSID Case No
ARB/05/07, at para 90 [Saipem].
239
(2017) 3(1) CJCCL
to ensure the coherence or development of ‘arbitral jurisprudence’”.55 e
point here is not so much that Saipem was wrongly decided and Romak
was correct on the point that motivated these observation: how tribunals
should treat precedents.56 Rather, the point is that the Saipem tribunal
was asking the wrong question – how should tribunals conceive of their
role? – while the Romak tribunal was asking the right one – what role
have the parties assigned to the tribunal?
erefore, nearly all questions of the applicable law – how it is
chosen, how its content is ascertained, and how it is applied to the facts –
are at heart matters of interpretation. e tribunal should determine the
intentions of the parties as expressed in the relevant agreements and, where
those agreements do not provide a clear answer, should act according to
the parties’ presumed intentions. Applying rules beyond those the parties
agreed to or in a manner not agreed to by the parties constitutes an excess
of authority, which, if “manifest”, constitutes grounds for annulment of
the tribunal’s award.
What laws do current IIAs call for? Most contain no choice of
55. Romak SA (Switzerland) v e Republic of Uzbekistan (2009), UNCITRAL
PCA Case No AA280 at para 171 [Romak]; see also AES Corporation v
e Argentine Republic (2005), ICSID Case No ARB/02/17 (concluding
that “[e]ach tribunal remains sovereign and may retain, as it is conrmed
by ICSID practice, a dierent solution for resolving the same problem” at
paras 30-31).
56. e status of precedents in ISA is outside the scope of this article, but
it is worth mentioning that citation of precedent is consistent with the
structural argument advanced in this article. However, precedents are
relevant not because tribunals have some obligation to decide consistently
with each other or to advance the development of IIL. Rather, “judicial
decisions” constitute a source of international law under the Statute
of the International Court of Justice (albeit as “subsidiary means”) for
determining the content of international law. erefore, even under a
narrow, orthodox denition of the sources of international law, precedents
are still within the scope of the law chosen by the choice of law provisions
of most IIAs.
240
Karton, Law and Interpretive Authority in Investor-State Arbitration
law provision whatsoever.57 For ICSID arbitrations, that leads to the
application of the residual rule of Article 42(1), which calls for the host
state’s law and the applicable rules of international law. For non-ICSID
arbitrations, this generally means that the parties to the dispute may agree
to the governing law or, failing such agreement, the tribunal chooses.58
Among the IIAs that contain choice of law provisions, the wording
varies. However, the choice of law provisions fall into six categories.59
e most common type of clause calls for the application of four
sources of legal rules: the IIA itself, the municipal law of the host state,
the provisions of any investment agreement or contract relating to the
investment, and applicable principles of international law. e second
type of choice of law clause is similar in that it lists various sources of law,
but it provides the tribunal should “take [these sources] into account”, as
opposed to applying them, and may also provide that the list of sources
is non-exhaustive. e third type calls for application of the IIA itself and
international law. e fourth type is found in Indian BIT practice, and
calls for application of the treaty text alone.60 e fth type, which appears
only in more recent treaties and goes into more detail as to the relevant
sources of law, species dierent laws to apply to dierent matters. For
example, it might specify that the IIA text and international law apply
57. Christoph Schreuer, “Jurisdiction and Applicable Law in Investment
Treaty Arbitration” (2014) 1:1 McGill Journal of Dispute Resolution 1 at
12.
58. See UNCITRAL Arbitration Rules GA Res 68/109, UNCITRAL, 2013,
UN Doc A/68/462 art 35(1) (most non-ICSID ISAs are conducted
under the UNCITRAL Rules, which determine the applicable law in this
manner).
59. Andrew Newcombe & Lluis Paradell, Law and Practice of Investment
Treaties: Standards of Treatment (Alphen aan den Rijn: Kluwer, 2009)
at 78. See generally Hege Elisabeth Kjos, Applicable Law in Investor-
State Arbitration: e Interplay Between National and International Law
(Oxford: Oxford University Press, 2013).
60. Ibid at 80 (It also provides that arbitrations may be submitted only to
ad hoc arbitration under the UNCITRAL Rules, and not to ICSID,
presumably to avoid Article 42(1), second sentence, ICSID Convention,
which provides a backup choice of law that includes reference to
international law).
241
(2017) 3(1) CJCCL
to claims relating to breaches of treaty obligations, but that claims
relating to investment authorizations are governed by the law specied
in the authorization or, failing that, a combination of the law of the
host state, the IIA text, and international law.61 Finally, the sixth category
of choice of law provisions may contain similar language to any of the
previous types, but then adds that an interpretation of an IIA provision
made jointly by the contracting states is binding on a tribunal.62 ese
provisions vary between IIAs as to matters like the time limit for issuance
of a joint interpretation and whether the tribunal must, if so requested by
a state party, ask for a joint interpretation.
e apparent variety between IIAS does not, in the end, make much
of a dierence. “e only signicant dierence between the various rules
on applicable law in treaties lies in the absence of a reference to host
state law in some of them. e narrower clauses refer only to the treaty
itself and to applicable rules of international law”. 63 More generally, the
default rule under Article 42(1) of the ICSID Convention captures most if
not all elements contained in the more elaborate choice of law provisions
in some IIAs. What all of these categories, except the last one, have in
common is that they do not make any attempt to guide the tribunal’s
ascertainment or application of the chosen law; all they do is make an
exclusive or non-exclusive list of the laws tribunals must consider.
e only exception to this pattern is the choice of law provisions
that call for joint interpretations of the IIA by the states party to it. ese
do have the potential to signicantly aect outcomes; however, as will
be discussed below, states have the power to shape treaty interpretation
61. See e.g. Dominican Republic-Central America-United States Free Trade
Agreement, 5 August 2004, 32 ILM (entered into force 1 March 2006).
62. See North American Free Trade Agreement Between the Government of
Canada, the Government of Mexico, and the Government of the United
States, 17 December 1992, Can TS 1994 No 2 (entered into force 1
January 1994) [NAFTA] (NAFTA Chapter 11 is probably the most
prominent example. Article 1131(2) provides that tribunals are bound
by interpretations of the NAFTA Free Trade Commission, a joint body
composed of representatives of the three NAFTA contracting states:
Canada, the USA, and Mexico).
63. Schreuer, supra note 57 at 12-13.
242
Karton, Law and Interpretive Authority in Investor-State Arbitration
through subsequent agreements and subsequent practice, regardless of
whether the treaty contains a specic provision calling for binding joint
interpretations.64 Except for clauses relating to joint interpretations,
IIAs typically give no guidance as to how the applicable law ought to be
interpreted, nor does the default choice of law provision in the ICSID
Convention. I call choice of law clauses of this standard kind “unqualied”
choices of national or international law.65
How should tribunals proceed when faced with an unqualied
choice of national or international law? Such a choice of law provision
leaves unspecied a number of issues relating to interpretation of that
law, such as how its rules are to be ascertained, whether reference may be
made to other laws or to general principles, and how conicts are to be
resolved between these dierent sources of rules of law (i.e. which norms
take precedence in case of conicts).
Since the choice of law clause constitutes an agreement of the parties
to the arbitration agreement, these questions should themselves be
answered by reference to a set of interpretive principles. Unfortunately,
in most cases involving unqualied choices of law, neither the text of the
IIA nor any of the other means of interpretation prescribed by Articles
31-32 of the Vienna Convention on the Law of Treaties66 (“VCLT”) (such
as an examination of the IIA’s travaux) sheds any light. e tribunal’s only
safe option is to presume that, unless the parties carve out some area of
the law or qualify the choice of law in some other way, choice of a legal
system means a choice of all of that legal system’s substantive rules.
With respect to national laws applied by ISA tribunals (for example,
to determine whether the host state violated the terms of an investment
authorization or investment contract governed by national law) this
means applying all of the various sources of law recognized as valid within
64. See infra notes 78-92 and accompanying text.
65. One sees the same thing in contract drafting practice, where most choice
of law clauses simply name the law of some state without qualication.
See Joshua Karton, “e Arbitral Role in Contractual Interpretation”
(2015) 6:1 Journal of International Dispute Settlement 1 at 27-30.
66. Vienna Convention on the Law of Treaties, 23 May 1969, 1155 UNTS 331
arts 31-32 (entered into force 27 January 1980) [VC LT].
243
(2017) 3(1) CJCCL
the chosen national legal system – not just relevant code provisions, but
also to all relevant statutes and regulations, along with case law and
customs recognized as authoritative within the named legal system.67
Since all state laws exist within a normative hierarchy that national
courts would unhesitatingly apply (although they may disagree on the
outcomes in individual cases), ISA tribunals should also take into account
the relationship between the various national rules of law, including
constitutional norms that might invalidate some other rule of law.68 As a
corollary, for matters governed by national law, the chosen national law
should be applied to the exclusion of any other national or transnational
laws; when parties select a single state’s law, they presumably intend that
no other national laws should be applied.
With respect to international law, it means roughly the same thing:
ISA tribunals should apply international law as a whole, not just rules
developed in the investment context, and recognize supervening principles
of public international law, such as ius cogens or rules of international
human rights law, that may have priority over the investment treaty.
In all this, it is the intentions of the state parties to the investment
treaty that matter, rather than the intentions of the parties to the dispute
(i.e. the investor and the host state). Investment arbitrators are agents of
67. is specication is made in some IIAs; see US, Oce of the United
States Trade Representative, 2012 US Model Bilateral Investment Treaty
(2012), online:
ACIEP%20Meeting.pdf> (denes national law as “the law that a domestic
court or tribunal of proper jurisdiction would apply in the same case”
at 34, n 22. Without further specication, this must be taken to mean
that the tribunal should situate itself in the position of a domestic court,
consulting whatever sources of law are authoritative in that state in
whatever hierarchy would be observed within that state) [US Model BIT].
68. International Law Commission, Draft Articles on Responsibility of States for
Internationally Wrongful Acts, with Commentaries, ILC 2001 A/56/10, art
3 (this is qualied by the principle that states may not plead conformity
with their own municipal laws to excuse a violations of their international
law obligations; “the characterization of an act of a State as internationally
wrongful is governed by international law. Such characterization is not
aected by the characterization of the same act as lawful by internal law”,
art 3).
244
Karton, Law and Interpretive Authority in Investor-State Arbitration
the treaty parties, not the disputing parties.
is rather bald statement will be instinctively rejected by many in
the ISA community, who see the system of investment treaties as designed
in large part to protect the legitimate expectations of investors. erefore,
two immediate qualications are required. First, I do not mean to say
that the legitimate expectations of investors are never relevant. ey may
matter greatly in a variety of circumstances, especially for issues that turn
on the investment contract that the investor signed. In addition, the fact
that some act of the host state was contrary to previous representations
by the state to the investor – representations that gave rise to legitimate
expectations by the investor that the state would follow through on those
representations – may mean that the action constituted a treaty breach,
specically the fair and equitable treatment standard contained in most
IIAs.69 Recent treaty practice adopts this vision of legitimate expectations.
For example, the Canada-European Union Comprehensive Economic and
Trade Agreement (“CETA”) provides:
When applying the above fair and equitable treatment obligation, a Tribunal
may take into account whether a Party made a specic representation to an
investor to induce a covered investment, that created a legitimate expectation,
and upon which the investor relied in deciding to make or maintain the
covered investment, but that the Party subsequently frustrated.70
But the expectations of the investor cannot be applied to shape the
meaning of treaty obligations themselves: the choice of law in the IIA
constitutes a condition of the host state’s standing oer to arbitrate, and
investors accept that oer when they launch an arbitration under the IIA.
e second qualication is that the power of the state parties to the
69. A consistent line of case law, especially under NAFTA Chapter 11,
embraces this point of view. See William Ralph Clayton, William Richard
Clayton, Douglas Clayton, Daniel Clayton andBilconof Delaware Inc v
Government of Canada (2009), UNCITRAL PCA Case No 2009-04 at
para 589; Mobil Investments Canada Inc & Murphy Oil Corporation v
Canada (2012) ICSID Case No ARB(AF)/07/4 at para 152; Glamis Gold
Ltd v United States of America (2009), UNCITRAL Arbitration Rules at
para 22; Waste Management Inc v United Mexican States (2004), ICSID
Case No ARB(AF)/00/3, at paras 98-99.
70. 30 October 2016, art 8.10(4) (not yet entered into force) [CETA].
245
(2017) 3(1) CJCCL
IIA to shape the tribunal’s interpretation and application of the IIA does
not imply a requirement to interpret IIAs in a manner favourable to the
host state in any given dispute. States enact IIAs in order to create a
fair and predictable climate for foreign investments, with the ultimate
(and sometimes directly expressed) goal of increasing FDI ows. In
order to achieve those objectives, IIAs grant rights directly to investors,
rights would be largely meaningless if states did not delegate to neutral
and independent arbitral tribunals the power to enforce those rights.71
Accordingly, some state conduct harmful to investors, even if conducted
for a legitimate regulatory purpose, can and should be held to violate
state obligations under the IIA and to give rise to an obligation to pay
compensatory damages.
ese qualications aside, the contractarian structure of arbitral
authority in ISA means that states eectively control the interpretive
process. ey are, to use the memorable phrase of Methymaki and
Tzanakopoulos, “masters of puppets”.72
V. Implications of the Proposed Approach
In this Part, I explain how the abstract points made in Part III apply
to the interpretation of IIAs by arbitral tribunals. e most direct
consequence of my focus on choice of law is that states have the ultimate
power not only to choose the governing substantive rules of law, but also
to direct their interpretation. is Part rst describes how states can make
use of that power to bind ISA tribunals to interpret IIAs in particular
ways. Next, it explains how tribunals should interpret IIAs under the
commonly-employed choice of law provisions that currently exist in IIAs.
71. Roberts, “Power and Persuasion”, supra note 40 at 183.
72. Eleni Methymaki & Antonios Tzanakopoulos, “Masters of Puppets?
Reassertion of Control through Joint Investment Treaty Interpretation”
(2016) Oxford Legal Studies Research Paper No 10/2016 at 1-2, online:
246
Karton, Law and Interpretive Authority in Investor-State Arbitration
A. States in the Driver’s Seat
In the last few years, a number of states have pursued a number of
strategies in an eort to “reassert control” over ISAs.73 Some have
“exited” the ISDS system by withdrawing from the ICSID Convention
or from individual treaties they view as problematic, or by not renewing
investment treaties when they expire.74 More often, states have attempted
to renegotiate the terms of existing treaties or draft new terms for treaties
going forward. Leaving aside treaty terms that would change the structure
of the ISDS system,75 states have generally adopted one of two strategies:
dening substantive obligations more narrowly and precisely than in past
IIAs, and carving out specic exceptions to state liability in areas such
as taxation, nancial services, public health and the environment, and
culture.76 Taken together, these strategies suggest that states in general
desire greater detail in investment treaties, with the aim of constraining
arbitral discretion and promoting greater certainty.77
Despite some notable diplomatic successes, such as the conclusion
of CETA, these eorts have been somewhat underwhelming. Some
redrafting attempts fail on their own terms to introduce greater precision
73. See Andreas Kulik, ed, Reassertion of Control over the Investment Treaty
Regime (Cambridge: Cambridge University Press, forthcoming 2017).
74. Roberts, “Power and Persuasion”, supra note 40 at 191.
75. See European Commission, “Fact Sheet on Investment Provisions in
the EU-Canada free trade agreement (CETA)” (February 2016), online:
(such as the EU’s proposal for a permanent investment court, which has
been incorporated into the recently-concluded CETA).
76. Mitchell Moranis, “Between power and procedure: the changing balance
of investment treaty protections” (2015) 32:1 Arbitration International 81
at 83.
77. Ibid at 101.
247
(2017) 3(1) CJCCL
and predictability.78 But no matter how detailed and specic treaty
language becomes, it can never resolve all questions as to the content of
obligations arising under the treaty. ere are two reasons for this, rst,
all treaty language, no matter how apparently clear, must be interpreted.
As former President of the ICJ, Dame Rosalyn Higgins put it:
Reference to the ‘correct legal view’ or ‘rules’ can never avoid the element
of choice (though it can seek to disguise it), nor can it provide guidance
to the preferable decision. In making this choice one must inevitably have
consideration for the humanitarian, moral, and social purposes of the law.79
Second, not all potential areas of dispute can possibly be predicted at the
time an IIA is negotiated or renegotiated. ere will always be unforeseen
gaps.
In other words, attempting to eliminate interpretive uncertainty ex
ante by drafting more specic treaties is ultimately a doomed enterprise,
based on a misunderstanding of how law works. Treaty text is never:
… reducible to a xed, immutable expression of the rule … the engagement of
actors with a legal text is historically contingent: it is structured by the frame
in which it is situated, and it is measured against rules contained within that
frame, not to mention the past practices of other actors or disputants.80
As Schwebel writes of the EU’s attempts to reassert control by redrafting
treaty provisions with greater precision:
ere is … a troubling message throughout the EU’s [proposals] … the
notion that international law is simply a set of rules to be applied by judges
mechanically. Under this view, the more the [treaty] text claries what the law
is, the less doubt will exist, the more rigorous and consistent the analysis of
78. Federico Ortino, “Rening the Content and Role of Investment ‘Rules’
and ‘Standards’: A New Approach to International Investment Treaty
Making” (2013) 28 ICSID Rev 152 at 158; but see Moranis, ibid
(arguing that “States are beginning to ll the gaps in treaties, providing
greater detail and setting clear limits to their obligations and investors’
rights” at 83).
79. Rosalyn Higgins, Problems & Process: International Law and How We Use It
(Oxford: Oxford University Press, 1994) at 5.
80. Hernandez, supra note 6 at 171 citing Ingo Venzke, How Interpretation
Makes International Law: Between Normative Twists and Semantic Authority
(Oxford: Oxford University Press, 2012) at 49.
248
Karton, Law and Interpretive Authority in Investor-State Arbitration
TTIP awards will be.81
is view is at best misguided and at worst wilfully obtuse. As Schwebel
concludes, “[t]he fact is that applying rules always involves a certain
degree of choice”.82
A potentially more fruitful approach, adopted by some states but
still signicantly underused, is to promulgate interpretations of treaties
ex post.83 As discussed above, some IIAs expressly provide that the states
parties may jointly issue interpretations of a treaty that are binding upon
tribunals. But even where no such provision exists, rst principles dictate
that states, acting together, are masters of the treaties they make:
ultimately the power of authoritative interpretation of a norm … rests with
the organ which promulgated the norm and which has the power to revoke it.
In the context of a treaty, the organ that has this power is the peculiar organ
formed by all the states parties to it.84
When states agree on an authentic interpretation of the treaty, it has both
retroactive and prospective eect: “[w]e are in the realm of the lawmaker
restricting the range of possible meanings of a norm once and for all, by
selecting one of them to control indenitely, rather than selecting one
among them to apply to a specic case”.85 us, the only legal eect of
an IIA provision regulating joint interpretations is potentially to limit
the scope of states’ power to issue joint interpretations, for example by
imposing temporal limits or by requiring that the joint interpretations
81. Stephen M Schwebel, “e Outlook for the Continued Vitality, or
Lack ereof, of Investor-State Arbitration” (2016) 32:1 Arbitration
International 1 at 7.
82. Ibid at 8.
83. Roberts, “Power and Persuasion”, supra note 40 at 179.
84. Methymaki, supra note 72 at 5 citing Question of Jaworzina (Polish-
Czechoslovakian Frontier) (Advisory Opinion) [1923] PCIJ Ser B No 8,
37: (“the right of giving an authoritative interpretation of a legal rule
belongs solely to the person or body who has the power to modify or
suppress it” at 5) [emphasis in original].
85. Ibid at 5 [emphasis in original].
249
(2017) 3(1) CJCCL
take a particular form or be arrived at through a particular process.86
is conclusion is conrmed by customary international law
principles of interpretation, as recognized by tribunals going back
more than a century. In 1911, the US-Mexico International Boundary
Commission in e Chamizal Case87 found that joint interpretations
were not only binding on the tribunal, but indeed also on the parties to
the treaty. e Commission found it:
impossible to come to any other conclusion than that the two nations have, by
their subsequent treaties and their consistent course of conduct in connection
with all cases arising thereunder, put such an authoritative interpretation upon
the language of the treaties of 1848 and 1853 as to preclude them from now
contending that the uvial portion of the boundary created by those treaties is
a xed line boundary.88
e binding force of subsequent agreements and subsequent practice
of the parties is also codied in the VC LT , which expresses customary
principles of interpretation. Article 31(1) requires that treaties be
interpreted “in good faith in accordance with the ordinary meaning to be
given to the terms of the treaty in their context and in light of its object
and purpose”.89 At a minimum, joint interpretations of a treaty by the
states party to it form part of that context, and are therefore relevant to
all matters of interpretation. But in addition, paragraph 3 of Article 31
provides expressly that, together with the treaty’s context, interpreters
must also take into account:
(a) Any subsequent agreement between the parties regarding the interpretation
of the treaty or the application of its provisions; (b) Any subsequent practice
in the application of the treaty which establishes the agreement of the parties
regarding its interpretation; (c) Any relevant rules of international law
86. See NAFTA, supra note 62 (Chapter 11, art 1131(2) provides that
interpretations of the NAFTA Free Trade Commissions are binding upon
tribunals, and art 1132, which provides that disputing parties may require
tribunals to request from the Free Trade Commission interpretations on
the scope of a reservation or exception set out in one of the annexes to
Chapter 11).
87. (1911), XI RIAA 309.
88. Ibid at 328.
89. VC LT , supra note 66, art 31(1) [emphasis added].
250
Karton, Law and Interpretive Authority in Investor-State Arbitration
applicable in the relations between the parties.90
e distinction in paragraph 3 between subsequent agreements and
subsequent practice makes no dierence in terms of their legal eects, but
only in terms of the necessary evidence; a subsequent agreement under
subparagraph (a) constitutes “an ipso facto authentic interpretation”, while
a party relying on subsequent practice under subparagraph (b) must show
that the practice of the states parties substantiates a particular common
understanding.91 If the states that are party to an IIA promulgate any kind
of document setting out their understanding of the meaning of treaty
terms, such document would qualify as both subsequent agreement and
subsequent practice. Either way, writes Villiger, “the parties’ authentic
interpretation of the treaty terms is not only particularly reliable, it is also
endowed with binding force”.92 Here, I refer to subsequent agreements
and subsequent practice that have the purpose and eect of specifying
the meanings of treaty provisions collectively as “joint interpretations”.
Under the VC LT , states’ powers to issue joint interpretations are
eectively unlimited. In the investment arbitration context, it has been
argued that the direct rights that are vested in investors by IIAs may impose
limits on the absolutely binding character of joint interpretations.93
Investors may rely to their detriment on meanings of the applicable IIA
as they are understood at the time the investment is made. In addition,
a tension arises between states’ dual roles, as parties to the IIA and as
respondents in individual arbitrations. Roberts explains:
Viewing investment treaty arbitration solely through a public international
law, state-to-state prism is unsatisfactory because investment treaties create
90. Ibid.
91. First Report of Special Rapporteur Georg Nolte, Subsequent Agreements
and Subsequent Practice in Relation to Treaty Interpretation, ILC, 65th
Sess, UN Doc A/CN.4/660 (2013) at para 70; see also Rahim Moloo,
“When Actions Speak Louder an Words: e Relevance of Subsequent
Party Conduct to Treaty Interpretation” (2013) 31:1 Berkeley Journal
of International Law 39 at 58 (describing subsequent agreements and
subsequent practice as existing along a single evidentiary continuum).
92. Mark Villiger, Commentary on the 1969 Vienna Convention on the Law of
Treaties (Leiden: Martinus Nijho, 2009) at 429.
93. See Roberts, “Power and Persuasion”, supra note 40 at 207-15.
251
(2017) 3(1) CJCCL
reciprocal rights and duties for the treaty parties and rights for nonstate actors
(investors). To increase condence in and enforcement of those rights, states
have delegated the power to resolve investor-state disputes to arbitral tribunals.
If the treaty parties could agree at any time on a binding interpretation of
the treaty, they could use that authority to undermine not only investors’
expectations but also tribunals’ dispute resolution powers.94
For this reason, Roberts and others argue that states’ power to issue joint
interpretations of IIAs may be limited on two broad bases: reasonableness
and timing. With respect to reasonableness, if a joint interpretation selects
one of a set of reasonable interpretations of a disputed treaty provision,
that is indisputably within the states parties’ power, but adoption of
an unreasonable or unexpected interpretation may constitute a de facto
amendment of the treaty, which would be unfair to the investor to apply
retroactively. With respect to timing, changing the terms of the treaty
after an investment is made may involve harm to investors who have
detrimentally relied on the treaty; xing the treaty’s terms after a claim is
led may harm not only the investor, but also the integrity of the arbitral
process.95 For this reason, Roberts concludes that “the persuasiveness of
treaty party interpretations should be understood as a function of their
timing and reasonableness”.96
I reject limits based on reasonableness. States do indeed delegate to
tribunals the power to interpret treaties, but that delegation is limited by
subsequent agreements and subsequent practice. Writes Crawford:
In the context of investment treaty arbitration there is a certain tendency to
believe that investors own bilateral investment treaties, not the states parties to
them … at is not what international law says. International law says that the
parties to a treaty own the treaty and can interpret it. One might say within
reason, but one might not question their application of reason as they see t.97
e contractarian structure of authority in ISA means that investment
arbitrators are, if anything, more bound by joint interpretations than
94. Ibid at 183 [emphasis in original].
95. Ibid at 212.
96. Ibid.
97. James Crawford, “A Consensualist Interpretation of Article 31(3) of the
Vienna Convention on the Law of Treaties” in Georg Nolte, ed, Treaties
and Subsequent Practice (Oxford: Oxford University Press, 2013) 29 at 31.
252
Karton, Law and Interpretive Authority in Investor-State Arbitration
adjudicators applying other kinds of treaties. A joint interpretation
may be objectively unreasonable, but it is not within the power of ISA
tribunals to declare it to be so.
In addition, there should not be any temporal limit on states’
power to issue interpretations of treaties, unless a limit is imposed by
the IIA itself or a legal stabilization clause in an investment agreement
or other contract between the respondent state and the investor. Absent
such a provision, investors have no legitimate expectation that the legal
environment of their investment will remain stable for the entire life
of that investment. Some recently-enacted treaties make this principle
explicit. For example, CETA provides that, “[f]or greater certainty, the
mere fact that a Party regulates, including through a modication to its
laws, in a manner which negatively aects an investment or interferes
with an investor’s expectations, including its expectations of prots, does
not amount to a breach of an obligation…”.98 In other words, states
may regulate without fear that the mere fact that their regulations reduce
foreign investors’ prots will lead to a nding that they have breached the
treaty. erefore, a fortiori states must be able to enter into subsequent
agreements or establish subsequent practice in order to clarify or change
the meaning of treaty obligations, even if this restricts the range of
possible interpretations tribunals may adopt. 99
More importantly for present purposes, the choice of law provisions
in IIAs or in the ICSID Convention state only that tribunals shall or
may apply certain laws; they do nothing to x the content of those
laws in place. Accordingly, arbitrators are bound to apply the law as it
stands at the time they render their decision. is arguably raises due
process concerns, especially where states issue joint interpretations after
a dispute arises. However, nothing about a change in the underlying law
constitutes a procedural violation that might justify annulment of an
award or refusal of enforcement. On the contrary – a tribunal’s failure to
apply the governing law as authoritatively interpreted by the states party
to the treaty would constitute an excess of powers. In addition, as the
98. CETA, supra note 70, art 8.9(2).
99. Methymaki, supra note 72 at 19.
253
(2017) 3(1) CJCCL
investor’s home state has an opportunity to protect its investor by arguing
that a subsequent agreement or practice alleged by the respondent states
constitutes an impermissible moving of the goalposts. If it decides not to
do so, that “must mean something”.100 It is therefore unfortunate that ISA
tribunals tend toward “a certain reluctance … to embrace wholeheartedly
… the unquestionable vesting with binding force of joint interpretations
by the states parties of their own treaties”.101
e correct approach is demonstrated by a recent decision of the
High Court of Singapore annulling an investor-state award in favour of
a Macanese investor under the PRC-Laos BIT on the basis of excess of
jurisdiction (“Sanum Investments”).102 e basis of the annulment was an
exchange of letters between China and Laos, which reected the common
position of China and Laos that the BIT did not extend to Macau. e
court held that the letters constituted a subsequent agreement of the
states parties under Article 31(3)(a) VC LT , establishing conclusively that
the Macanese investor could not take advantage of the BIT.103
It made no dierence that exchange of letters came after the tribunal
had issued its decision upholding its jurisdiction; the Singaporean court
dismissed the investor’s due process concerns, reasoning that “parties
relying on the provisions of BITs” should be aware of the potential impact
of Article 31(3)(a). e letters reected the “common understanding” of
the parties rather than a retroactive amendment of the PRC-Laos BIT.104
e Sanum Investments decision reects the conception, advanced in this
article, that while investors are third-party beneciaries of IIAs and derive
certain direct rights from them, investors have no claim to interpretive
power over them.
ese two avenues – redrafting of substantive IIA provisions and
promulgation of binding interpretations after ratication of the IIA – have
been employed by states and explored in the literature. e contractarian
100. Ibid at 20.
101. Ibid at 15.
102. Government of the Lao Peoples’ Democratic Republic v Sanum Investments
Ltd [2015] SGHC 15 [Sanum Investments].
103. Ibid at paras 69-70.
104. Ibid at paras 76-77.
254
Karton, Law and Interpretive Authority in Investor-State Arbitration
theory advanced here points the way to a third strategy that, thus far, has
been largely overlooked: states may shape the application of the governing
law by adding content to the choice of law provisions in IIAs. For
example, if states want arbitrators to follow or not to follow precedents,
they can so provide in their treaties. e US Model BIT provides that “An
award made by a tribunal shall have no binding force except between the
disputing parties and in respect of the particular case”.105 is provision is
modeled on Article 59 of the ICJ Statute, which has in no way prevented
the ICJ from regularly treating its prior judgments as persuasive. Indeed,
a de facto system of precedent has already existed in ISA for some time.106
us, if tribunals are to be restrained from considering prior case law, the
Model BIT should be redrafted accordingly; conversely, if the states party
to the IIA want tribunals to take prior decisions into account, the clause
can so state.
Similarly, if the states party to an IIA want to restrict the tribunal
to certain international law doctrines and not others, it is within their
power to do so. But even most recently-drafted treaties, including those
promulgated by states intent on restricting arbitrators’ zone of discretion,
do not take advantage of this opportunity. For example, Article 8.31(1)
of CETA provides:
When rendering its decision, the Tribunal established under this Section shall
apply this Agreement as interpreted in accordance with the Vienna Convention
on the Law of Treaties, and other rules and principles of international law
applicable between the Parties.107
Such a provision is eectively meaningless – a missed opportunity. e
105. US Model BIT, supra note 67, art 34(4).
106. See e.g. Eric De Brabandere, “Arbitral Decisions as a Source of
International Investment Law” in Tarcisio Gazzini & Eric De Brabandere,
eds, International Investment Law: e Sources of Rights and Obligations
(Leiden: Martinus Nijho, 2012) at 245; Lucy Reed, “e De Facto
Precedent Regime in Investment Arbitration: A Case for Proactive
Case Management” (2010) 25:1 ICSID Rev 95; Andrea Bjorklund,
“Investment Treaty Arbitral Decisions as Jurisprudence Constante” in
Colin Picker et al, eds, International Economic Law: e State and Future
of the Discipline (London: Bloomsbury, 2008) at 265.
107. CETA, supra note 70, art 8.31(1).
255
(2017) 3(1) CJCCL
VC LT already applies (as treaty law and as an expression of customary
international law) to the interpretation of all IIAs except the small
minority that do not take the form of treaties. Stating that tribunals
should apply “other rules and principles of international law applicable
between the Parties” does nothing to constrain arbitrators’ authority to
determine which rules and principles are applicable.108 Despite all its
verbiage, it is no dierent from a simple choice of “international law”.
States easily provide that particular treaties or areas of international
law outside of the IIA text do or do not apply. ey could contract out
of the interpretive rules in the VC LT . ey could even provide that
interpretations other than those of the states parties are binding. For
example, the choice of law provision in the IIA could declare opinions of
the International Law Commission are binding. Such a provision could
be accompanied by treaty language to the eect that failure to abide by
reports of the ILC would constitute manifest excess of the tribunal’s
powers, so that awards could be annulled or refused enforcement. (Note
that I am not arguing that this would necessarily be a good idea, only that
it would be eective.)
e upshot is not that any particular interpretive approach is
optimal, or even that any one approach can be considered optimal for all
combinations of states parties. It is simply to encourage treaty drafters to
be more creative and more assertive in shaping the interpretive authority
of arbitrators beyond simply making unqualied choices of whole legal
systems. Given the contractarian structure of interpretive authority in
ISA, arbitrators would be bound by more specic choice of law provisions.
B. In Most Cases, International Investment Agreements
Should be Interpreted Extensively
e previous section considered the ways in which states might act to
take advantage of the contractarian structure of arbitral authority in ISA.
But renegotiating treaties takes a great deal of time, and in the current
108. See Schwebel, supra note 81 at 8-9 (arguing that, by contrast, the
CETA text is probably overzealous in its attempts to restrain tribunals’
interpretations of national laws and regulations).
256
Karton, Law and Interpretive Authority in Investor-State Arbitration
political climate there may be little support for new multilateral or bilateral
IIAs. Even if the political will existed and the collective action problems
could be overcome, the transaction costs involved in renegotiating the
thousands of extant treaties would be prohibitive.109 Even issuance of
joint interpretations requires agreement between states with potentially
opposed interests, as well as timely coordination between the states.
In the meantime, arbitrations continue to be commenced and
tribunals must resolve them. To do so, they will have to interpret the
IIAs that now exist. e contractarian structure of interpretive authority
in ISA indicates that, in doing so, they should be guided by the choice
of law provisions in those IIAs. As discussed above, most IIAs make, in
eect, an unqualied choice of “international law” as the governing law.
How should a tribunal interpret an IIA subject to such an unqualied
choice of international law? is section sets out some brief rules of
thumb as to how tribunals should proceed in this, the most common
choice of law scenario under existing IIAs.
I argue that an unqualied choice of international law implies three
more specic choices, all of which reect the presumptive intention of the
states party to the IIA as to how the IIA should be interpreted. First, (and
least controversial) it constitutes a choice of international law rules of
interpretation, as expressed in the VC LT . Second, it constitutes a choice
of all international law, including customary and treaty law from outside
the IIA and outside the IIL context – at a minimum, all of the sources
of law envisaged by Article 38 of the ICJ Statute. ird, it constitutes
a choice of only international law – which means, for example, that
tribunals have no power to conduct a comparative public law analysis
to determine the meaning of treaty obligations.110 Regardless of whether
a comparative public law analysis would increase the real or perceived
legitimacy of ISA, it is outside the arbitral remit.
In what follows, I will discuss these three points in more detail. In
the process, I will show how an unqualied choice of international law
109. Roberts, “Power and Persuasion”, supra note 40 at 192.
110. As urged by commentators such as Kingsbury & Schill, supra note 4;
see Roberts, “Power and Persuasion”, ibid; see also note 4, supra, and
accompanying text.
257
(2017) 3(1) CJCCL
as the governing law mandates an extensive approach to interpretation –
one that draws on sources of international law beyond the treaty text and
that gives voice to the object and purpose of the treaty.
It is undisputed that international law rules of interpretation as
expressed in Articles 31-32 of the VC LT apply to the interpretation
of IIAs,111 although tribunals are inconsistent in how they apply those
rules and may pay mere lip service to the VC LT .112 What remains more
contested is the role of international law beyond the VC LT in interpreting
IIAs and lling gaps within them. is is the question of restrictive versus
extensive interpretation, which lies at the heart of interpretive disputes
in many areas of international law adjudication: whether treaties such as
IIAs should be interpreted restrictively, according to the literal meaning
of their text and in isolation from broader international law except where
necessary to ll gaps, or extensively, with broad reference to the treaties’
object and purpose and to international law more generally.
Given the incomplete nature of most IIAs, including the recently-
negotiated ones, some resort to international law beyond the IIA is
unavoidable and uncontroversial. A good example is the customary
international law rules on attribution described by the International
Law Commission in its Draft Articles on the Responsibility of States for
Internationally Wrongful Acts with Commentaries.113 ese are frequently
111. J Romesh Weeramantry, Treaty Interpretation in Investment Arbitration
(Oxford: Oxford University Press, 2012) at 24-30.
112. Ibid at 157-64 (observing that, while declaring that they would apply
the VC LT , tribunals often rely instead on fairness, policy implications,
practical consequences, eciency, or reasonableness); see also Mahnoosh
H Arsanjani and W Michael Reisman, “Interpreting Treaties for the
Benet of ird Parties: the ‘Salvors Doctrine’ and the use of Legislative
History in Investment Treaties” (2010) 104:4 American Journal of
International Law 597 at 599; omas Wälde, “Interpreting Investment
Treaties: Experiences and Examples” in Christina Binder et al, eds,
International Investment Law for the 21st Century: Essays in Honour of
Christoph Schreuer (Oxford: Oxford University Press, 2014) (observing
that “[i]t is dicult to nd a tribunal which formally and properly applied
the Vienna Rules step by step” at 746).
113. ILC, 54th Sess, UN Doc A/56/10 (2001) arts 4-5.
258
Karton, Law and Interpretive Authority in Investor-State Arbitration
cited by ISA tribunals and, to my knowledge, no commentator has
criticized tribunals for citing them.
But beyond such “easy cases”, debate continues. Ongoing
disagreement as to restrictive versus extensive interpretation may be
justied with respect to permanent international judicial bodies, but
given the contractarian structure of interpretive authority in ISA, an
unqualied choice of international law in an IIA mandates an extensive
approach to interpretation of the IIA. International law from outside the
applicable IIA should be applied not only to shed light on the meaning of
the IIA – to resolve disagreements as to the meaning of the text – but also
as a source of obligations of the parties to the arbitration beyond those
created by the IIA. is may appear, at rst glance, to be a pro-investor
position, since only states, not investors, have obligations under IIAs
and, more generally in international law, non-state entities enjoy rights
more than they incur obligations. However, international law principles
may also limit states’ obligations or provide excuses for breaches of treaty
obligations.
One’s attitude toward the interpretation and application of
international law is inextricable from one’s attitude toward the
independence of international law in relation to its main subject: states.
If international law is always and entirely a product of state consent, and
merely determines the reciprocal rights and duties of states that belong
to an international community without limiting their sovereignty,114
then the interpretation and application of international law should be
limited to what is explicitly mentioned in the text of treaties and in
written instruments in general. If, on the other hand, international law
is an autonomous legal system that requires the consent of its subject
to exist and determine its rules and principles, but that may have an
impact beyond the express consent of states, then its interpretation
and application might extend beyond the explicit words of written
114. See e.g. James Leslie Brierly, e Law of Nations: An Introduction to the
International Law of Peace (Oxford: Clarendon Press, 1963) at 1 (this may
be described as the “classical view”); Lassa Francis Lawrence Oppenheim,
International Law: A Treatise, vol I (Peace) (New York: Longman, Green,
1912) at 366-67.
259
(2017) 3(1) CJCCL
instruments and the clearly-dened customary law of state practice.115
e VC LT crystalizes the understanding that treaties should be
interpreted based on “their context and in the light of its object and
purpose”.116 It also recognizes that treaties, like other legal instruments,
are not self-executing, nor are can they ever be entirely autonomous;
they must therefore be interpreted within their broader legal context,
including preambles, annexes, and agreements relating to treaties, as well
as instruments connected to them and accepted by the parties.117 Finally,
the V C LT requires that, in addition to the treaty’s context, subsequent
agreements between the parties, or practice regarding the treaty’s
interpretation and “relevant rules of international law applicable in the
relations between the parties” must also be taken into consideration.118
is is not to say that the VC LT is entirely contextualist in its
approach; to the contrary, the rst rule of VC LT Ar ticle 31 is that treaties
should be “interpreted in good faith in accordance with the ordinary
meaning to be given to the terms of the treaty”. At the same time, though,
Article 31 does not impose pure textualism.119 e ordinary meaning
of the treaty terms must be considered in their context and in light of their
object and purpose. In this way, the VCLT adopts an approach that sits
between pure textualism and pure contextualism. Most importantly for
present purposes, the VC LT directs adjudicators to consider more than
just the immediate context of the treaty – the circumstances surrounding
its conclusion, including its drafting history – and to take into account
115. See Antonio Augusto Cançado Trindade, International Law for
Humankind: Towards a New Jus Gentium (Leiden: Martinus Nijho,
2010) at 429-49. On developments in treaty interpretation generally, see
Richard K Gardiner, Treaty Interpretation (New York: Oxford University
Press, 2008) and Gerald Fitzmaurice, Olufemi Elias & Panos Merkouris,
eds, Treaty Interpretation and the Vienna Convention on the Law of Treaties:
30 Years On (Boston: Martinus Nijho Publishers, 2010).
116. VC LT , supra note 66, art 31(1).
117. Ibid, art 31(2).
118. Ibid, arts 31(3)(a)-(c).
119. See e.g. Julian Davis Mortenson, “e Travaux of Travaux: Is the Vienna
Convention Hostile to Drafting History?” (2013) 107:4 American Journal
of International Law 780.
260
Karton, Law and Interpretive Authority in Investor-State Arbitration
the treaty’s broader political and legal context, including conduct of the
states parties not directly related to the treaty, as well as the full scope of
international law rules applicable between the states parties.
Some argue that the VC LT mandates a two-stage interpretive process
similar to the process of contractual interpretation in many common law
jurisdictions; in the rst stage, only the ordinary meaning of the words
is to be considered; if, and only if, that meaning is vague or ambiguous
or leads to results that contradict other provisions of the treaty may the
adjudicator proceed to the second stage of interpretation, in which it
may consider the treaty’s object and purpose.120 ere is some support
for this position in the travaux of the International Law Commission at
the time of the drafting of the VC LT , as some ILC members emphasized
“the primacy of the text as the basis for the interpretation of a treaty,
while at the same time giving a certain place to extrinsic evidence of the
intentions of the parties and to the objects and purposes of the treaty as
means of interpretation”.121
However, applying this line of argument to the VCLT itself shows that
it does not call for such strict textualism. Article 31(1) of the VC LT states
simply that “[a] treaty shall be interpreted in good faith in accordance
with the ordinary meaning to be given to the terms of the treaty in their
context and in the light of its object and purpose”. It says nothing about
distinct stages of interpretation, nor does it limit the object and purpose
of the treaty to a subsidiary or supplementary role. Rather, the ordinary
meaning of the treaty text should be considered together with (“in light
of”) the treaty’s object and purpose. e fact that other interpretive aids
are expressly relegated to a subsidiary status122 shows that the treaty’s
context and its object and purpose should not be left to a second stage of
interpretation.
120. See e.g. Ulf Linderfalk, On the Interpretation of Treaties: e Modern
International Law as Expressed in the 1969 Vienna Convention on the Law
of Treaties (Dordrecht: Springer, 2007) at 203.
121. Draft Articles on the Law of Treaties with Commentaries, ILC, 18th Sess,
UN Doc A/CN.4/191 (1966) 187 at 218 [Draft Articles on Treaties].
122. VC LT , supra note 66 (specically, “the preparatory work of the treaty and
the circumstances of its conclusion”, art 32).
261
(2017) 3(1) CJCCL
Indeed, the International Law Commission itself pointed out that
“the ordinary meaning to be given to the terms of the treaty in their
context and in the light of its object and purpose should be the rst
element to be mentioned”.123 Consequently, preference must be
given to the ordinary meaning, which “is not to be determined in the
abstract but in the context of the treaty and in the light of its object
and purpose”.124 Article 31(1) does not foreclose consideration of the
treaty’s object and purpose in the rst instance; all it forecloses is “an
investigation ab initio into the intentions of the parties”.125 e V CLT
therefore prescribes a method of interpretation that takes into account,
together with the ordinary meaning of the text, the purpose of a treaty
as a whole, including its preamble, and the area being interpreted
(international law in general or its specic subdivisions), including its
evolution through time (“emergent purpose”). ese contextual materials
will be particularly useful to adjudicators when they are called upon to
ll gaps and apply treaties to circumstances not considered at the time of
the treaty’s conclusion.
An example of this approach in action can be seen in the jurisprudence
of international human rights courts. e object and purpose of human
rights treaties is, generally speaking, the eective protection of individuals.
Accordingly, the Inter-American Court of Human Rights (“IACtHR”)
and the European Court of Human Rights (“ECtHR”) have both
developed their jurisprudence by going beyond the mere grammatical
interpretation of their respective treaties and seeking interpretations that
advance such eective protection.
In Loizidou v Turkey,126 the ECtHR armed that “the object and
purpose of the Convention as an instrument for the protection of
individual human beings requires that its provisions be interpreted
and applied so as to make its safeguards practical and eective”.127 It
added that a restrictive approach to the interpretation of the European
123. Draft Articles on Treaties, supra note 121 at 220.
124. Ibid at 221.
125. Ibid.
126. (1995), 20 ECHR (Ser A) 99.
127. Ibid at para 72.
262
Karton, Law and Interpretive Authority in Investor-State Arbitration
Convention on Human Rights would “seriously weaken” the role of the
ECtHR and would “diminish the eectiveness of the Convention as a
constitutional instrument of European public order (ordre public)”.128
e corollary position that the Convention on Human Rights is a “living
instrument” that requires dynamic interpretation to ensure that it
continues to achieve its object and purpose, rst expressed in Tyler v the
United Kingdom,129 is continually rearmed by the ECtHR.130
Arguably inuenced by the ECtHR,131 the IACtHR has gone even
further in crystalizing and extending the notion that human rights
treaties must be interpreted to be eective in protecting individuals.
Invoking what it called the “pro homine principle”,132 the IACtHR has,
for example, held that States cannot damage an individual’s “life plan”
without breaching their international law obligations;133 that indigenous
communities have special rights to their lands;134 that the Court may take
into consideration indigenous legal traditions;135 and that international
law prohibits forced disappearances.136
All of these judgements advanced the protection of human rights
beyond the initial set of rights spelled out by the Inter-American Convention
on Human Rights. e IACtHR, in its jurisprudence interpreting treaties
pro homine, has made reference to other treaties, and to principles
codied or developed in the context of international humanitarian law,
128. Ibid at para 75.
129. (1978), 21 ECHR (Ser A) 612 (the Convention “is a living instrument
which … must be interpreted in the light of present-day conditions” at
para 31).
130. See e.g. Rantsev v Cyprus and Russia, No 25965/04, [2010] I ECHR 1, at
paras 273-75.
131. Magnus Killander, “Interpreting Regional Human Rights Treaties” (2010)
7:13 SUR – International Journal of Human Rights 145.
132. Sometimes called the pro personae principle.
133. Loayza Tamayo Case (Peru) (1998), Inter-Am Ct HR (Ser C) No 42 at
paras 144-54.
134. e Mayagna (Sumo) Awas Tingni Community Case (Nicaragua) (2001),
Inter-Am Ct HR (Ser C) No 79.
135. Aloeboetoe et al Case (Suriname) (1993), Inter-Am Ct HR (Ser C) No 11.
136. Villagrán-Morales et al Case (Guatemala) (1999), Inter-Am Ct HR (Ser C)
No 63.
263
(2017) 3(1) CJCCL
international environmental law, international investment law, and
the international law of economic, social and cultural rights.137 e
IACtHR has held that international human rights law is a part of public
international law, but is lex specialis in cases that come before the Court;
that is, international human rights law prevails over conicting principles
of general public international law, but only when its provisions are more
favourable to the rights bearers in a specic case.138
ISA tribunals should take a page from the international human rights
courts (although not necessarily from international human rights law).
e primary lesson is that an IIA which makes an unqualied choice of
international law cannot be interpreted according to the text of the treaty
alone; instead, tribunals must interpret and apply it in such a way as to
preserve its eectiveness, including by drawing on doctrines that do not
appear in the treaty and may have been developed in dierent contexts.
us far, many tribunals have been tentative in their treatment of
international law beyond the investment context, restricting themselves
to citing the VC LT , the ILC’s Draft Articles on State Responsibility, and
a small handful of famous judgments like the Barcelona Traction139
ruling on nationality and locus standi or the Chorzów Factory formula for
compensation.140
Such hesitation is unwarranted. Interpretation in international law
is a necessarily subjective process, bound up as it is with uncertainty as
to the sources of international law. But the IIA’s choice of law provision
points the way. As Kelsen put it, “the work of interpretation is one of
discovering the intention of the parties not only by reference to rules of
interpretation, but to rules of international law bearing upon the subject-
137. See Lucas Lixinski, “Treaty Interpretation by the Inter-American Court of
Human Rights: Expansionism at the Service of the Unity of International
Law” (2010) 21:3 European Journal of International Law 585 at 603.
138. Ibid [emphasis in original].
139. Barcelona Traction, Light and Power Company, Limited (Belgium v Spain),
[1964] ICJ Rep 6.
140. Factory At Chorzów Case (Germany v Poland) (1928), PCIJ (Ser A) No 17.
264
Karton, Law and Interpretive Authority in Investor-State Arbitration
matter of the disputed contractual stipulation”.141 If states call for a treaty
to apply something as so open-ended as “international law” simpliciter,
they thereby acquiesce, at minimum, to the arbitrators’ determination of
the sources of that law.142 ISA tribunals should therefore take a wide view
as to the scope of applicable norms in any given investment arbitration.
In doing so, arbitrators will “give prudent eect to the truest expression
of state intent in sacricing sovereignty vis-à-vis foreign investments
(thereby potentially fostering greater state commitment to the system)”.143
None of this is to suggest that all international law is always relevant,
or that the treaty text is subordinate to general international law. e IIA
remains lex specialis. Just like a contract in private law, it constitutes a
derogation from all conicting rules of general international law except
non-derogable ius cogens. As an example of such derogations, Kurtz cites
the customary rules on diplomatic protection,144 which entitles states to
bring actions against other states for injuries caused to their nationals
by internationally wrongful acts;145 these principles are excluded by
141. Hans Kelsen, Pure eory of Law (2d English ed) (Berkeley: University of
California Press, 1970) at 355.
142. is point may also be shown by a counter-example, the phrase in many
IIAs negotiated by the United States that the fair and equitable treatment
standard be dened according to the customary international law standard
for minimum treatment of aliens. See e.g. the nal negotiated text of the
Trans-Pacic Partnership Agreement, 4 February 2016 (not yet in force),
online:
partnership/tpp-full-text> [Trans-Pacic Partnership Agreement 2016].
Article 9.6(2) states that, “for greater certainty”, the obligation of states
to provide to covered investments fair and equitable treatment and full
protection and security “prescribes the customary international law
minimum standard of treatment of aliens as the standard of treatment to
be aorded to covered investments. e concepts of “fair and equitable
treatment” and “full protection and security” do not require treatment in
addition to or beyond that which is required by that standard, and do not
create additional substantive rights.
143. Kurtz, supra note 8 at 281.
144. Trans-Pacic Partnership Agreement 2016, supra note 142 at 283-84.
145. Draft Articles on Diplomatic Protection with Commentaries ILC, 58th Sess,
UN Doc A/61/10 (2006) art 1; see also Ahmadou Sadio Diallo (Guinea v
Congo), Preliminary Objections, [2007] ICJ Rep 582, at para 39.
265
(2017) 3(1) CJCCL
the ICSID Convention, which prohibits the extension of diplomatic
protection to individuals who have brought direct claims against other
states under IIAs subject to the ICSID Convention, “unless such other
Contracting State shall have failed to abide by and comply with the
award rendered in such dispute”.146
However, to the extent that the treaty is silent, or is vague or ambiguous
(all of which are common in the existing IIAs), tribunals should consider
the full sweep of international law to ll the gap or clarify the vagueness
or ambiguity. In this exercise, tribunals should not limit themselves to
doctrines specic to IIL or even to international economic law. IIL norms
should only take precedence over other international law norms to the
extent that the IIL norm is more favourable to the object and purpose of
the IIA. us, tribunals’ authority to pluck principles from other areas of
international law is not unlimited. ey may venture outside the treaty
text, but must remain inside the treaty’s legal and political framework.
In this light, it is important to remember that the object and purpose
of IIAs, and of investment arbitration in particular, are to ensure fair,
predictable, and non-discriminatory treatment of foreign investments,
no more and no less.147 It has been argued that, since investors gain direct
rights as subjects of IIAs, just like individuals gain direct rights as subjects
146. Ibid, art 27(1).
147. As the Suez/Vivendi tribunal put it, “a recognized goal of international
investment law is to establish a predictable, stable legal framework for
investments”. Suez, Sociedad General de Aguas de Barcelona, SA and
Vivendi Universal, SA v Argentine Republic (2015), Case No ARB/03/19 at
para 189 (International Centre for Settlement of Investment Disputes).
266
Karton, Law and Interpretive Authority in Investor-State Arbitration
of human rights treaties, the purpose of IIAs is to protect investor rights.148
Indeed, especially in the earlier years of ISA, some tribunals “seemed not
solely to simply assume that IIAs were concluded exclusively to protect
investors, but also to make this assumption their ultimate guide to the
interpretation of the agreement, establishing the peculiar presumption in
dubio pro investore”.149
However, the analogy to human rights is misplaced. ese tribunals
confused the promotion of investment with the promotion of investors.
e purpose of IIAs is not to protect the prot margins of investors,
although of course the promotion of investment is an intended and
welcome consequence. Rather:
from a teleological point of view, investment treaties were initially concluded in
order to induce FDI ows to states with developmental needs…. is diers a
lot from the very idea at the heart of human rights or other individual-centered
148. See generally Martins Paparinskis, “Analogies and Other Regimes of
International Law” in Zachary Douglas et al, eds, e Foundations
of International Investment Law: Bringing eory into Practice
(Oxford: Oxford University Press, 2014) at 79-85. See also Martins
Paparinskis, “Investment Treaty Arbitration and the (New) Law of
State Responsibility” (2013) 24 European Journal of International Law
617 at 617; Anastasios Gourgourinis, “Investors’ Rights Qua Human
Rights? Revisiting the ‘Direct’/‘Derivative’ Rights Debate” in Malgosia
Fitzmaurice and Panos Merkouris, eds, e Interpretation and Application
of the European Convention of Human Rights: Legal and Practical
Implications (Leiden: Martinus Nijho, 2013) at 147; Francisco Gonzáles
de Cossio, “Investment Protection Rights: Substantive or Procedural?”
(2010) 25 ICSID Review 107. e issue has been taken up in some
arbitrations, with mixed outcomes; see e.g. Wintershall Aktiengesellschaft
v Argentina (2008), ICSID Case No ARB/04/14, at para 110; Archer
Daniels Midland Company and Tate & Lyle Ingredients Americas, Inc v
Mexico (2007), ICSID Case No ARB(AF)/04/05, at paras 161-80; ADF
Group Inc v USA (2003), ICSID Case No ARB(AF)/00/1, at para 152;
Corn Products International, Inc v United Mexican States (2008), ICSID
Case No ARB(AF)/04/1, at paras 167-179; and Cargill Inc v Mexico
(2009), ICSID Case No ARB(AF)/05/2, at paras 403-28.
149. Methymaki, supra note 72 at 3 (this is intended to be analogous to the pro
homine principle adopted by human rights courts); see above, notes 122-
27 and accompanying text for clarication.
267
(2017) 3(1) CJCCL
regimes (eg, the law of consular relations), namely the protection of individuals
because of considerations of humanity and due process.150
Equally, the object and purpose of IIAs is not to insulate states from
liability, although of course states can and should draft IIAs to preserve
their ability to regulate in the public interest. e reason states enter
into investment treaties is analogous to the reason states enforce contracts
within their borders: to ensure the predictable and fair enforcement of
legal obligations, with the expectation that a private market will ourish
as a result. erefore, in interpreting and applying IIAs, tribunals should
adopt the interpretation that most fosters a fair, predictable, and non-
discriminatory (not necessarily protable) regulatory environment for
foreign investment.
VI. Conc lusion
I have argued in this article that the interpretive authority of investment
arbitrators is constrained primarily by the choice of law provisions in the
IIAs that they interpret. e contractarian structure of arbitral authority
in ISA means that arbitrators must not only apply the chosen laws as
opposed to other laws, but must also interpret the chosen laws in the
ways intended by the states parties to the IIA. I oered some rules of
thumb for determining the states parties’ presumed intention in (the
majority of) cases where they have provided no guidance beyond naming
“international law” as the law governing the dispute. I also oered
something of a roadmap, drawing from the same theoretical principles
about the structure of arbitral authority, for states to more eectively
guide the discretion of arbitrators.
e main benet of the proposed approach is greater certainty and
consistency, assuming tribunals were to follow it – not consistency in
the sense of absolute uniformity, which is impossible and probably
undesirable, but a consistency of approach that is most likely to conform
to the intentions of states when they issue regulations that may aect
investors covered by the state’s international law obligations.
More broadly, the proposed approach would also help to rationalize
150. Methymaki, ibid at 16.
268
Karton, Law and Interpretive Authority in Investor-State Arbitration
the interests of states and investors. On the one hand, investors do
have legitimate expectations from investment treaties and the dispute
resolution systems created by them, expectations on which they should
be able to rely with some degree of certainty. On the other hand, ultimate
control remains with states, as it should: they are the ones issuing a
standing oer to arbitrate, and are entitled to set the terms of that oer.
ey can restrict or expand the range of available rules of decision in the
arbitration agreement, or simply dene it more precisely. ey may issue
joint interpretations of treaty terms that are binding on tribunals, even
after disputes arise. To the extent that such actions would narrow the range
of claims that investors may make, or reduce the damages recoverable
for breaches, they would still benet investors to the extent that they
would create greater certainty. Widespread renunciation of IIAs fueled by
political backlash, or even just a drastic narrowing of states’ obligations
under IIAs, may be even more harmful to the global investment climate
and the global rule of law.
To continue reading
Request your trial