JOIA2009002

Journal of International Arbitration 26(1): 25–58, 2009.
© 2009 Kluwer Law International. Printed in The Netherlands.
“Back to the Future” for Investor-State Arbitrations:

Revising Rules in Australia and Japan to
Meet Public Interests “Back to the Future” for Investor-State Arbitrations: Revising Rules in Australia and Japan to Meet Public Interests
Luke Nottage and Kate Miles*

Journal of International Arbitration
Michael
Moser
Dominique
The more things change, the more some stay the same. This article first highlights renewed concerns about
Journal of InternationalHascher
Arbitration
2009 Volume 26 Issue 1
delays and, especially, costs in international commercial arbitration (ICA). Many now urge quite radical solu-
tions to make ICA more efficient, including allowing parties to authorize arbitrators to facilitate settlement
(Arb-Med). At the same time, there are growing calls for more transparency, non-party participation, and other
rule changes to promote the legitimacy of the burgeoning field of investor-state arbitration (ISA). Such reforms
are justified by the greater variety of public interests involved in ISA, despite some possible losses in efficienc.
We should resist a tendency simply to extend the solutions devised or proposed for ICA, particularly in the
form of Rules of arbitral institutions, to contemporary ISA. However, some reforms incorporating proper safe-
guards may also be advisable in both fields, such as Arb-Med processes, or arbitrator remuneration providing
better incentives to streamline proceedings. Many reforms can be implemented by institutions devising tailored
ISA Rules, to be added as options for investors in bilateral or regional investment treaties or Free Trade Agree-
ments (FTAs). Our article therefore proposes a variety of improvements.These are based on comparisons of the
main Rules adopted for ISA (ICSID and the UNCITRAL Rules), the arbitration Rules of institutions like
the Australian Centre for International Commercial Arbitration (ACICA) and the Japan Commercial Arbi-
tration Association ( JCAA), and some of the provisions already found in Australia’s FTAs or governing trade
disputes under the World Trade Organization (WTO) system. Hopefully, these improvements will enable ISA
to keep developing through bilateral initiatives such as the proposed Australia-Japan FTA, emerging regional
initiatives, and ultimately a new multilateral framework for investment. Although reforms are currently needed
to bolster the legitimacy of ISA, longer-term reforms may instead re-emphasize efficiency, rather like ICA has
done after decades spent achieving global acceptance.
1. International Commercial Arbitration Backdrop: Delays and Costs

International commercial arbitration (ICA) is facing another round of soul-searching,
the current focus of which is the increasing costs and delays of proceedings. As one
prominent British practitioner confessed:
* Respectively, Associate Professor and Lecturer, Sydney Law School. Kate Miles comes to this topic from the
perspective of public international law. Luke Nottage comes to it from the perspective of international commercial
arbitration and private law, and serves on the ACICA Rules subcommittee that has begun examining Investor-State
Arbitration Rules and other topics. Both take full responsibility for the views expressed in this article.
This is a revised version of a paper presented at the ANZSIL conference, Canberra, June 28, 2008. We pre-
sented the skeleton of this article to the Australasian Forum for International Arbitration, Sydney, November 23,
2007; and a full draft to the Sydney Centre for International Law, April 2, 2008. We thank participants for helpful
comments or information, especially Malcolm Holmes QC, Damian Sturzaker, and Dr. Brett Williams; Prof.
Richard Garnett and other members of ACICA’s subcommittee examining various changes or additions to its
Rules; Prof. Okuda Yasuhiro and members of the Investment Treaty Arbitration Study Group of the Asian Society
of International Law’s Japan Chapter; Assoc. Prof. Susan Franck, Profs. Tom Ginsburg and Campbell McLachlan,
and Hunter Nottage. A version of the third Part of this article appears in 20 J.C.C.A. Newsletter 1 ( July 2008),
available at <http://jcaa.or.jp/e/> and another is forthcoming (in Japanese translation) in J.C.A. Jyanaru.
26 journal of international arbitration
We are finding increasingly that given lawyers’ remuneration at high hourly rates and the proce-
dures and methods employed, particularly by common lawyers, largely from the City of London
and New York, that bills of costs in the millions of dollars are commonplace even in cases of only
moderate length and complexity. It is now simply uneconomic for small and medium-sized com-
panies engaged in international business to go to international arbitration in the main Western
European cities under the aegis of leading institutions for the resolution of their disputes.1
This problem of spiralling costs in ICA may be particularly acute in the United
Kingdom. Despite the Arbitration Act 1996,2 the legal and business community still
appears to be struggling to break from a long tradition of judicial supervision of arbitra-
tion.3 English lawyers have long tended to “formalize” arbitration.4 But the bane of bur-
geoning costs appears more widespread. A recent study by the University of London and
Pricewaterhouse Coopers found that seventy out of eighty in-house counsel even at
major international corporations cited cost as one of their top three concerns. Half of
those interviewed ranked it as their primary concern. This disadvantage of ICA was con-
sidered far more important than delays.5 Some may be starting to appreciate one lesson
from waves of civil procedure reforms carried out since the late 1990s in a number of
states including the United Kingdom and Japan: speeding up proceedings does not nec-
essarily translate into major cost savings.6 A common response by lawyers, for example, is
to “front-load” the expenses to meet tougher timetables. Perhaps unsurprisingly, then, cost
savings have not readily followed efforts to expedite ICA by revisions in the late 1990s to
the Rules of the world’s major arbitration institutions and other efforts by the broader
arbitration world.7
Consequently, a number of proposals are emerging to counter the over-formalisation
of ICA. And ICA is well-placed to reinvent itself in this way for the twenty-first century.
Not only is commercial arbitration rooted primarily in the economic interests of private
parties and in private autonomy, but it can also draw on a broader pool of experiences
and expectations, now that more countries have come in from the “periphery” of the
arbitration world.8
At its “core,” for example, the International Chamber of Commerce (ICC) pub-
lished a Report on Techniques for Controlling Time and Costs in Arbitration,9 noting that
2% of costs comprised the administration fees charged by the ICC and 16%, the fees of
1
A. Marriott, Breaking the Deadlock, 22 Arb. Int’l 411, 413– 14 (2006), available at <www.ialecture.com>.
2
See, e.g., 23 Arb. Int’l (2007).
3
Marriott, supra note 1; see also Y. Dezalay & B.G. Garth, Dealing in Virtue: International Commercial
Arbitration and the Construction of a Transnational Legal Order (1996).
4
J. Flood & A. Caiger, Lawyers and Arbitration: The Juridification of Construction Disputes, 56 Mod. L. Rev. 412
(1992).
5
Available at <www.pwc.com/arbitrationstudy/>; L. Mistelis, International Arbitration: Corporate Attitudes and
Practices, 15 Am. Rev. Int’l Arb. 525 (2004).
6
M. Zander, The State of Justice (2000); L. Nottage, Civil Procedure Reforms in Japan: The Latest Round, 22
Ritsumeikan L. Rev. 81 (2005).
7
L. Nottage, The Procedural Lex Mercatoria: The Past, Present and Future of International Commercial Arbitration
CDAMS Discussion Paper (2003), available at <www.cdams.kobeu.ac.jp/archive/dp03–1.pdf>.
8
L. Nottage, The Vicissitudes of Transnational Commercial Arbitration and the Lex Mercatoria: A View from the Periphery,
16 Arb. Int’l 53 (2000).
9
Publication No. 843, 2007, available at <www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>.
“back to the future” for investor-state arbitrations 27
the arbitrators. The Report concentrated on various ways in which arbitrators and parties
can manage proceedings more efficiently to reduce the remaining bulk of costs involved
in parties presenting their case. However, one international law firm from the United
States points out that the proposals involve parties carefully agreeing beforehand on cost-
reduction techniques, and identifies some more innovative responses.10 They note, for
example, that “Many arbitrators, particularly in Switzerland, Germany, and Austria, are
now more active and confident in promoting, and even ‘pushing,’ settlement during the
arbitration procedure. This is still controversial, however, and it is not possible to identify
it as a clear trend.”11
Nonetheless, more recently established arbitration institutions, such as the Australian
Centre for International Commercial Arbitration (ACICA), are now considering ways to
encourage parties and their legal advisors to experiment with such hybrid “Arb-Med”
processes, involving arbitrators in settlement facilitation. The English law tradition, unlike
strands within the civil law tradition (including parts of Asia),12 has tended to oppose such
processes as contrary to natural justice and arbitrator impartiality. One compromise, how-
ever, is to encourage parties to agree on Arb-Med without allowing “caucusing,” a prac-
tice which entails private meetings between the arbitrator and each party.13 Risks of
challenge seem minimal compared to the potential benefits, in the present era of height-
ened concerns about the costs of ICA.
2. Public Interest Concerns in Investor-State Arbitration

History is also repeating itself somewhat in an overlapping, yet distinct, field:
investor-state arbitration (ISA). Many of the large-scale arbitrations of the 1950s and 1960s
involved claims of expropriation against newly independent states.14 Investment arbitra-
tion was often conducted following ad hoc rules negotiated between the parties (some-
times involving the arbitrators), following underlying agreements to refer the matter to
arbitration. Such experience generated momentum to conclude the 1958 New York
Convention on the Recognition and Enforcement of Foreign Arbitral Awards (NYC),15
and then the 1976 UNCITRAL Arbitration Rules (UR).16 Although mainly reflecting
experiences in ad hoc arbitration, the UR were first used extensively in the 1980 Iran-United
10
Winston & Strawn L.L.P., What Can Be Done About Arbitration Costs?, Briefing: International Arbitration
Practice (September 2007), available at <www.winston.com/siteFiles/publications/Arbitration_Costs.pdf >.
11
Id. at 4.
12
See, e.g., W.S. Chang, Combination of Arbitration with Conciliation and Remittance of Awards, with Special Reference
to the Asia-Oceania Region, 19 J. Int’l Arb. 51 (2002).
13
L. Nottage, Arb-Med in Australia: The Time Has Come, 5 Austl. A.D.R. Rep. 18 (2007), available at
<www.arbitrators.org.au>; G. Kaufmann-Kohler, When Arbitrators Facilitate Settlement: Towards a Transnational Standard,
26 Arb. Int’l (forthcoming 2009), available at <www.ialecture.com>.
14
K.-H. Boeckstiegel, Settlement of Disputes Between Parties from Developing and Industrial Countries, 15 ICSID
Rev. 275 (2000).
15
Convention on the Recognition and Enforcement of Foreign Arbitral Awards, opened for signature June 10,
1958, 330 U.N.T.S. 3, 21 U.S.T. 2517, T.I.A.S. No. 6997.
16
Arbitration Rules of the United Nations Commission on International Trade Law, U.N. Doc. A/Res/31/98
(1976).
States Claims Tribunal (again, often dealing with investment disputes).17 Various arbitra-
tion centres, especially newer ones like the Kuala Lumpur Regional Centre and ACICA,
then made them available to parties either directly or with modifications tailored to insti-
tutional arbitration. Arbitral institutions also attracted a broad array of commercial dis-
putes, beyond the state-oriented focus of investment arbitration, thus expanding the ICA
world as well.
In conjunction with this institutional expansion, the last few decades have also wit-
nessed a period of extensive investment liberalization. As well as arbitration provisions
being added into investment contracts between firms and host states, several countries
have enacted legislation on foreign direct investment (FDI) consenting to submission to
arbitration of any investor claims. Many more states have entered into bilateral investment
treaties (BITs), and now some broader free trade agreements (FTAs) that include invest-
ment chapters and submission to arbitration clauses. These provisions are directed at the
investor’s home country, but are essentially for the benefit of investors, who can invoke
these agreements to initiate arbitration proceedings against the host state. A commitment
to arbitration contained in a BIT is now the most common route for generating the
“consent” to arbitration on the part of host states, as required by Article 25(1) of the 1965
Washington Convention on the Settlement of Investment Disputes between States and
Nationals of Other States.18 Almost all BITs allow foreign investors to commence ISA
under the arbitration Rules of the International Center for the Settlement of Investment
Disputes (ICSID), set up by the Convention under the aegis of the World Bank. How-
ever, in many treaties the foreign investor is also given the alternative of invoking the
UR, and, in some treaties, the ICC, the Rules of the Stockholm Chamber of Com-
merce, or even (if both sides agree) other institutional Rules.
There has also been an exponential growth in ISA over the last few years. Until
1990, only a small number of ISA cases were decided under the ICSID Rules, but
around 250 had been filed by 2008, and around half of these are still pending.19 ISA is
also now a common talking point in major arbitration conferences and law journals.20
The large commercial law firms have identified it as a promising field of new work, even
though the world of ISA differs considerably from that of ICA. In ISA, the substantive
law applied is mostly public international law, not contract law, the traditional area of
expertise for these firms. Further, many arbitrators are in fact law professors. They tend
to be seen as more neutral, and as having a capacity to encompass the broader conceptu-
alization of issues that is needed in this important formative phase of ISA, as indeed was
required in the formative era of ICA itself.
17
K.-H. Boeckstiegel, Enterprise v. State: The New David and Goliath?, 23 Arb. Int’l 93 (2007).
18
Convention on the Settlement of Investment Disputes between States and Nationals of Other States, March
18, 1965, 575 U.N.T.S. 159; 17 U.S.T. 1270.
19
A.H. Ali & A. de Gramont, ICSID Arbitration in the Americas, 2008 Arb. Rev. Americas, Global Arb. Rev.,
available at <www.globalarbitrationreview.com/handbooks/4/sections/7/chapters/50/icsid-arbitration-americas>;
ICSID, List of Pending Cases, available at <http://icsid.worldbank.org/ICSID/FrontServlet?requestType=Cas-
esRH&actionVal=ListPending>.
20
See, e.g., ICCA International Congresses 2002, 2004, and 2006 (conference volumes edited by Albert van
den Berg, mostly available at <www.kluwerarbitration.com>).
There still remains considerable uncertainty and debate about important procedural
and substantive issues in ISA. This is to be expected. An initial period of flux tends to
characterize new fields of dispute resolution, especially when high political and economic
stakes are involved. The first decade of dispute resolution under the World Trade Organ-
ization (WTO) framework is a case in point. The development of WTO law, however,
was assisted by a permanent appellate body able to review questions of law and legal
interpretation in decisions reached by the panels, which are appointed on an ad hoc basis
by disputing states. Many commentators now call for a similar mechanism to add more
consistency to awards in ISA (outlined below). Meanwhile, despite the lack of an appeals
facility, a disparate body of non-binding precedent is being built up and applied in the
field of ISA.21
The large volume of pending ICSID cases, however, has raised concerns. Host
states, as respondents in ISA, tend to be less concerned about delays than private investor
claimants. However, both sides are certainly worried about the growing costs of running
high-value and often politically-charged ISA cases.22 Although systematic empirical evi-
dence of tribunal and legal costs remains patchy,23 perceptions of ballooning costs remain
important as they tie into a broader “legitimacy crisis” in investment arbitration.24
This situation provides part of the backdrop to the United Nations’ review of the
1976 UR, formally underway in Vienna and New York since 2007. However, the UR
also remain important for regular ICA cases, either in ad hoc arbitrations or through insti-
tutions that allow parties to substitute the UR for their own Rules, or which (like
ACICA) have modelled those institutional Rules on the UR. Particularly in the early
stage of the UR revision discussions, therefore, there has been a certain tendency to
assume that the UR should simply be updated to incorporate provisions added since the
late 1990s to the Rules of major institutions, such as the ICC. Those measures generally
help to speed up proceedings and to reduce costs. Yet this tendency downplays the fact
that the UR are relevant to ISA as well as ICA, whereas the Rules of institutions like the
ICC are primarily for ICA. This would not be a problem if the features of ICA and ISA
were identical, but they are not, despite some overlaps and shared history.
In particular, ISA clearly involves a set of much stronger public interests, although
ICA also does preserve some public interest elements through, for example, its conceptions
21
G. Kaufmann-Kohler, Arbitral Precedent: Dream, Necessity or Excuse?, 23 Arb. Int’l 357 (2007); C. McLachlan
et al., International Investment Arbitration: Substantive Principles (2007). Lawyers and policy-makers from the
civil law tradition, or public international lawyers, may be less concerned about non-binding precedents than those
trained in the common law tradition.The latter, especially in its Anglo-Commonwealth subtradition, applies the doc-
trine of stare decisis in the hope of maximizing predictability, especially in private law matters. See Part II. A of L.
Nottage, The Japanisation of American Law? Substantive Similarities, Compared to Formal Anglo-New Zealand Law (Sydney
Law School Research Paper, (forthcoming, 2009)) available at <www.ssrn.com> with an earlier version available at
<www.law.usyd.edu.au/anjel/content/anjel_research_pub.html>.
22
Marriott, supra note 1.
23
S. Franck, Empirically Evaluating Claims About Investment Treaty Arbitration, 86 N. C. L. Rev. 1, 66–70 (2007).
24
S. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public Law Through Inconsistent Deci-
sions, 73 Fordham L. Rev. 1521 (2005).
of mandatory minimum standards of substantive and procedural “public policy.”25 The

International Institute for Sustainable Development (IISD) and others have recently high-
lighted that ISA often involves:
– public service sectors, such as the provision of water supplies, oil and gas, elec-
tricity, transport systems, telecommunications, and waste management services;
– government regulations (in substance or process) for the public benefit that
directly or indirectly affect private investors;
– governments increasingly subject to good governance obligations such as trans-
parency and due process; and
– large sums of money, both in the damages claimed and relating to costs of
defending disputes.26
The existence of the public interest element in ISA has led to calls for greater trans-
parency, broader stakeholder participation, and enhanced natural justice safeguards in ISA
proceedings.27
3. Incorporating Revised ISA Rules into Treaties or Rules

ICSID has already partly heeded such concerns by amending in 2006 some of its
arbitration Rules, which were already designed solely for ISA. For example, transparency
has been enhanced by allowing third parties to attend hearings if parties consent (Rule
32(2)), and by allowing arbitrators to accept amicus curiae briefs after consulting with the
parties (Rule 37(2)).28 Yet, somewhat ironically, ICSID itself did not widely publicize
25
G. Triggs, New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards 1958: 50th Anniver-
sary: Celebration, Reflection and Reform, Keynote address at the commemorative event co-hosted by ACICA and Free-
hills, Sydney, February 27, 2008; abridged in 6 Austl. A.D.R. Rep. (2008), available at <www.arbitrators.org.au>.
26
IISD, Revising the UNCITRL Arbitration Rules to Address Investor-State Arbitrations (2007),
available at <www.iisd.org/pdf/2007/investment_revising_uncitral_arbitration_september.pdf>.
27
See, e.g., K. Tienhaara, What You Don’t Know Can Hurt You: Investor-State Disputes and the Protection of the Envi-
ronment in Developing Countries, 6 Global Environm’l Politics 73 (2006); K. Miles, Bringing the Public Interest into
Private Business, in International Economic Law and National Autonomy: Convergence or Divergence?
(Meredith Lewis and Susy Frankel eds., forthcoming 2009) (Manuscript on file with Author); M. Sornarajah, The
Clash of Globalizations and the International Law on Foreign Investment, 10 Can. Foreign Pol’y 1 (2003); G. Van
Harten, Investment Treaty Arbitration and Public Law (2007); H. Mann et al., IISD Model International
Agreement on Investment for Sustainable Development (2d ed. 2006), available at <www.iisd.org/pdf/2005/
investment_model_int_handbook.pdf>; IISD (A. Cosby et al.), Investment and Sustainable Development: A
Guide to the Use and Potential of International Investment Agreements (2004), available at <www.iisd.org/
pdf/2004/investment_invest_and_sd.pdf>. Many of these proposals come from those with a background in public
international law. However, we now find growing sympathy from leaders in the field of ICA. See, e.g., G.A. Alvarez
& W.W. Park, The New Face of Investment Arbitration: Capital Exporters and Host States Under NAFTA Chapter 11, in
International Commercial Arbitration: Important Contemporary Questions 392 (A. van Den Berg ed., 2003);
N. Blackaby, Investment Arbitration and Commercial Arbitration (or the Tale of the Dolphin and the Shark), in Pervasive
Problems in International Arbitration 217 (L.A. Mistelis & J.D.M. Lew eds., 2006).
28
M. Egonu, Investor-State Arbitration Under ICSID: A Case for Presumption Against Confidentiality? 24 J. Int’l
Arb. 479, 483 (2007). For a succinct summary of other changes, see also J. Savage, Investment Treaty Arbitrations and
Asia: Review of Developments in 2005 and 2006, 3 Asian Int’l Arb. J. 1, 13 (2007).
how and why some changes were made and not others. IISD argues for more dramatic
reforms. The Institute envisages changes to the UR, and has proposed a draft model for
any future multilateral agreement on investment (MAI).29 In the late 1990s, the collapse
of the Organization for Economic Cooperation and Development (OECD) negotiations
for the MAI was in part due to similar public interest concerns, including dissatisfaction
with the dispute resolution system of ISA.30 However, the idea of a MAI has re-emerged
in WTO negotiations and other fora.31
Although we do not follow all the IISD proposals, the public interest elements dis-
tinctive to ISA do demand changes that go beyond streamlining procedures to cut costs
and delays, the primary and more justifiable aim for contemporary ICA. Indeed, several
of the proposals introduced below involve more generous timeframes or safeguards, with
significant cost implications. Nonetheless, at this stage, they seem a fair price to pay to
meet the growing expectations of procedural justice among stakeholders in the still emer-
gent field of ISA. It can then further experiment in refining the balance struck in the per-
ennial interplay between legitimacy and efficiency more familiar now in the world of
ICA.32 And in the longer term, after ISA achieves the more global acceptance built up
over many decades for ICA, another round of reflection and revisions to the ISA system
may refocus instead on delays and cost efficiency. Meanwhile, under the new system pro-
posed below, arbitrators and advocates would bear a responsibility to minimize cost and
delay implications of the new provisions aimed at better reflecting the diverse public
interests in ISA.
To keep testing and tweaking the new system, it is also important to encourage, or
at least allow, some diversity in Rules designed for contemporary ISA. First, therefore,
ICSID should consider another round of more far-reaching reforms to its own Rules.
Secondly, perhaps a more likely scenario, states and private investors should be offered
UR devised primarily for ISA (called, e.g., “UR-ISA Rules”). An alternative set of UR
revisions (“UR-ICA Rules”) can instead emphasize efficiency and private interests over
procedural justice and public interests. However, parties and states could be permitted to
opt out of these and into the UR-ICA Rules even for ISA cases, if they consider the bal-
ance achieved in the latter Rules to be more appropriate than that achieved in the UR-
ISA Rules. Thirdly, arbitral institutions like the ICC, ACICA and the Japan Commercial
Arbitration Association ( JCAA) should develop Rules more specifically tailored to ISA,
while again preserving the parties’ freedom to opt in to these institutions’ more generic
Rules (such as ACICA’s 2005 Rules, themselves based on the UR). However, guidelines
should be widely publicized explaining why the public interests in ISA generally point
towards the use of the more tailored sets of Rules.
29
IISD, supra note 26, at 4–5; Mann et al., supra note 27.
30
S. Brunner & D. Folly, The Way to a Multilateral Investment Agreement (NCCR Trade Regulation Working
Paper No. 2007/24, 2007), available at <http://ssrn.com/abstract=1092836>; K. Miles, Transforming Foreign Invest-
ment: Globalisation, the Environment, and a Climate of Controversy, 7 Macquarie L. J. 81 (2007).
31
See, e.g., <www.globalpolicy.org/socecon/bwi-wto/maimia/index.htm>.
32
W.W. Park, Two Faces of Progress: Fairness and Flexibility in Arbitral Procedure, 23 Mann et al., supra note 27.
Arb. Int’l 499 (2007).
Thus, for example, in the FTA now being negotiated between Australia and Japan,
its investment chapter could allow the claimant investor to proceed “under ICSID Arbi-
tration Rules, UR-ISA Rules, ACICA-ISA Rules or JCAA-ISA Rules.” By allowing
states, investors, and arbitral institutions to differentiate themselves somewhat and to pro-
mote experimentation with new sets of Rules, albeit with some guidance and recom-
mendations for specific sets of ISA Rules, we should generate a better sense of how to
rebalance the system even further over the longer term.
3.1. Why to retain ISA

The more tailored and balanced Rules suggested below should also allay some con-
cerns about using the ISA model at all in investment disputes. For example, ISA was left
out of the investment chapter of the Australia-United States FTA.33 An official reason was
that each “developed country” trusted the others’ courts for investors to initiate claims
against the host state, and fora such as the International Court of Justice for the investor’s
home state to pursue claims against the host state. In fact, around the time Australia inten-
sified negotiations for that FTA, the United States had started to face claims brought by
Canadian investors under the North American Free Trade Agreement (NAFTA).34 Ulti-
mately, the United States successfully defended challenges to a Mississippi jury verdict35
and to rulings in Massachusetts.36 Nonetheless, these claims shocked many in the United
States, who had assumed that ISA was more likely to work in their favour regarding their
investments in Mexico and had not contemplated the possibility that the ISA system
could constrain the powers of U.S. courts as well as government agencies.37 As such, the
2002 Trade Act (19 U.S.C. 3802) called for significant modifications to ISA.
Accordingly, revised provisions for ISA were included in the State Department’s
“model BIT,” further modified to reflect experiences and public comment from the
United States-Uruguay BIT negotiating process.38 ISA has been included in every subse-
quent BIT and FTA concluded by the United States, but with additional safeguards pri-
marily favouring host states.39 These include greater transparency (e.g., public initial
pleadings and hearings, and acceptance of amicus briefs, in the United States-Chile FTA),40
33
United States-Australia Free Trade Agreement, May 18, 2004, 43 I.L.M. 1248 (entered into force January 1,
2005).
34
North American Free Trade Agreement, December 17, 1992, 32 I.L.M. 612 (entered into force January 1,
1994).
35
Loewen, ICSID Case No. ARB (AF)/98/3, Final Award June 26, 2003, tribunal chaired by Sir Anthony
Mason (former Chief Justice of Australia).
36
Mondev, ICSID Case No. ARB (AF)/99/2, Award of October 11, 2002.
37
P. Sands, Lawless World: America and the Making and Breaking of Global Rules (2005).
38
W.W. Park, Procedural Evolution in Business Arbitration: Three Studies in Change, in Arbitration of Interna-
tional Business Disputes: Studies in Law and Practice 1, 33–39 (W.W. Park ed., 2006).
39
C.J. Tams, An Appealing Option? The Debate About an ICSID Appellate Structure, 57 Essays in Transnat’l
Econ. L. 31 (2006).
40
United States-Chile Free Trade Agreement, June 6, 2003, 42 I.L.M. 1026 (entered into force January 1,
2004).
and even in some instances the potential for an appellate review process going beyond the
limited procedural grounds under the ICSID Convention (e.g., Article 28(10) of the
United States-Uruguay BIT).41
Despite this, in the negotiations with the United States for their own FTA con-
cluded on March 3, 2004, the Australian government was happy to go along with the
new U.S. cautiousness towards ISA. One less overt but more pragmatic reason appears to
have been that Australia has traditionally been a net FDI importer vis-à-vis the United
States, although flows (and even total stocks) have become roughly balanced in recent
years.42 Thus, U.S. investors would have been more likely to bring claims against the Aus-
tralian government than vice versa, yet find themselves without the benefits of ISA pro-
cedures. This factor may help explain why it was the Australian government that
reportedly asked for ISA not to be included in the FTA with the United States.43
If so, then for countries Australia might presume to be a net FDI exporter, we would
expect the government instead to push for ISA provisions, even if those countries have
developed and familiar legal systems. Consistent with this hypothesis, we find ISA provi-
sions in the Singapore-Australia FTA (SAFTA),44 the Thailand-Australia FTA (TAFTA),45
and the very recent Australia-Chile FTA.46
In early 2008, by contrast, the word in Tokyo was that the Australian government
was pushing the same official line as with the United States: no need for ISA at all in its
current FTA negotiations with Japan. The Japanese government has not faced domestic
opposition to ISA, but it may be flattered that Australia considers the country to have a
“developed” court and legal system and indeed it does.47 Yet a key motivation for
Australia’s stance seems likely to be that it remains a net FDI importer, despite some
41
Concerning the Encouragement and Reciprocal Protection of Investment United States-Uruguay, November
4, 2005, 44 I.L.M. 268 (entered into force November 1, 2006). Art. 28(10) states: “If a separate multilateral agreement
enters into force between the Parties that establishes an appellate body for purposes of reviewing awards rendered
by tribunals constituted pursuant to international trade or investment arrangements to hear investment disputes, the
Parties shall strive to reach an agreement that would have such appellate body review awards rendered under Article
34 in arbitrations commenced after the multilateral agreement enters into force between the Parties.”
42
Foreign Investment Review Board, Annual Report 2006–2007, 62–63 (Chart 4.1), available at
<www.firb.gov.au/content/publications.asp?NavID=5>.
43
However, overt rationales were that ISA was inadvisable between countries with “advanced legal systems,” as
well as the “discomfort over NAFTA jurisprudence” despite more attempts to provide guidance to arbitral tribunals.
T. Westcott, Foreign Investment Issues in the Australia-United States Free Trade Agreement 80 (2005),
available at <www.treasury.gov.au/documents/958/pdf/06_foreign_investment_policy_ausfta.pdf>.
44
Singapore-Australia Free Trade Agreement, February 17, 2003, 2257 U.N.T.S. 103 (entered into force July
28, 2003).
45
Australia-Thailand Free Trade Agreement, July 5, 2004, 2005 Austl. T.S. (entered into force January 1, 2005).
46
The respective Ministers concluded negotiations on May 27, 2008, and the Agreement was tabled in the
Australian Parliament on June 17. The government’s aim was to sign it at the end of July and bring it into force from
January 1, 2009. See <www.dfat.gov.au/GEO/chile/fta/index.html>. Further, although Australia has not concluded
many BITs in recent years, it concluded a BIT with Mexico in 2005 that also includes ISA provisions: see
<www.austlii.edu.au/au/other/dfat/nia/2006/24.html> (also available at <www.dfat.gov.au/treaties>).
47
J.J. Spigelman, Judicial Exchange Between Australia and Japan, 22 J. Japanese L. 1 (2006). Ironically, Japan’s
impressive judicial system is one reason for the relative paucity of ICA proceedings brought in Japan, as opposed to
arbitrations fought by Japanese corporations outside their country. L. Nottage, Japan’s New Arbitration Law: Domesti-
cation Reinforcing Internationalisation?, 7 Int’l Arb. L. Rev. 54 (2004).
recent high-profile forays by Macquarie Bank and others into infrastructure and tourism
developments in Japan.48
If trust in the other’s courts is truly a major reason for excluding ISA from invest-
ment chapters in bilateral FTAs, then Australia should offer Japan a bilateral enforcement
of judgments regime like that proposed to New Zealand in 2005. That proposes enforce-
ability of judgments from an Australian court in the other country, even if the foreign
defendant refuses to submit to that court’s jurisdiction.49 Yet negotiations have stalled,
despite almost two decades of soft “business law harmonization” between Australia and
New Zealand, and a shared common law tradition. Perhaps there is still not much trust
in each others’ courts after all! On that logic, there should be less resistance to adding ISA
to the investment chapter that the two countries are considering adding to their long-
standing CER Agreement. However, the net FDI importer (New Zealand) may take a
leaf out of Australia’s book and propose excluding ISA.
3.1.1. Short-term Interests in Including ISA Protections

Such pragmatism on the part of government negotiators is understandable. An
investor-state dispute resolution mechanism provides an efficient method of bringing
claims against a host state. Excluding such a mechanism renders the process of recouping
losses from the host state more difficult, and it therefore reduces the likelihood of claims
being brought at all, and this obviously impacts on the public purse. However, the shorter-
term interests of firms themselves may diverge from those of their own government.
One group of host country firms comprises those more likely to become subject to the
impacts of inbound foreign investment, rather than the originators of investment directed
abroad. Their interests are admittedly harder to gauge in the abstract. By contrast, a second
group of firms keen to engage more in outbound FDI have clear interests in their own
government pressing for ISA protections in their favour.
Within the first group, there may be influential firms or entire sectors that are
strongly opposed to FDI inflows, or at least to ISA. Some may be focused on their own
domestic markets, fearful rightly or wrongly of being swamped by foreign competition.
But others may also be globally active firms, such as Australia’s large exporters of natural
resources, that are enjoying such a world-wide boom that they will (and do) attract FDI
even without getting their governments to commit to ISA protections.50 However, those
48
More generally on FDI nowadays into Japan, see C. Pokarier, Open to Being Closed? Foreign Control and Adap-
tive Efficiency in Japanese Corporate Governance, in Corporate Governance in the 21st Century: Japan’s Gradual
Transformation (L. Nottage, L. Wolff, & K. Anderson eds., 2008).
49
See generally <www.beehive.govt.nz/node/23850>. According to a July 31, 2007 joint ministerial statement,
“Ministers re-affirmed their governments’ commitment to conclude a CER Investment Protocol and noted a
New Zealand response to Australia’s offer was pending”; available at <www.trademinister.gov.au/releases/2007/
wtt070731_joint_statement.html>. The statement also mentions agreement in principle on the Trans-Tasman Judg-
ments Enforcement regime, but publically available details remain sparse.
50
A key empirical question is how inelastic the supply of inbound FDI is relative to the level of investment pro-
tection. See T. Ginsburg, International Substitutes for Domestic Institutions: Bilateral Investment Treaties and Governance, 25
Int’l Rev. L. & Econ. 107, 118 (2005). For a useful comparative and historical summary of how various interests
tend to align in the market for FDI generally, and in treaty negotiations, see also T. Pollan, Legal Framework for
the Admission of FDI 1, 3 (2006).
companies or sectors will not always enjoy such a favourable trading environment, and if
it becomes harder for them to attract FDI it may prove very difficult for them to get the
government to renegotiate treaties to add in ISA. In addition, there will be firms or sec-
tors that are already very keen to attract FDI, and would prefer their government to signal
their joint commitment by incorporating ISA protections in investment treaties. This
therefore boils down to an empirical question about net benefits to the host economy as
a whole, but governments rarely seem to attempt a systematic answer before embarking
on investment treaty negotiations regarding ISA.
At a much more aggregate level, anecdotal and quantitative evidence is admittedly
mixed regarding the FDI effects of investment treaties, including the impact of providing
substantive and procedural advantages for foreign investors. The effects are even less cer-
tain regarding ISA itself.51 Japan attracted considerable FDI without including ISA in its
“Treaties of Friendship and Navigation.” One example is the 1976 Australia-Japan Basic
Treaty of Friendship and Cooperation.52 Article IX(3) guarantees substantive protections
such as “fair and equitable treatment” and non-discrimination regarding the business
activities (including investments) of each other’s nationals. Japanese corporations also
invested heavily world-wide without many BITs having been concluded by their state,
although the dozen or so since 1977 all include ISA, and those with Korea and Singapore
extended “fair and equitable treatment” to the pre-establishment phase of the invest-
ment.53 Japanese investors have not invoked these ISA options much yet, but have been
involved directly or indirectly in several well-known international investment arbitra-
tions.54 The Japanese government now promotes ISA as a key component of its FTA
strategy, although it does not always succeed in including arbitration provisions.55
Franck also points out that China, Brazil, and Ireland have enjoyed even more
dramatic inflows of FDI without offering ISA protections. Brazil and Ireland have not
51
See, e.g., J. Yackee, Do BITs Really Work? (Univ. of Wisconsin Legal Studies Research Paper No. 1054, 2007)
available at <http://ssrn.com/abstract=1015083 (2007)>.
52
Also known as the “Nara Treaty”: see <www.dfat.gov.au/GEO/japan/japan_treaty.html>.
53
S. Yanase, Bilateral Investment Treaties of Japan and Resolution of Disputes with Respect to Investment, in International
Commercial Arbitration: Important Contemporary Questions 426, 429–30 (A.Van Den Berg ed., 2003).
54
For example, the Dutch subsidiary (Saluka Investments BV) of a major Japanese securities firm (Nomura)
pursued a claim against the Czech Republic. See Savage, supra note 28, at 12. In Indonesia, a major Japanese general
trading company (Tomen Corp.) was a 9% partner in the Karaha Bodas power project consortium, led by two U.S.
companies, which was awarded USD 261 million in a UR arbitration by a tribunal chaired by Yves Derains. A major
Japanese bank (Sakura) helped finance the Dieng and Patuha power projects, where a consortium led by CalEnergy
(in the United States) obtained USD 570 million out of USD 3 billion against Indonesia’s utility company (PLN) in
a first UR arbitration. That tribunal was chaired by Jan Paulsson, with the Indonesians appointing Prof. Priyatna
Abdurrasyid, and CalEnergy appointing A.A. de Fina, an Australian: see also ICT Pty. Ltd. v. Sea Containers Ltd.,
(2002) N.S.W.S.C. 77. This consortium then initiated a second arbitration against the Indonesian government itself,
alleging sovereign performance guarantees. See L.T. Wells & R. Ahmad, Making Foreign Investment Safe: Prop-
erty Rights and National Sovereignty 13, 14 (2007).
55
See, e.g., Toshi Kyotei no Susumekata [Ways to Promote Investment Treaties], 47-page report for the
fifthTrade Policy Division meeting of the Industrial Structure Deliberative Council (Sangyo Kozo Shingikai), April 16,
2008; and Investment, in Japanese government’s Report on Compliance by Major Trading Partners with Trade
Agreements 5 – WTO, FTA/EPA, BIT (2007), available at <www.meti.go.jp/english/report/index_report.html>.
Savage, supra note 28, at 6, 13 (pointing out that Japan went along with objections from the Philippines, which had
recently lost in investment arbitrations, and omitted ISA in their FTA signed in 2005). By contrast, ISA was included
in the Japan-Malaysia FTA signed that year, although recently Chile initiated investment arbitration against Malaysia
under the BIT between those two countries. However, the METI Report (id. at 630) notes that the Japan-Malaysia
FTA excludes National Treatment and Performance Requirement obligations from its ISA provisions.
grown as strongly as China, but have still liberalized their economies and consequently
have attracted a great deal of FDI, again without relying on BITs. However, it has been
suggested that there may be indirect benefits to host state economies through ISA
enhancing local court systems and more broadly the rule of law.56 Other empirical work
counters this link, especially for developing countries, suggesting that ISA substitutes for
rather than complements transparency in local courts and other institutions.57 The link
proposed by Franck and others also requires a further controversial assumption about a
particular vision of the rule of law having a causal effect on economic growth.58
Accordingly, there certainly remains ongoing empirical and theoretical debate about
the net benefits of ISA for host states, including different sectors and organizations that
are more likely to be subjected to the effects of FDI inflows. However, it is worth noting
another potentially significant divergence between governmental and business interests
regarding ISA. Even where it is made available, if foreign investors bring a successful
claim, it is the host state that is directly liable, not the local investment targets. Indeed,
having obtained relief under ISA, a successful claim against the host state may obviate the
investors’ need to claim under any separate or related contracts with investment target
firms. Nonetheless, the host state’s compensation payments to foreign investors may allow
them to keep operating, maintaining competitive pressure on the local firms.
By contrast, firms primarily engaged in investment abroad (e.g., Macquarie Bank)
have more straightforward interests in encouraging their governments to include ISA
protections in investment treaties. Other investment target firms (e.g., those involved in
large infrastructure developments) may also be actual or potential investors in the country
with which Australia is negotiating an investment treaty.
At a more aggregate level, moreover, some such firms (e.g., in Australia) may be
tempted to route their investment (e.g., destined for Japan) through a subsidiary created
in a third country (e.g., Chile) with which it has negotiated a FTA liberalizing investment
flows (and perhaps indeed itself allowing ISA, as in the new Australia-Chile FTA). The
subsidiary could then try to invoke any ISA provisions in a treaty between that third
country and the ultimate destination country (as under Chapter 8 of the Chile-Japan
FTA, signed on March 27, 2007).59 Yet such forum-shopping would involve “investment
56
S. Franck, Foreign Direct Investment, Investment Treaty Arbitration, and the Rule of Law, 19 McGeorge Global Bus.
& Dev. L. J. 337 (2007). However, now that Brazil is a net FDI exporter particularly to other South American countries,
it is apparently including ISA in some of its BITs.The New Zealand-China FTA (April 7, 2008, expected to come into
force on October 1, 2008, available at <www.chinafta.govt.nz/>) contains extensive ISA provisions in Ch. 11 s. 2.
57
Ginsburg, supra note 50.
58
See generally F. Upham, Speculations on Legal Informality: On Winn’s “Relational Practices and the Marginalization
of Law”, 28 L. & Soc’y Rev. 233 (1994).
59
Australian investors via Chile may not succeed in invoking ISA because art. 86(2) of the Chile-Japan Economic
Partnership Agreement, available at <www.mofa.go.jp/policy/economy/fta/chile.html>, gives Japan the discretion, after
consulting with Chile, to “deny the benefits” of its investment Chapter (both substantive and procedural) if the vehicle
controlled by a non-party (like Australia) has no “substantial business activities” in Chile. (More generally, see G. Biggs,
Investor State Settlement of Disputes Under the Chile Japan Agreement, 19 J.C.A.A. Newsletter 1 (2007), available at
<www.jcaa.or.jp>. The same applies in the reverse situation of a Japanese investor routing an investment in Australia
through Chile, due to ACFTA, art. 10.13 (similar also to SAFTA, Ch. 7 art. 18 and TAFTA, art. 911). Tax or other prac-
tical considerations may also make such routing difficult. However, careful structuring of the investment might allow cir-
cumvention of such obstacles, and the host state may not invoke its discretion in a particular case. As countries enter into
more and more FTAs and BITs, moreover, the possibilities for forum-shopping expand considerably.
diversion” that is probably inefficient for the economies involved, and which certainly
increases transaction costs, especially for investor firms.60
3.1.2. Long-term Interests in Including ISA Protections

In addition to these more immediate interests of governments and various sectors or
firms, there are longer-term interests for all to consider when negotiating ISA processes,
even when their country is a net FDI importer. First, if developed countries begin to
exclude ISA as amongst themselves, there is a chance that ISA will become more biased
against developing countries or, at least, be perceived to be so. Thus, involving developed
and developing countries in applying and refining the ISA system is important for the
legitimacy of both its procedures and the substantive law principles it generates. That has
also been important for the now more “judicialized” WTO system.
Secondly, just as the WTO system has some direct as well as more diffuse overlaps
with ICA,61 ICA has even more overlaps with ISA. Leaders in ICA already, and perhaps
increasingly, serve as arbitrators or advocates in ISA proceedings, for example. The UR
revision process shows how Rules and practices in each field can influence for better or
worse those in the other. Such overlaps are particularly important for countries like Aus-
tralia and Japan, which are seeking to attract more ICA to their shores, rather than con-
tinuing to see cases go to “core” jurisdictions in Europe and the United States, or
nowadays to regional leaders such as Singapore.
Thirdly, the professed longer-term goal of countries like Australia and Japan is to
begin to link up the “spaghetti bowl” of bilateral FTAs. Japan, in particular, is the fulcrum
in many “ASEAN+” proposals or detailed negotiations already underway.62 It will be
easier to ensure ISA is included in such regional agreements involving both developing
countries in the Association of South-East Asian Nations and developed countries like
Japan or Australia, if the developed countries have already included ISA in their own
60
Forum-shopping, e.g., via an intermediate investment vehicle in Chile, may also be unnecessary if it is pos-
sible to invoke any most-favoured-nation (MFN) provision that could be included in the Australia-Japan FTA, even
if that FTA ends up excluding ISA, since Japan has consented to ISA in FTAs with third countries such as Chile. A
similar possibility arises regarding the Australia-PRC BIT and Australia-United States FTA, which lack ISA provi-
sions but include MFN clauses. However, arbitral tribunals have reached conflicting decisions on whether MFN
clauses, even if broadly worded, extend in part or in full to procedural rights like ISA, as opposed to substantive pro-
tections for investors. See P.J. Turner et al., Investment Treaty Arbitration: An Australian Perspective, 24 J. Int’l Arb. 103,
116–19 (2007). A recent analysis opposes extending MFN in this manner: A. Faya Rodriguez, The Most-Favored-
Nation Clause in International Investment Agreements: A Tool for Treaty Shopping?, 25 J. Int’l Arb. 89 (2008).
61
See, e.g., J. Bohannes & H. Nottage, Arbitration as an Alternative to Litigation in the WTO: Observations in the
Light of the 2005 Banana-Tariff Arbitrations, in The WTO in the Twenty-first Century: Dispute Settlement,
Negotiations, and Regionalism in Asia (Y. Taniguchi & J. Bohannes eds., 2007); B. Malkawi, Arbitration and the
World Trade Organization: The Forgotten Provisions of Article 25 of the Dispute Settlement Understanding, 24 J. Int’l Arb.
173 (2007). A more indirect example is the “cross-over” of one of the most prominent Asia-Pacific ICA experts, Prof.
Yasuhei Taniguchi, to serve on the WTO Appellate Body (2000–2007).
62
See, e.g., ANU/ERIA/JETRO conference Roadmap Towards Economic Integration in East Asia, Sydney,
February 25, 2008, available at <www.jetro.go.jp/australia/events/upcoming/eriaseminar.html>. See also the Austral-
ian prime minister’s proposal, for his recent visit to Japan, to engage both countries in a more integrated Asia-Pacific
Community drawing some inspiration from the European Union: L. Nottage, PM Takes Our Japan Trade to New Lev-
els, Australian Fin. Rev., June 2008, at 73.
bilateral treaties. Of course, developed countries that have excluded ISA in their bilateral
arrangements may attempt to exert diplomatic and economic pressure to have ISA
included in the regional agreements. However, this risks generating considerable resent-
ment towards ISA and may ultimately undercut the original exclusion of ISA from the
bilateral agreements.
This situation therefore generates a problem familiar to economists and political sci-
entists. Short-term considerations, which seem to favour the ad hoc exclusion of ISA, are
impeding the likely potential net benefits of a broader bilateral and ultimately multilateral
network of ISA. However, that potential will only come about, or be maximized in
terms of social welfare rather than dollars or yen, if the ISA system evolves to meet the
problems it encounters in practice, in particular to address the public interests involved. If
the system can be reformed, along the lines proposed below, it should create a system that
can balance legitimacy and efficiency.
3.2. How to retain ISA

Even accepting such arguments for including revised ISA provisions into interna-
tional agreements, there remains the question of how best to do so. One possibility is for
countries to negotiate such rules on a case-by-case basis in their BITs or bilateral FTAs.
A problem with this approach is the risk of ad hoc inconsistencies. For example, SAFTA
2003 allows investors to bring ISA cases under the UR or ICSID Rules; TAFTA 2005
only allows the UR; ACFTA (Article 10.16(3)) allows the UR, ICSID Rules, ICSID
Additional Facility Rules (perhaps acknowledging that Chile or even Australia might
withdraw from ICSID) or “any other arbitration rules” the countries may agree to; but
the Australia-United States FTA 2004 does not provide for ISA at all. Australia’s FTAs also
have quite significant differences regarding timeframes for ISA proceedings and interim
measures (as outlined below). The rationale for such differences has not been publically
articulated.
Inconsistencies even among one country’s bilateral FTAs will multiply if that coun-
try has less negotiating power, as with Australia compared, for example, to the United
States. The unique economic and diplomatic clout has allowed the United States to apply
successfully a “model BIT.” Many other countries are developing model BITs, including
Germany and the Netherlands, but they tend to be net FDI exporters. Even if some add
provisions on transparency and third-party transparency that may appeal to FDI importer
countries, as in the 2007 Norway model BIT, the wording and impact may be disparate,
especially when smaller economies are involved.63 A related problem for countries like
Australia comes from constraints on the budget and personnel available for negotiating
its proliferating international agreements. There was apparently little overlap in the staff
63
See, e.g., <http://ita.law.uvic.ca/investmenttreaties.htm>; R. Dolzer & C. Schreuer, Principles of Inter-
national Investment Law 353–419 (2008), (PRC 2003, United States 2004, France 2006, Germany 2005, United
Kingdom 2005).
assigned to negotiate its recent bilateral agreements, and key personnel negotiating the
FTA with Japan are different again.64
A further argument against leaving the development of ISA rules to ad hoc incor-
porations into bilateral agreements is the problem posed by regional agreements now
under consideration, especially in the Asia-Pacific region. These need a common starting
point, yet this becomes more and more ephemeral as countries like Australia negotiate
bilateral FTAs and BITs in a disparate manner.
A second possibility is therefore the elaboration of a multilateral framework for ISA.
One option is a protocol or revision to the ICSID Convention itself. However, as with
the 1958 New York Convention, there is always reluctance to alter formally a treaty that
already has so many accessions. There is a risk of confusion about whether the original or
revised treaty regime applies, even if enough parties sign up and then accede to the
revised treaty. It also seems too early for many countries to agree yet on the substantial
ISA rule changes proposed below. All the more so, regarding another option: a multilat-
eral investment treaty covering substantive principles, as well as dispute resolution proc-
esses. The controversy over the OECD’s MAI, and now regarding adding new fields to
the WTO regime, is too fresh in everyone’s mind.
This leaves the third possibility: devising new sets of ISA Rules to be incorporated
as options for foreign investors into bilateral and future regional international FTAs and
BITs. In particular, national arbitration centres like ACICA and JCAA can quite easily
take the lead in developing tailored ISA Rules. Those, in turn, can inform debates under-
way for revisions to the UR and ultimately again the ICSID Rules, which have struggled
to achieve major reforms due to the many states involved in those revision processes.
Investors will hopefully choose “ACICA-ISA” or “JCAA-ISA” Rules, recognizing the
chances of less backlash in the host state to any rulings in their favour by tribunals oper-
ating under fairer Rules, as well as the longer-term benefits of following such Rules.
After all, the regimes under both the ICSID Convention (Article 55) and the NYC still
allow losing states to resist execution of arbitral awards by pleading sovereign immunity.
There remains considerable scope to do so, even when the investor seeks execution
against assets in third countries allegedly owned by the losing states, which would be gov-
erned by those third countries’ legislation on sovereign immunity.65 However, the incen-
tives for host states to keep fighting awards at the execution phase should diminish if those
awards have resulted from proceedings governed by recently customized Rules seen as
64
DFAT has had primary carriage of FTA negotiations, but Treasury also gets involved, see, e.g., in the
Australia-United States FTA context: Westcott, supra note 43. The Attorney-General’s Department, Australia’s Justice
Ministry, also has international law experts but the extent of their involvement is unclear.
65
In the United States, for instance, the Foreign Sovereign Immunities Act (28 U.S.C. ss. 1602–11) allows eas-
ier execution of awards compared to foreign judgments (where the sovereign debtor has to provide a separate waiver
of sovereign immunity). However, the creditor still has to establish that the assets are owned by the losing state and
are being used in the United States for a “commercial activity” (s. 1605). See K.H. Cross, Arbitration as a Means of
Resolving Sovereign Debt Disputes, 17 Am. Rev. Int’l Arb. (2007), available at <http://ssrn.com/abstract=1014833>.
Such complications make it even more difficult for successful investors in multiple ICSID arbitrations to obtain
execution of their awards against Argentina, which apparently has not yet paid out under any. It may also explain why
Occidental Petroleum apparently settled its long-standing dispute against Ecuador even after its ICSID award was
upheld by courts in the United Kingdom (as outlined by Triggs, supra note 25).
fairer for the states involved. In turn, positive experiences applying those Rules should
encourage similar amendments to the UR and ICSID Rules. Such a bottom-up, volun-
tarist “emerging consensus” can then help build up a broader and firmer basis for elabo-
rating and imposing more standardized provisions through bilateral, regional, and
multilateral agreements among states.
For this to work best, however, we may need to recreate some specific advantages of
arbitrations conducted under the ICSID Rules largely “delocalized” from national court
involvement.66 In particular, ICSID Rules arbitrations involve a self-contained mecha-
nism for challenging the award. Section V of the ICSID Convention allows a party to
seek interpretation or revision (based on certain later-discovered facts), usually by the
same arbitral tribunal (Articles 50 –1), or annulment (Article 52) by a new tribunal but
primarily on limited procedural grounds. Article 53 clarifies that these are the only ave-
nues for appeal. Article 54(1) requires the host state to recognize and enforce pecuniary
obligations imposed by an ICSID award “as if it were a final judgment” of its local courts.
ICSID Arbitration Rules 50 to 55 elaborate on these aspects.
By contrast, investment arbitration awards that are not rendered under the ICSID
Rules do not benefit from the ICSID Convention’s self-contained regime.67 This applies
even to arbitrations administered by ICSID under its Additional Facility Rules, available,
for example, for investment disputes where one party is not (from) a Convention state,
e.g., involving Mexico. It also applies to investment arbitrations resolved under other
institutional Rules, or ad hoc under the UR. This opens up the possibility of challenge
in courts at the agreed “seat” of the arbitration, e.g., in Canada when U.S. investors claim
against Mexico in NAFTA arbitrations, pursuant to national arbitration legislation.68 The
award can also be challenged in courts of the enforcing state, such as Mexico. However,
most states have now acceded to the NYC and over the decades their courts have gener-
ally learned and applied its principles, including limited grounds for refusing enforce-
ment. As the cornerstone of their legislation applicable to arbitrations with their seat in
that state, many have also adopted the 1985 UNCITRAL Model Law on International
Commercial Arbitration (ML), which replicates for setting aside applications the limited
grounds found in NYC, Article V. These tendencies help explain why many ISAs are
conducted outside the ICSID Rule framework, but the majority of investors still apparently
choose ICSID Rules.
If we wish tailored ISA Rules to compete on a level playing field with ICSID Rules
in this respect, therefore, it is important first to build into those ISA Rules themselves
self-contained mechanisms for review by arbitrators and direct enforceability. The Rules
should also clarify that parties agree to waive rights to apply for courts to set aside at the
seat or to refuse enforcement. Because arbitration law (even based on the ML) may not
66
L. Reed et al., Guide to ICSID Arbitration 8 (2004).
67
See Dolzer & Schreuer, supra note 63, at 277–90.
68
See, e.g., United Mexican States v. Metalclad Corp. (2001) B.C.S.C. 664. The Supreme Court of British
Columbia partially upheld the NAFTA award under the International Commercial Arbitration Act 1996, s. 34 but
also annulled key parts of the award (adopting the ML, like Australia’s International Arbitration Act, and Japan’s Arbi-
tration Act 2003).
clearly allow such waivers, it may also be prudent to build such mechanisms into the rel-
evant treaties (and corresponding domestic legislation). Foreign investors and states would
also have to agree to arbitrate under the specified ISA Rules with the seat (and hearings)
in one or other of the treaty states. Otherwise, a disgruntled or obstructive party may
seek to have the award set aside by courts of third states specified as the seat, which states
cannot be forced to uphold waivers of such rights. However, even without such compli-
cations added to bilateral treaties and legislation, hopefully we can build other attractions
into the new ISA Rules to encourage investors to choose them over the venerable
UNCITRAL Rules, and even the ICSID Rules.
4. Specific ISA Rule Revision Proposals

This Part addresses the key areas for reform in ISA Rules. Disclosure, third-party
access and appellate review are the ISA issues that have most exercised IISD and other
commentators on the contemporary ISA system. However, a number of other procedural
matters also need consideration, such as arbitrator appointment, time limits, and voting
procedures. The final column in this article’s Appendix summarizes our tentative propos-
als for further reforms to ICSID Rules, and revisions to the UR or Rules designed pri-
marily for ICA by arbitral institutions such as ACICA and JCAA.
4.1. Disclosures, Third-party Participation, and Appellate Review

4.1.1. Fact of the Arbitration
Most importantly, the public interest elements in ISA point to the need for disclo-
sures of various kinds. The first relates to the fact of arbitration.69 Although the UR and
ICSID make provision for the disclosure of the fact that a notice of intention to submit
a claim for arbitration has been filed, under the current JCAA Rule 40(2) and ACICA
Rule 18.2, the fact of arbitration is to remain confidential. In our view, this requires dis-
closure. Possibly, even the full notice of arbitration should be made available on the offi-
cial website of the arbitral institution.
4.1.2. Access to Documents, Access to Hearings, and Amicus Briefs

Currently, confidentiality of documents is preserved under JCAA Rule 40(2) and
ACICA Rule 18.2. Under ICSID Rule 6.2, disclosure is permissible by the parties. At
this stage, our proposal allows disclosure of the pleadings with consent of the parties, but
restricts access to the evidentiary documents presented during the hearings to the parties
involved in the dispute. While there are good reasons to consider disclosing the pleadings
69
Sornarajah, supra note 27; Luke E. Peterson, All Roads Lead out of Rome: Divergent Paths of Dispute Settlement
in Bilateral Investment Treaties, in International Investment for Sustainable Development: Balancing Rights and
Rewards (L. Zarsky ed., 2005).
in the matter, full access to all the evidence would become unwieldy. We also suggest that
arbitral institutions should give further consideration to allowing disclosure of the initial
pleadings even without the consent of the parties.
Non-party access to the hearings, as well as acceptance of amicus briefs, remain par-
ticularly controversial issues. Currently, the various arbitration regimes do not accommo-
date the need for non-party access that is generated by the public interests involved in
ISA. UR Rule 25.4, ACICA Rule 18.1, and JCAA Rule 40(1) all maintain the privacy
of the proceedings. They also preclude any opportunity for the receipt of amicus curiae
briefs. ICSID Rule 32.2 only allows non-party access if the parties agree; if one party
objects, the non-party is excluded from the oral hearings. ICSID’s new Rule 37.2, how-
ever, is a step in the right direction. It allows the arbitrators to receive amicus briefs even
without the consent of the parties. The arbitrators are still required to consult with the
parties, but they can override their wishes. Procedural safeguards have been put in place
to limit non-party submissions, such as the need to establish a sufficient interest in the
subject matter of the dispute.70 Rule 37 further prescribes additional matters that the tri-
bunal will consider in determining whether to accept a non-party submission. One is
that such non-parties should be able to assist the tribunal by bringing a perspective, par-
ticular knowledge, or insight that is different from that of the disputing parties. Another
is that their submission would address a matter within the scope of the dispute.
In our view, the arbitrators should have the capacity to permit both the acceptance
of amicus briefs and non-party access to the hearings, despite objections from the parties.
However, much in line with the new ICSID Rule 37.2, we also consider that consulta-
tion with the parties should be mandatory before making their decision and that appro-
priate safeguards would need to be put in place. These include measures to protect
confidential commercial information and a screening process for organizations seeking to
submit amicus briefs.71
4.1.3. Publication of the Award

Basic public law transparency principles also point to the need for publication of
arbitral awards in ISA. The potentially far-reaching implications of the awards for host
states and, in particular, the pressure on public funds to meet any damages awarded neces-
sitate the compulsory publication of arbitral awards in ISA.72 This suggestion is very dif-
ferent from the current norm in international arbitration. UR Rule 32.5 maintains the
confidentiality of the award, as do ACICA Rule 18.2 and JCAA Rule 40(2). Under Rule
48.4, ICSID is only permitted to publish the award with the consent of the parties. It is,
however, required to publish excerpts of the legal reasoning from an award. We do not go
so far as to argue for disclosure of settlements (discussed further below), but awards ought
to be publically available regardless of the views of the parties.
70
Tienhaara, supra note 27.
71
IISD 2007, supra note 26.
72
Van Harten, supra note 27.
4.1.4. Appellate Body

In addition to these measures, establishing a full-scale appellate body for ISA needs
to be given further serious consideration.73 An appellate facility in the mould of the
WTO Appellate Body would review questions of law and legal interpretation, not just
procedural irregularities, to help remedy problems of inconsistency amongst awards.74 It
would also help to alleviate illegitimacy allegations regarding ISA. This issue has again
been highlighted by the recent decision of the annulment committee in CMS Gas Trans-
mission Co. v. Argentine Republic, in which errors were found in the original award but the
limited annulment conditions were not satisfied by the occurrence of errors of law.75
Concerns have been expressed that permitting appeals would remove the finality of the
award, prolong the dispute, increase the period of uncertainty for the parties, and increase
costs.76 Others dismiss such concerns, arguing that tight procedural rules and timeframes
can help to address these objections.77
One possible compromise might be an obligation to refer controversial issues of sub-
stantive law to an appellate body, which would then provide advice to the tribunal on inter-
pretation and application of the law. The inspiration would be the European “preliminary
reference” procedure, whereby national courts seek (and later themselves apply) authoritative
interpretations of EU law from the European Court of Justice (EC Treaty, Article 234).78
73
See generally K.P. Sauvant, Appeals Mechanism in International Investment Disputes (2008). Since 2003,
over a dozen treaties concluded by the United States have provided for further discussion about adding an appeals
mechanism, supra note 41. Apparently, however, no formal negotiations have been initiated: B. Legum, Options to
Establish an Appellate Mechanism for Investment Disputes, in Appeals Mechanism in International Investment
Disputes 231 (K. P. Sauvant ed., 2008).
74
See, e.g., the Lauder Cases: In the Matter of a UNCITRAL Arbitration between Ronald S. Lauder and the
Czech Republic, Final Award, September 3, 2001; CME Czech Republic B.V. (The Netherlands) v. Czech Repub-
lic, Partial Award, September 13, 2001, Final Award, March 14, 2003. However, some commentators suggest that
examples of true conflicts in ISA are rare, and appellate review may therefore be overkill: J. Paulsson, Avoiding Unin-
tended Consequences, in Appeals Mechanism in International Investment Disputes 241 (K.P. Sauvant ed., 2008).
75
C.H. Brower, Structure, Legitimacy, and NAFTA’s Investment Chapter, 36 Vand. J. Transnat’l L. 37 (2003);
Mann et al., supra note 27; Van Harten, supra note 27; Miles, supra note 27; CMS Gas Transmission Co. v. Argen-
tine Republic, ICSID Case No. ARB/01/8, Annulment Proceeding, Decision of the Ad Hoc Committee on the
Application for Annulment of the Argentine Republic, September 25, 2007; the Tribunal states at para. 158:
“Throughout its consideration of the Award, the Committee has identified a series of errors and defects. The Award
contained manifest errors of law. It suffered from lacunae and elisions. All this has been identified and underlined by
the Committee. However the Committee is conscious that it exercises its jurisdiction under a narrow and limited
mandate conferred by Article 52 of the ICSID Convention.The scope of this mandate allows annulment as an option only
when certain specific conditions exist. As stated already (paragraph 136 above), in these circumstances the Committee
cannot simply substitute its own view of the law and its own appreciation of the facts for those of the Tribunal.”
76
Summarized by Tams, supra note 39.
77
M.R. Sammartano, The Fall of a Taboo: Review of the Merits of an Award by an Appellate Arbitration Panel and a
Proposal for an International Appellate Court, 20 J. Int’l Arb. 387 (2003).
78
Kaufmann-Kohler, supra note 21, at 378. This proposal is further developed now by C. Schreuer, Preliminary
Rulings in Investment Arbitration, in Appeals Mechanism in International Investment Disputes 207 (K.P. Sauvant
ed., 2008). Another proposal that emerged from the OECD in 2006 was a more superficial review of awards similar
to that conducted by the ICC “Court” of Arbitration: K.Yannaca-Small, Improving the System of Investor-State Dispute
Settlement, in Appeals Mechanism in International Investment Disputes 223, 226 (K.P. Sauvant ed., 2008). An
advantage of developing sets of ISA Rules, as we propose here, is that it allows experimentation by both institutions
and parties themselves with different models of appellate review. Most of the debate continues to revolve around
appeals within ICSID, which decided not to proceed partly because diverse variants were proposed. However, some
are now at least recognizing the need to reconsider the issue for other ISA Rules: A. Qurishi & S.G. Khan, Implica-
tions of an Appellate Body for Investment Disputes from a Developing Country Point of View, in Appeals Mechanism in
International Investment Disputes 267 (K.P. Sauvant ed., 2008).
Costs and delay issues are also tied into the method of appointment and remunera-
tion of arbitrators, and they have implications for public interest issues as well (as discussed
further below).79 As the costs of defending claims in ISA continue to increase, concerns
have been expressed that they are reaching prohibitive levels for developing states, and
that this will have a chilling effect on the enactment of public welfare regulation. That is,
developing states will not even risk a challenge from investors and will instead refrain
from enacting new legislation.80 Particularly with a permanent appellate facility, the
remuneration system for arbitrators could be shifted from an hourly rate to tenured judi-
cial appointments. This would remove any temptation to prolong proceedings, as well as
safeguarding the independence of the decision-makers in ISA.81 Indeed, the WTO expe-
rience illustrates how an appellate body system can lead to proportionately fewer cases
being appealed.82 Arguably, this is related to increasingly predictable substantive law out-
comes. As such, over time, an appellate system may, in fact, generate significant cost and
time savings, as more and more disputes are settled in the clearer shadow of the law.
4.1.5. Advisory Centre

Furthermore, a related but broader improvement would be the creation of an Advi-
sory Centre for Developing States on Investment Treaty Arbitration.83 Concerns have
been raised about the lack of financial resources and technical capacity in developing
states facing substantial investor claims, and the consequential inability to mount an ade-
quate defence.84 This obviously raises fairness issues and these will continue to impact on
the perceived illegitimacy of international arbitration as a dispute resolution model for
investor-state disputes unless ISA is modified to take better account of the needs of devel-
oping states. To address this imbalance, Gottwald proposes the establishment of a Legal
Assistance Center for Developing Nations in Investment Treaty Arbitration.85 A model for
such an entity already exists in the form of the WTO Advisory Centre on WTO Law.86
The Centre is funded through contributions from developing and developed countries,
but it provides legal advice on WTO law, support during dispute settlement proceedings,
and training of government officials in WTO law to developing states only.87 An Advisory
79
See generally Van Harten, supra note 27.
80
See, e.g., L. Zarsky, Havens, Halos and Spaghetti: Untangling the Evidence About Foreign Direct Investment and the
Environment, in Foreign Direct Investment and the Environment 47 (1999); N. Bernasconi-Osterwalder & E.B.
Weiss, International Investment Rules and Water: Learning from the NAFTA Experience, in Fresh Water and Interna-
tional Economic Law (E.B. Weiss, L. Boisson de Chazournes, & N. Bernasconi-Osterwalder eds., 2005).
81
See, e.g.,Van Harten, supra note 27; Miles, supra note 27.
82
W. Davey, Reforming WTO Dispute Settlement, (University of Illinois, Public Law & Legal Theory Research
Paper, (No. 04 –01, 2004) available at <http://ssrn.com/abstract=495386>.
83
E. Gottwald, Leveling the Playing Field: Is It Time for a Legal Assistance Centre for Developing Nations in Investment
Treaty Arbitration, 22 Am. U. Int’l L. J. 237 (2007).
84
See e.g., C. Garcia, All the Other Dirty Little Secrets: Investment Treaties, Latin America, and the Necessary Evil of
Investor-State Arbitrations, 16 Flo. J. Int’l L. 301 (2004).
85
Gottwald, supra note 83.
86
See <www.acwl.ch/e/index_e.aspx>.
87
Gottwald, supra note 83.
Centre of this kind for ISA would be a welcome addition to the international arbitration
community.
4.1.6. Australia-Chile FTA

Interestingly, unlike the agreements recently entered into by the United States, the
recent ACFTA does not even envisage the possibility of any more extensive appellate
review mechanism. This is despite two unusual provisions. Article 10.20(3) requires the
tribunal to “decide as a preliminary issue any objection by the respondent that the claim
is manifestly without legal merit.” This is without prejudice to its authority to address
objections that the tribunal lacks jurisdiction, which the tribunal must decide on an
expedited basis (Article 10.20(4)). Both provisions seem to favour the host state, which
may be keen to knock out weak claims by foreign investors at an early stage. Article
10.20(3), in particular, misses an opportunity to set up an EU-style “preliminary refer-
ence” procedure for rulings by arbitrators selected from a permanent body of experi-
enced arbitrators from Australia, Chile, and agreed third countries. This could have been
the first step to developing a permanent appellate body reviewing ad hoc arbitrators’ rul-
ings on at least some matters of substantive law.
Instead, the ACFTA adds provisions on public access, unique for its FTAs and BITs
so far. The tribunal can accept amicus briefs (Article 10.20(2)), but these must be in both
Spanish and English, which seems burdensome for (what the Agreement calls) “the sub-
mitter” and therefore something better left to the arbitrators’ discretion. The tribunal
must give the parties the opportunity to respond to briefs that it lets in, but there is no
express requirement allowing parties (especially investors) to present prior arguments
about whether the submitter or its material are likely to be useful to the tribunal in the
specific dispute. Article 10.21(1) then provides for disclosure of certain documents to “the
non-disputing Party.” This is defined in Article 10.1(m) as the “Party not a party to an
investment dispute,” who is then empowered to make submissions to the tribunal “on any
question of interpretation of this Agreement,” upon written notice to the disputing par-
ties (Article 10.21(2)). These documents include not only the notice of arbitration, but
also pleadings, minutes, or transcripts of hearings “where available,” the tribunal’s rulings
and “any other document submitted to the tribunal, including redacted versions of con-
fidential documents submitted in accordance with Article 10.22.”
Except for the last-mentioned “other documents,” these must also be made available
“to the public at their cost” (Article 10.22(1)). The tribunal must also conduct hearings
in public, unless any disputing party advises that it intends to “use information designated
as confidential business information or information that is privileged or otherwise pro-
tected from disclosure under a Party’s law in a hearing.” Then “the tribunal shall make
appropriate arrangements to protect the information from disclosure, including closing”
the (related) hearing (Article 10.22(2)). The tribunal must also protect disclosures of such
information, including presumably any provided by submitters of amicus briefs, under
procedures detailed in Article 10.22(5). However, the respondent (host state) may not rely
on this “to withhold from the public information required to be disclosed by its laws”
(Article 10.22(7)); on the other hand, it need not “disclose information which would
impede law enforcement” or the like (Article 10.22(3)). Oddly, there is no provision cor-
responding to the former (Article 10.22(7)) that applies to the claimant (foreign investor).
These provisions are also noticeable for imposing strict requirements on the arbitrators,
rather than leaving at least some matters to their discretion.
This brief analysis of the new ACFTA shows the complications that can arise when
provisions are drafted into treaties, or left out, on an ad hoc basis. Because the provisions
encompass relatively new ideas and even bilateral agreements are relatively difficult to
update in the light of developments in theory and practice, it reinforces our argument
(above) that tailored Rules may be a better option for finding the right balance.
4.1.7. Summary
The establishment of a central, permanent appellate body for ISA, together with the
appropriate rules on transparency and receipt of amicus briefs, would address many of the
primary concerns that have been expressed about the current dispute settlement frame-
work for ISA. Yet the notion of in-built appellate review is almost unheard of in Rules
designed for ICA; it is even lacking in the UR, which derived partly from an earlier era
involving arbitration of investment disputes. In the field of ISA, so far, a separate ICSID
arbitral tribunal only reviews awards for serious procedural irregularities (ICSID Conven-
tion, Article 52). The proposed appellate body therefore requires major Rule amend-
ments, even for ICSID, as do our proposals regarding public access and disclosure.
By contrast, relatively minor changes would be required to ICSID Rules regarding
the following topics. These have drawn less debate, but they are also important, particu-
larly for those familiar with or inspired by recent developments in ICA.We focus on how
the generic Rules of institutions like ACICA and JCAA require quite extensive tailoring
for ISA, if they want to run such cases effectively and thereby help restore the legitimacy
of the system. Occasionally, however, their Rules instead provide some useful scope for
improvements to the ICSID or UR systems.
4.2. Arbitrator appointment and remuneration

One set of issues revolves around the default number of arbitrators and how they are
appointed. The UR Rule 4, probably reflecting its origins in the first era of larger-scale
investment disputes, sets the number at three. JCAA Rule 24(1) sets the number at one.
So do many arbitral institutions, especially since the 1990s, against the backdrop of rising
concern about delays and costs, but particularly in ICA cases. ACICA Rule 8 adopts a
compromise, leaving the institution to determine an appropriate number of arbitrators if
the parties cannot agree. ICSID (ICSID Convention, Article 37(2)(b)) maintains the
default rule of three arbitrators, one appointed by each party and with the Chair
appointed by the parties (or, if they cannot agree within ninety days of the request for
arbitration, by ICSID: Article 38). Having three arbitrators seems preferable for modern
ISA Rules, given the large amounts and complexity usually involved. In addition, allow-
ing the party-appointed arbitrators to appoint the Chair, as under the UR, ACICA, and
JCAA Rules, somewhat reduces a safeguard for the parties under the ICSID Rules.88
On the other hand, it should be possible to reduce the ninety-day time limit for
ICSID appointing the Chair (or for when one party refuses to appoints its arbitrator). For
example, ACICA Rule 10.2 allows the institution to appoint an arbitrator in lieu after
thirty days, and to appoint the Chair in lieu after thirty days. ICSID may have set and
maintained the ninety-day limit because when it appoints the Chair, the latter must be
from its Panel of Arbitrators (ICSID Convention, Article 40(1)). When the parties agree
on appointments, they are not so constrained. However, a solution for ICSID or those
devising ISA Rules like that institution would be to maintain a panel with members truly
ready and available to accept appointments as Chairs within quite short notice. Unfortu-
nately, that relies on the goodwill of Convention member states, since almost all the panel
members are selected by them rather than ICSID’s own Chair (ICSID Convention, Arti-
cle 14). By contrast, institutions like ACICA or JSAA are free to choose whether or not
to have a panel of arbitrators (both do), and how to admit members to the panel. Thus,
they can manage their panels so as to abide by a tighter time limit. They might even not
require institutional appointments from a panel, but that efficiency gain probably trades
off too much legitimacy.89
Similarly, it makes sense to follow ICSID regarding arbitrator nationality, when
other arbitral institutions seek to devise rules for ISA. ACICA Rule 9.3 only requires that
institution, when appointing the sole arbitrator (but not the third: cf. Rule 10) in lieu of
the parties, to take into account “the advisability of appointing an arbitrator of a nation-
ality other than the nationalities of the parties.” JCAA Rules 25 and 26 require the JCAA
to “consider” this factor, if requested, when it appoints either a sole or a third arbitrator.
The ICSID Convention (Articles 38 and 39) allows parties to agree that one or more of
their nationals will form the entire tribunal, but otherwise requires a majority of non-
nationals either agreed by them, or otherwise as selected by ICSID within the ninety-day
limit.
88
Incidentally, Australia and Japan could already reconsider the practice emerging under some FTAs or BITs,
nominating ICSID’s Secretary-General as the “appointing authority” for an arbitration under the UR (e.g., ACFTA,
art. 10.19(3)). In the Australia-Japan FTA context, for example, even if ACICA or JCAA does not end up with tailored
ISA Rules provided as an option for foreign investors in the treaty, those organizations might be assigned a role as
appointing authority in lieu of (or in conjunction with) ICSID.
89
Considerable concerns have emerged about two types of conflict of interest in ISA disputes: (i) lawyers serv-
ing as arbitrator in one case while as counsel for the investor or state in another case, and (ii) appointing arbitrators
from major law firms that generally represent large corporate investors (but not necessarily the investor involved in
that case). See H. Mann, Transparency and Consistency in International Investment Law, in Appeals Mechanism in Inter-
national Investment Disputes 213, 215–17 (K.P. Sauvant ed., 2008). The 2004 IBA Guidelines on Conflicts of
Interest in International Arbitration do not specifically address (ii), and would put (i) on the “waivable Red List.”
Given the different public interests involved in ISA, the implications of these conflict issues need further considera-
tion. The IBA Guidelines appear to have focused on ICA situations. The IBA Committee responsible for drafting the
Guidelines comprised 19 experts, of which at least eight came from firms with more than 275 lawyers: M.J. Bond,
A Geography of International Arbitration, 21 Arb. Int’l 99, 105 (2005). In this context, arbitral institutions could add
considerable “value-added” by developing a broad-based panel of readily available arbitrators specializing in ISA and
by maintaining standards higher than those set in the IBA Guidelines.
The possibility of this further constraint on ICSID’s scope to select arbitrators may
be another reason for that time limit. Again, however, a tighter time limit should be man-
ageable by institutions if they adopt ICSID’s approach regarding arbitrator nationality for
their ISA Rules. The approach seems preferable not only for perceptions of fairness in
ISA cases of high public interest, but because the substantive law applied involves public
international law rather than regular ICA. There, the tribunal often ends up applying the
substantive contract law of one of the parties, making it attractive to have more arbitrators
from that country.
Another area that raises considerations of efficiency, but indirectly also legitimacy, is
the remuneration of arbitrators under the current system of ad hoc appointments. UR
Rule 39(1) generally required it to be “reasonable,” taking into account, for example, the
amount in dispute. The Rules within arbitral institutions have become increasingly spe-
cific.90 ACICA Rule 40 requires an hourly rate, and allows for the institution to deter-
mine the amount if the parties cannot, considering also for example “the standing and
experience of the arbitrator.” The JCAA’s separate Regulations for Arbitrator’s Fees adopt
hourly rates, increased in 2008, but with caps based on the amount in dispute. The ICC
moved to a similar hybrid approach in the context of concerns that growing costs of arbi-
tration were linked to arbitrators not having sufficient incentives to limit the time they
spent on their cases. By contrast, ICSID’s Schedule of Fees simply sets a daily rate (USD
2,400 as of 2006;91 USD 3,000 under the Schedule in effect from 2008, which also dou-
bled the administration fee charged by ICSID and reintroduced a charge for its staff to
attend ICSID hearings outside Washington, DC).
Although this rate can be varied by party agreement (and apparently has been, quite
frequently), the ICSID default per diem seems low by ICA standards.92 Even so, other
institutions considering ISA Rules should set a similarly low rate, calculated instead on an
hourly basis. This could be combined with a cap based on the amount in dispute, but
bearing in mind that amounts in investment disputes are often large, even for quite
straightforward cases and those that proceed to awards.93 Such reforms would send a message
that the institution and their arbitrators were concerned with efficiency and cost savings
in proceedings. It might also signal that they were not operating a system to maximize
financial gain, reassuring the broader community as well as commercial and state parties.
In other words, this seems one area where efficiency and legitimacy make better bedfellows.
4.3. Time limits and ADR

Turning to rules more directly addressing the other major persistent concern in con-
temporary international arbitration: delays. What about the imposition of time limits,
90
K. Sachs, Time and Money: Cost Control and Effective Case Management, in Pervasive Problems in Interna-
tional Arbitration 103–16 (L.A. Mistelis & J.D.M. Lew eds., 2006).
91
Id. at 106.
92
Id. at 110–11.
93
Franck, supra note 23.
especially after the tribunal is formed? Time limits have been set for the rendering of
awards by some arbitration rules designed for ICA, and even occasionally by legislation
(e.g., the old Thai Arbitration Act 1987, section 21: 180 days). However, they raise the
risk of even well-meaning parties finding themselves lacking in mandate to complete the
arbitration, and hence unable to obtain an enforceable award. This risk is heightened
when the dispute types and amounts are disparate. Hence, time limits for rendering
awards are still mainly found only in smaller-value disputes pursuant to Expedited Rules
(e.g., JCAA Rule 65: three months, which Rules parties can opt out of anyway: Rule
59(1)(1)).
The UR do not provide for such limits. Nor does the ICSID system, and nor do
many FTAs or BITs. However, FTAs often seem to require “arbitral tribunals” to render
their decisions on trade law disputes between the states within fixed timeframes (e.g.
SAFTA, Chapter 16, Article 7(11): sixty days from tribunal formation; TAFTA, Article
1809(3): 120 days; ACFTA, Article 21.10(2): generally 180 days). There is also a looser
and more generous timeframe provided for complaints brought before the WTO. The
Dispute Settlement Understanding (DSU) expects a decision from a panel within nine
months, or from the Appellate Body within twelve months, of panel formation. How-
ever, the average time seems to have remained around fifteen months, despite the rate of
non-appealed cases doubling.94 Given the new issues of law still emerging in ISA cases, it
seems inadvisable to set strict time limits for rendering awards, at least in ISA Rules.
States can always experiment with time limits added into individual bilateral treaties.
Institutions with ISA Rules could also publish related guidance or practice notes encour-
aging arbitrators and counsel to complete cases within recommended timeframes, and
adjust other Rules (e.g., on arbitrator remuneration) to provide further incentives.
Indeed, even Karl-Heinz Böckstiegel seems to acknowledge the need for flexibility in
timeframes under ISA.95 Yet, Professor Böckstiegel is widely known as the “timely arbi-
trator.”96 He developed “chess clock” time limits initially to manage disputes brought
before the Iran-United States Claims Tribunal and this technique was extended by him
and others to regular ICA cases.
Should there be time limits for specific phases of the ISA process? Apart from the
timeframe for appointing those to resolve the dispute, which has proved particularly
problematic under the WTO panel procedure,97 one of the most common periods set in
ICA nowadays is that for exchanging written submissions. The UR left this issue to arbi-
trator discretion, and the 1996 UNCITRAL Notes on Organizing Arbitral Proceedings
offer no further guidance.The Notice of Arbitration (Rule 3) can be followed by a State-
ment of Claim within a time set by arbitrators (Rule 18), who also determine the time
allowed for the Statement of Defence (Rule 19) and any further Statements (Rule 20).
94
Davey, supra note 82, at 26.
95
K.-H. Böckstiegel, Enterprise v. State: The New David and Goliath?, 23 Arb. Int’l 93 (2007).
96
J. Paulsson, The Timely Arbitrator: Reflections on the Böckstiegel Method, 22 Arb. Int’l 19 (2005).
97
Davey, supra note 82.
The ACICA system is similar. However, after receiving Notice (Rule 4, which again
need not include the Statement of Claim), the respondent must provide an Answer
within thirty days (Rule 5.1), and arbitrators must generally set a forty-five-day limit for
all subsequent Statements (Rule 26). The JCAA envisages around seven weeks, with no
separate Statement of Claim. The claimant’s request (Rule 14) is forwarded by JCAA
“without delay” (Rule 15) to the respondent. The “Basic Date” (Rule 2(1)) is generally
three weeks from that, and the Answer must be within four weeks of that date (Rule 18).
ICSID Rule 26(1) allows the arbitrators to set time limits “where required,” rather
like the UR, and allows them to delegate this power to the Chair. It seems better to leave
this matter to the arbitrators in other ISA Rules, supplemented again by some guidance
or best practice notes. Even in disputes between states under FTAs that contain overall
time limits for rendering decisions (SAFTA 2003, TAFTA 2005, and ACFTA 2008), no
periods are set for ISA. Under the WTO’s Dispute Settlement Understanding, the
“working procedures” proposed generally for panels envisage at least six to twelve weeks:
three to six weeks for the complainant’s first written submissions, two to three weeks for
the response, then two to three weeks for any written rebuttals from the states involved.
However, in the review of the DSU process initiated in the late 1990s, these have been
criticized as too generous.98
In fact, ISA under SAFTA 2003, Article 14(4)(c) requires a separate written Notice
at least thirty days before the Claim, which thus extends time periods. By contrast,
TAFTA 2005 and ACFTA 2008 have no such provision. But delays are built into all three
agreements in that before initiating the ISA phase, it seems that the foreign investor and
the host state must negotiate. Under SAFTA, Article 14(3) and ACFTA, Article 10.16(1),
this is for six months (unless, it seems, they agree otherwise because there is no hope of
settlement). Footnote 6 to TAFTA, Article 917(2) allows “in principle” for three months
(with no express reference to party agreement). Perhaps Australia envisaged more disputes
with Thailand, and thus wanted ISA to become available earlier. However, setting a sig-
nificant period of time to resolve disputes amicably, in the shadow of a more formal proc-
ess, is a useful and common feature of disputes involving states. It still characterizes the
WTO system (DSU, Article 4(7): sixty days before the states can initiate panel proce-
dures), and is also widely found now in FTAs for disputes between states themselves.
In ICA, commercial parties increasingly achieve a similar result by ad hoc “multi-
tiered dispute resolution clauses.”99 Arbitral institutions specializing in ICA, like ACICA
and JSAA, have not gone the next step of requiring periods for good faith negotiations
before proceeding to arbitration. In ISA Rules, however, they should provide for at least
two months of negotiations to maximize the chances of resolving the investment dispute
quickly and cheaply. The Rules of institutions should require negotiations to be conducted
in “good faith” (not specified in SAFTA, TAFTA, or ACFTA), and their arbitrators
98
Id. at 27.
99
See, e.g., Mistelis, supra note 5; also see generally M. Pryles, Multi-Tiered Dispute Resolution Clauses, 18 J. Int’l
Arb. 159 (2001).
should be encouraged to enforce these obligations more vigorously.100 These institutions

could also use this period to begin considering who might be suitable and available to
serve as arbitrators if the parties cannot reach a settlement.
This “cooling off ” period could be shortened significantly, say to two months for
good faith negotiations, if ISA Rules provided for the arbitrators to become more
involved in encouraging the parties to settle their dispute even during the arbitral pro-
ceeding. As mentioned above, such “Arb-Med” remains controversial even in ICA. It
raises concerns about natural justice provided generally in the UR (Rule 15.1) and
ACICA Rule 17.1; for example, an Arb-Med via caucusing can also be undermined by
other Rules (such as arbitrator impartiality, or UR Rule 15.3 and ACICA Rule 17.4
requiring all information supplied by one party to be communicated to the other). Such
Arb-Med could also be a ground for annulling the award under ICSID (ICSID Conven-
tion, Article 50(1)) because of a “serious departure from a fundamental rule of proce-
dure.” This might explain why it seems much rarer in ICSID cases for arbitrators actively
to facilitate settlement compared to ICA.
However, more seems to be at work, for hardly any cases have been brought under
the quite detailed ICSID Conciliation Rules. Böckstiegel speculates that foreign investors
are often not minded to settle disputes about expropriation, because they do not intend
to invest again in that country. Yet more and more investment disputes are less about
expropriation, and more about “fair and equitable treatment” or other matters that could
be resolved without irretrievably cutting off all relations between investors and host states.
Böckstiegel also points out that host state officials may not have authority, or be prepared
to risk taking responsibility, for a settlement that “would normally entail having to admit
that the state is at fault to some extent.”101 Yet similar factors used to prevent states or
state-owned enterprises using ICA instead of court proceedings in purely commercial
100
S. Franck, Integrating Investment Treaty Conflict and Dispute Systems Design, 92 Minn. L. Rev. 161, 198 (2007);
but see, e.g., P. Lalive, Some Objections to Jurisdiction in Investor-State Arbitration, in International Commercial Arbi-
tration: Important Contemporary Questions 376, 387–88 (A. van den Berg ed., 2003). Rules (or treaties) could
probably go further to clarify that good faith negotiations must be attempted, for up to the specified time limit,
before the matter can proceed to arbitration. (Under the rules of ICA institutions and legislation nowadays, that is
the usual consequence of mult-itiered ADR clauses: see also art. 13 of the 2002 UNCITRAL Model Law on Inter-
national Commercial Conciliation, available at <www.uncitral.org/uncitral/en/uncitral_texts/arbitration/
2002Model_conciliation.html>.) Otherwise, there is the risk of ISA tribunals interpreting such requirements as
“directory and procedural rather than mandatory and jurisdictional” (S.G.S. v. Pakistan, 8 ICSID Rep. 383, para.
184, (2005), tribunal chaired by Florentino Feliciano, formerly on the WTO Appellate Body; see also the first Lauder
case, supra note 74). Even with such specifications, however, the mandatory requirements may be undermined if tri-
bunals apply broadly worded MFN provisions: Faya Rodriguez, supra note 60; Turner et al., supra note 60.
101
K.-H. Böckstiegel, The Role of Arbitrators in Investment Treaty Arbitration, in International Commercial Arbi-
tration: Important Contemporary Questions 366, 375 (A. van den Berg ed., 2003). A recent study highlights
how foreign (mainly U.S.) investors in Indonesia certainly proceeded quickly to arbitration. They showed little or no
inclination to negotiate or settle the disputes regarding the Karaha Bodas and Dieng projects, partially suspended fol-
lowing the Asian financial crisis. However, a major factor seems to have been those investors’ limited experience in
overseas infrastructure developments, followed by shifts in corporate strategy back against overseas expansion. The
withdrawal of Enron from the Pasuruan gas-fired power plant project also occurred in the context of Enron’s shift
away from physical investments in power generation and into (ultimately disastrous) trading in energy futures. The
authors also emphasize how political risks insurance, especially in the Dieng dispute, provided incentives not to seek
renegotiation or settlement. See Wells & Ahmad, supra note 54, at 15. In such cases, ADR before or during ISA pro-
ceedings does seem very unlikely to succeed. Nonetheless, it should remain more promising when disputes involve
investors and insurers with longer-term commitments to the host state or its region.
disputes. Despite that, attitudes have changed over time. Further, taking an investment
dispute through to an adverse ISA award may entail even more risks of displeasure vis-à-
vis the public officials involved.
Host states, in particular, may feel that the public interest issues at home require the
more formal proceeding and outcome provided by the ICSID Arbitration Rules. Yet
ICSID’s Conciliation Rules, for example, add some formalities that help address such
issues. Rule 33 demands publication of any settlement agreement reached. In any case,
there should be less objection to negotiated settlements if reached during the more formal
arbitration proceedings. A relatively minor problem in this respect may be that the ICSID
Arbitration Rules do not expressly provide that the arbitrators must record any settlement
in the form of an award, as under UR Rule 34(1), ACICA Rule 35.1, or JCAA Rule
54(2). Thus, any ISA Rules developed by arbitral institutions should adopt the ACICA
Rule, similar to the UR and the JCAA Rule. By requiring the arbitrators to record the
parties’ settlement, rather than leaving that to their discretion, the public can be informed
of the outcome.
Nonetheless, there appear to be more fundamental problems regarding arbitrators
attempting to facilitate settlement in ISA. Franck examined 102 ISA awards publically
available as of June 1, 2006 from eighty-two cases.102 Of those cases, fifty-two had been
finally resolved by that date (plus another involving Ecuador, where the award was not as
widely available); twenty-three were ongoing; but only six had settled or been discontin-
ued. Further, of the fifty-three awards, only three incorporated settlements reached by
the parties. Franck observes that all three were preceded by critical preliminary decisions
by the arbitrators. She speculates that, conversely, uncertainty about the substantive law as
well as questions of costs may be impeding settlement.
Reforms such as a permanent appellate review body and changes to arbitrator remu-
neration models (proposed above) may reduce such uncertainties. However, experience
in ICA shows how cautious Arb-Med can help further narrow divergences in the parties’
expectations about potential rulings by the arbitrators. Particularly in ISA, the natural jus-
tice and arbitrator impartiality concerns involved in promoting Arb-Med can be mini-
mized by prohibiting ex parte contact between a party and the arbitral tribunal (especially
if a sole arbitrator). Excluding caucusing may not be as effective in achieving settlement,
but experience in ICA shows that this approach is better than no settlement assistance at
all.103 This is also consistent with inter-state disputes under the WTO or FTAs (e.g., DSU,
Article 18(1)). Arbitrators may also draw on experiences in ICA indicating that settle-
ment facilitation may generate most net benefits after initial submissions have been
exchanged, but they can also successfully promote settlements at much later stages of ISA
proceedings.104
102
Franck, supra note 23, at 70–75.
103
See supra; M. Schneider, Combining Arbitration with Mediation, in International Dispute Resolution:
Towards an International Arbitration Culture 57 (A.J. van den Berg ed., 1998).
104
E.g., at the end of the evidentiary hearings in an investment treaty arbitration under generic London Court
of International Arbitration (LCIA) Rules: J. Bedard, The Sachs-Walde-Reichert Method, 2 Global Arb. Rev. 21
(2007).
Updating ISA Rules so arbitrators encourage settlement on this basis at any stage of
the proceedings should only be a small step for those familiar with the ICSID system.
After all, it already generally envisages a “pre hearing conference,” where if all parties so
request the issues may be considered “with a view to reaching an amicable settlement”
(Rule 21). However, that provision also highlights the importance of the parties’
informed consent to Arb-Med. The consent should be in writing, unlike under the
JCAA system where (similarly to Japan’s Arbitration Act 2003, otherwise based on the
UNCITRAL Model Law) the consent to Arb-Med may also be given orally ( JCAA Rule
47). Given that extra procedural safeguard, ISA Rules then might go a step further and
require arbitrators to try to settle disputes (i.e., to volunteer for consent-based Arb-Med, as
under German Arbitration Institute Rule 32.1), rather than leaving that to their discretion.
4.4. Interim measures and tribunal voting

Encouraging more proactive arbitrators through Arb-Med may lead to small but sig-
nificant improvements in the time and especially in the costs involved in resolving ISA
disputes. Another way in which arbitrations can be resolved quickly is when interim
measures are ordered. Such orders can have major practical effects. More generally, like
partial awards, interim measures can provide a “reality check” for the parties regarding the
underlying substantive issues and factual evidence, and thus help the case to settle early.
However, the scope of such measures appears quite narrow under the UR, as Rule 26
requires them to be related to the “subject matter” of the dispute. JCAA Rule 48 is to
similar effect. ACICA Rule 28 is expressed more broadly, and also adds other clarifica-
tions such as the standards for triggering the measure including a reasonable possibility of
the requesting party succeeding on the merits.These provisions are based on deliberations
at UNCITRAL as of 2005, which resulted the following year in a Revised Article 17 of
the UNCITRAL Model Law. However, due to natural justice concerns, ACICA Rule
28 does not allow for even the weak form of “preliminary orders” that the Revised
Model Law allows arbitrators to issue ex parte in limited circumstances.105 Something like
the current ACICA Rule 28, therefore seems most appropriate for revamped ISA Rules.
ICSID Rule 39 currently has a potentially broader scope than the UR, since it sim-
ply provides for “provisional measures” without further limitation or definition, and it
excludes ex parte orders. One more significant difference from ACICA Rule 28 is that the
ICSID arbitrators can recommend measures on their own initiative (Rule 39(3)). This
may be worth adding to ISA Rules, subject to the voting rules discussed below. The
major difference is that ACICA Rule 28.8 (following the Model Law) allows the parties
to approach a local court for interim measures, not just the arbitrators. ISA Rules should
105
New Zealand’s Arbitration Amendment Act 2007 goes a step further, however, incorporating the entire
Revised Model Law art. 17, including preliminary orders. See A. Kawharu, New Zealand’s Arbitration Law Receives a
Tune-Up, 25 Arb. Int’l 405 –422 Pt IV. (2008) Diverse views remain about such ex parte measures, which also relate
to the question of whether and how to allow Arb-Med (discussed above). See L. Nottage, Reforming International Com-
mercial Arbitration (ICA) Law: The U.N., New Zealand—Why Not Australia?, 7 Austl. A.D.R. Rep. 15–19 ( June 2008),
available at <www.arbitrators.org.au>.
follow ICSID Rule 39(6), which only permits this if “they have so stipulated in the
agreement recording their consent.” TAFTA 2005 provides no such stipulation. SAFTA
2003, Article 14.5 allows an approach to local courts for interim measures, but only prior
to the ISA commencing. ACFTA, Article 10.18(4) goes a step further by allowing this if
“the action is brought for the sole purpose of preserving the claimant’s or the enterprise’s
rights and interests during the pendency of the arbitration.” There seems therefore to be
considerable disparity in views on this issue, and restricting access to interim measures
along the lines of ICSID Rule 39(6) may better address fairness concerns.
Procedures for voting by the tribunal are related to the appropriate scope for interim
measures, as well as other matters discussed above, such as time limits. The UR requires
tribunal rulings on matters of procedure to be by majority vote or, if there is none or the
tribunal allows this, by the Chair (subject to the tribunal’s right to revise his or her ruling:
Rule 31.2). The JCAA, like many other arbitral institutions nowadays focusing on ICA,
gives more power to the Chair. If there is no majority and there is no express possibility
for revision by subsequent majority vote (Rule 7.2), the Chair decides. ACICA Rule
17.3 simply allows questions of procedure to be decided by the Chair, or indeed any
other arbitrator if the tribunal so allows this, but all such decisions are subject to possible
revision by the tribunal. ICSID Rule 19 leaves procedural orders to the tribunal, which
must decide by majority (Rule 16(1)). However, under Rule 20(1) the Chair should early
on “endeavour to ascertain the views of the parties.” This system does not seem to provide
for deadlocks, so the UR regime provides a better starting point for tailored ISA Rules.
Probably more important, however, are the provisions on voting for awards. The
UR require a majority vote (Rule 31.1), like ICSID Rule 16(1) again. By contrast, as
with other arbitral institutions focusing on ICA, ACICA and JCAA give the Chair the
ability to impose his or her own decision (Rules 32 and 7(2), respectively). Although this
may be much more efficient, it seems fairer for large and complex ISA cases to require
the Chair to “negotiate” with the other arbitrators to establish a majority position. This
issue has proved a major sticking point in negotiations within UNCITRAL regarding
reform of the UR. That bolsters our argument (above) that those Rules should now be
split into two sets “UR-ISA” designed for investment disputes, and “UR-ICA” for oth-
ers. The latter could retain greater power for the Chair.
5. Conclusions: Back to the Future for ISA

This article is not as ambitious as it may seem. First, we do not cover all possible
Rule changes.We focus only on some major ones, particularly from the perspective of those
more familiar with ICA. They may be tempted to import too readily their views and
experiences from that domain into the overlapping but distinct field of contemporary ISA.
Secondly, we do not attempt draft wording for the proposed Rule changes, although
we indicate which of the main existing Rules we compare appear to provide the best
starting point. In some relatively new areas, such as the admissibility of amicus curiae
briefs, considerable discretion may have to be left to the arbitrators.
Thirdly, and most importantly, we concentrate only on a range of possible improve-

ments to ISA processes. Our innovative but subtle proposal is to suggest ISA Rule changes
that individual foreign investors would not be bound to adopt, if instead they preferred
the existing UR or ICSID Rules regime.We do not examine here the many more radical
proposals made regarding the substantive law to be applied in ISA cases. Those include
integrating substantive social and environmental responsibility provisions into interna-
tional investment agreements, establishing investor liability within the home state for
activities carried out in host states, requiring consideration in ISA of competing host state
obligations under international environmental and human rights treaties, and proposing a
new multilateral investment agreement for sustainable development.106 Those proposals
also address, often more directly, the diverse public interests that are involved in interna-
tional investment disputes.
Still, the procedural improvements explored in this article promise several benefits,
at very little risk. Foreign investors obtain the advantage of opting in to tailored ISA
Rules that address many long-standing concerns, heightening the chances of host states
actually paying out promptly on awards that go against them. Investors professing a com-
mitment to corporate social responsibility, in particular, can signal their “good global cit-
izenship” by taking such a course.107 It is also in the interest of all investors to see some
form of ISA survive. Otherwise, more countries may begin simply to withdraw from the
ICSID Convention itself, following Bolivia’s complete denunciation in May 2007 (under
Article 71) or Ecuador’s notice in December 2007 that it will not consent to ISA (Article
25(4)) regarding natural resources investments.108 Procedural reforms also promote socio-
legal theory arguing that “any transnational regime should encourage innovation, exper-
imentation, and the capacity to imagine alternative futures for managing the relationship
between politics and markets.”109
A more balanced and therefore sustainable ISA process should also have spin-off
benefits for the development of ICA, particularly in countries on the periphery seeking
to move towards the core of the arbitration world, like Australia and Japan. ACICA or
106
IISD 2006, supra note 29; J.A. Zerk, Multinationals and Corporate Social Responsibility: Limitations
and Opportunities in International Law (2006); Miles, supra note 27.
107
See, e.g., Economist Intelligence Unit, Japan behind U.S., Europe in Adopting CSR, M2 Presswire 1 (2005);
R. Welford, Corporate Social Responsibility in Europe, North America and Asia: 2004 Survey Results, 17 J. Corp. Citizen-
ship 33 (2005).
108
See the official communications from the respective governments, available at <http://icsid.worldbank/org>.
Venezuela has reiterated a desire, shared with Nicaragua in 2007, to denounce the Convention, and meanwhile is
encouraging foreign investors in Venezuela not to commence any ICSID arbitrations if they wish to maintain lucra-
tive contracts in that country. However, the legal effect of denunciation remains contested: J. Fouret, Denunciation of
the Washington Convention and Non-Contractual Investment Arbitration: “Manufacturing Consent” to ICSID Arbitration?, 25
J. Int’l Arb. 71 (2008). A key issue is when and how “consent” under the ICSID Convention is completed. If that
involves not only the “offer” in the host state’s FTA or national legislation, but also the investor actually bringing the
claim (see, e.g., ACFTA, art. 10.17(2)), then the host state arguably can avoid ICSID arbitration by denouncing the
Convention before claims are filed, even regarding investments made after the FTA or legislation came into effect.
109
D. Schneiderman, Constitutionalizing Economic Globalization: Investment Rules and Democracy’s
Promise 8 (2008). However, Schneiderman focuses on substantive reforms, such as entrenching broader antidiscrim-
ination laws in host states and improving political risk insurance mechanisms. He also concentrates on national or
local responses to the new global FDI regime, while briefly acknowledging the potential for some supranational
responses (id. at 232–36). Compare, e.g., Constitutionalism, Multilevel Trade Governance, and Social Regula-
tion (C. Joerges & E.-U. Petersmann eds., 2006).
JCAA would only need one major ISA case to improve dramatically its standing among
the world’s arbitral institutions.These two organizations would also benefit more diffusely
by leading the way toward tailored ISA Rules, providing a model to be followed by other
institutions, both private (e.g., the ICC) or intergovernmental (especially UNCITRAL
and ICSID).
Most reforms necessarily involve a trade-off with efficiency. To ease that pain, devel-
oped countries should consider funding an “Advisory Centre on Investment Treaty Arbi-
tration” to assist developing countries. On the other hand, efficiency gains can also come
from certain reforms that we have proposed.Those include establishing an appellate body,
which would promote predictability, and capping arbitrator remuneration, which would
promote savings in cost and time. Once the ISA system has addressed its legitimacy issues,
it can undergo another round of reforms focusing more on measures to improve effi-
ciency. That is what happened with ICA. The experience simply reflects the balance con-
stantly negotiated between private and public interests not only in arbitration, but in
other forms of international dispute resolution as well.
Appendix
Rule UR ACICA JCAA ICSID – old ISSD Our proposals

[new] proposals
1. Disclosure OK No: r. 18.2 No: r. 40(2) OK: reg. 23 New r. 3.5, OK: fact of
of notice/ (confidentiality) notice of arbitration, plus
fact of arbitration to sufficient details
arbitration go to (as ICSID),
UNCITRAL possibly full notice
Secretariat of arbitration
and posted on unless objection
website
2. Disclosure OK No: r. 18.2 No: r. 40(2) – by New r. 15.3, NOT by

of (confidentiality) arbitrators: all docs to go arbitrators
documents No: r. 6.2 to Secretariat NOT by
from – otherwise and posted on institution
arbitration by parties: website, Otherwise OK
OK subject to by parties
redaction of Consideration still
confidential needs to be given
business to disclosure of
information pleadings
3. Non-party No: r. 25.4 No: r. 18.1 No: r. 40(1) No: r. 32.2 New r. 25.4, Extend same rules
access to (privacy) (privacy) (if one party open to as re amicus curiae
hearings objects; public,
tribunal must subject to
help protect protection of
proprietary confidential
information) business
information
(Continued)
[new] proposals
4. No: r. 25.4 But No: r. 18.1 No: r. 40(1) OK: r. 37.2 New r. 15.4, As [new] ICSID
Opportunity note that r. 15.1 (even without tribunal may
for amicus has been party consent, allow amicus
curiae interpreted as but tribunal submission,
giving the must consult subject to
arbitrators and safeguards
authority to safeguards)
accept amicus
briefs
5. Disclosure No: r. 32.5 No: r. 18.2 No: r. 40(2) – by New r. 32.5, Awards should be
of award (confidentiality (confidentiality) arbitrators: either party published
of award) No: r. 6.2 without OK for arbitrators
(still) consent; if have good faith
– by parties: tribunal to belief (or it turns
OK send to out) that parties or
– by ICSID: Secretariat ICSID has
OK with and posted on disclosed
consent website. Duty for ICSID to
of parties: Extend to promptly publish
r. 48.4, disclosure of award
discretion to any
publish “legal alternative
rules” applied settlement
(New 2006
Rules require
ICSID to
publish
promptly,
“excerpts
of legal
reasoning”)
6. Appeals on No No No Convention No A central,

questions of art. 52: new permanent
law or legal ad hoc appellate facility
interpretation tribunal can with tenured
annul but appointments
only for should be
serious established
procedural
irregularities
7. Number of Default rule is Default is Default is one: r. Usually, 3: rr. As in ICSID

arbitrators 3: r. 5 “determined by 24(1) 1–3, etc.
ACICA”: r. 8
8. Nationality If appointing Same: r. 9.3 Similar: JCAA Basically, As in ICSID

of arbitrators authority is to shall “consider” majority of
determine sole non-national if a non-nationals:
arbitrator party requests: rr. r. 1.3
(only), then 25(3), 26(5)
should consider
“advisability” of
non-national:
r. 6.4
(Continued)
[new] proposals
9. Time Notice of Similar, but after Around 7 weeks: Similar to – 45 days for
limits arbitration (r. 3) receiving notice claimant’s request UR, but time response to notice;
can be followed (r. 4, which (r. 14) forwarded limits may be – 60 days for
by statement of again need not by JCAA “without set “where statements
claim within include delay” (r. 15) to required” – cf., WTO, many
time allowed by statement of respondent; “Basic by the FTAs for inter-
arbitrators claim), the Date” (r. 2(1)) is arbitrators, state disputes: also
(r. 18), who also respondent must 3 weeks from that who may consider: “cooling
determine the provide an (or later, if notice delegate this off ” period, e.g.,
time allowed for answer within received then); power to their 1–6 months after
statement of 30 days (r. 5.1), answer within Chair (r. 26) dispute notified,
defence (r. 19) and arbitrators 4 weeks of that Cf., also before being able
and any further must generally date (r. 18) various to file, e.g., with
statements set 45-day limit Note: award overarching duty to negotiate
(r. 20) for all within 3 months FTAs or BITs in good faith up to
subsequent of arbitrator 30 days, and/or
statements (r. 26) appointment (r. option (or duty?)
65) unless parties of attempting
opt out of such mediation first for
expedited rules up to 60 days
10. Arb-Med natural justice: = natural justice: Arbitrator natural justice: Give tribunal
hybrid r. 15.1 r. 17.1 may facilitate r. 50.1 power (not duty)
settlement if annulment to facilitate
parties consent in ground settlement, but (as
writing or even in WTO/FTAs) no
(cf., 2003 Act?) ex parte
orally: r. 47 communications
(caucusing)
11. Interim Simple, and Broader and Like UR: r. 48 Broader than As in ACICA, still
measures limited scope more elaborate UR not ex parte, but
(measures (but cannot (“provisional allow tribunal to
related to be ex parte, cf., measures” itself recommend
“subject revised ML): undefined); measures, and
matter”): r. 26 r. 28 tribunal restrict access
can itself instead to court
recommend unless in original
measures; but “consent”
not ex parte;
recourse to
court only
if in original
“consent”:
r. 39
12. Tribunal’s By majority, By Chair, or Tribunal As in UR, rather

voting re but if none another if makes than ICSID
procedure or tribunal tribunal allows procedural
authorizes, by (subject to orders: r. 19
Chair (subject revision by
to tribunal tribunal): r. 17.3
revision): r. 31.2
13. Tribunal By majority: If no majority, By majority: As in ICSID:

voting on r. 31.1 Chair has casting r. 16 forces arbitrators to
award vote: r. 32 engage in reasoned
deliberations

JOIA2009002

Uploaded by

Copyright:

Available Formats

You might also like

JOIA2009002

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

JOIA2009002

Uploaded by

Copyright:

Available Formats

Journal of International Arbitration 26(1): 25–58, 2009.

© 2009 Kluwer Law International. Printed in The Netherlands.

“Back to the Future” for Investor-State Arbitrations:

Luke Nottage and Kate Miles*

1. International Commercial Arbitration Backdrop: Delays and Costs

2. Public Interest Concerns in Investor-State Arbitration

of mandatory minimum standards of substantive and procedural “public policy.”25 The

3. Incorporating Revised ISA Rules into Treaties or Rules

3.1. Why to retain ISA

3.1.1. Short-term Interests in Including ISA Protections

3.1.2. Long-term Interests in Including ISA Protections

3.2. How to retain ISA

4. Specific ISA Rule Revision Proposals

4.1. Disclosures, Third-party Participation, and Appellate Review

4.1.2. Access to Documents, Access to Hearings, and Amicus Briefs

4.1.3. Publication of the Award

4.1.4. Appellate Body

4.1.5. Advisory Centre

4.1.6. Australia-Chile FTA

4.2. Arbitrator appointment and remuneration

4.3. Time limits and ADR

should be encouraged to enforce these obligations more vigorously.100 These institutions

4.4. Interim measures and tribunal voting

5. Conclusions: Back to the Future for ISA

Thirdly, and most importantly, we concentrate only on a range of possible improve-

Rule UR ACICA JCAA ICSID – old ISSD Our proposals

2. Disclosure OK No: r. 18.2 No: r. 40(2) – by New r. 15.3, NOT by

6. Appeals on No No No Convention No A central,

7. Number of Default rule is Default is Default is one: r. Usually, 3: rr. As in ICSID

8. Nationality If appointing Same: r. 9.3 Similar: JCAA Basically, As in ICSID

12. Tribunal’s By majority, By Chair, or Tribunal As in UR, rather

13. Tribunal By majority: If no majority, By majority: As in ICSID:

You might also like