Ref. Ares(2016)7145736 - 22/12/2016
Ref. Ares(2017)1963547 - 13/04/2017
Presentation on Issues in the EU/Canada Discussion Paper on
Establishment of a Multilateral Investment Dispute Settlement System
Adjudicators' Qualifications; Independence and Neutrality
] , Investment Division, OECD
14 December 2016
I would like to congratulate the European Commission and Canada for
developing the discussion note and bring governments together to discuss these
important issues. I would also like to thank the organisers for inviting us to
participate in this important meeting.
I have been asked to address two issues: first, qualifications of adjudicators and
second, the issues of independence and neutrality. I will deal with them in turn.
Governments and others addressed the desirable qualifications for investment
treaty adjudicators at our 2016 Investment Treaty Conference and the issue was
also discussed in October between governments at a meeting of our Investment
Treaty Dialogue. Those considerations together with a few additional ones will
hopefully provide a good initial basis for a discussion on qualifications today.
I will address four issues relating to qualifications: (i) treaty practice on
adjudicator qualifications; (ii) the nature of the adjudicator pool; (iii) possible
different or additional qualifications to those in the discussion paper; and (iv)
generalist vs specialist adjudicators.
Treaty practice on adjudicator qualifications
A first point is that, in current practice, the necessary qualifications for
investment arbitrators are expressed only in very general terms. Few investment
treaties specify required competencies for arbitrators. Those that do use very
general terms. So disputing parties and appointing authorities generally have
broad freedom to choose people as arbitrators or potential arbitrators.
Second, the EU, Canada and Viet Nam have innovated in their recent treaties.
They have adopted qualifications that are similar to those for international
courts. Adjudicators must either be qualified to be domestic law judges or be
jurists of recognised competence. Expertise in public international law is
required. Expertise in investment and trade law, including in dispute resolution,
is desirable. The discussion paper identifies similar criteria – it’s an approach
inspired by the qualifications for judges at international courts.
Let’s turn to the question of the current pool of arbitrators in ISDS.
The pool of arbitrators
It is useful to start from the characteristics of the known pool of ISDS
arbitrators based on available information. Noted characteristics from our 2012
analysis include elite status in the legal profession, very high levels of
compensation, a high representation of private lawyers with commercial
arbitration experience, less representation of government backgrounds, very few
if any serving government officials (such as investment treaty negotiators), a
high representation of European and North American countries and a 95%/5%
gender distribution. It appeared that over 50% of ISDS arbitrators had acted as
legal counsel for investor claimants in other cases, while approximately 10%
had done so for respondent states. Things may have evolved somewhat since
our analysis but the broad outlines are likely still accurate.
Participants yesterday and today have pointed to the importance of diagnosis --
of identifying the issues that need to be addressed.
The description of the current pool of ISDS arbitrators raises a number of
potential issues for governments including issues of diagnosis. For example, one
question for governments is whether an adjudicator pool composed of private
lawyers to a significant degree is well adapted to the questions at issue in ISDS.
It can be framed as an issue of competence or legitimacy. We know that at the
WTO the pool of panellists for trade cases comprises many government
officials, often trade diplomats here in Geneva. They have also been a major
presence on the WTO Appellate Body. We also know that that approach at the
WTO goes along with a strong Secretariat role.
I think it is fair to say that some of the support for more of a court model has
been inspired by a desire to change the composition of the group of
adjudicators. A change is seen by some as important in protecting the right to
regulate. It is also advanced as necessary for legitimacy. There are of course
different views on these questions. Some, like the Association of German
Judges, consider that the rules developed by the EU are not clear enough in this
regard and that a change in adjudicator personnel is not certain. They think that
the systems for nomination will be determinative in achieving what they see as
necessary change away from a pool composed primarily of commercial
I would observe here that the issue of a change in adjudicators is not necessarily
linked with a court model. It could be achieved within the context of the
existing system. In fact, without taking any position on whether a change of
adjudicators might be desirable, the governments in our investment Roundtable
have requested us to initiate a dialogue with appointing authorities under the
current system to learn more about their role and views as part of our work on
ISDS. The interesting input from the appointing authorities at this meeting --
Meg Kinnear (ICSID) and [Art. 4.1(b)
) -- makes the value of that
dialogue clear for governments and others.
Appointing authorities are a focus of interest in part because they have a key
role in determining the nature of the arbitrator pool -- for several reasons. First,
they play a direct role if the parties cannot agree on a chair – they either select
the chair or provide a list from which he/she is selected. As Meg's and [Art. 4.1(b)]
interventions here suggest, that direct role appears to growing in part as a result
of the polarisation of the pool of arbitrators.
Basic negotiation theory suggests a broader influence of appointing authorities
on the arbitrator pool in two ways: (i) disputing parties and their counsel
negotiate about an agreed chair against the background of what they would
expect to get from the appointing authority, which affects the negotiations; and
(ii) disputing parties and their counsel select their co-arbitrator against the
background of the expected chair.
Some appointing authorities have told us they are trying to address certain
issues about the arbitrator pool in different ways.
There are other measures government could take with regard to the pool of
adjudicators. Governments could also seek to develop potential adjudicators
with public sector or public law backgrounds. They could try to address
qualifications in their treaties.
So there are a number of potential approaches if the diagnosis were that a
change in the backgrounds and experience of adjudicators is needed -- although
a court model would likely be the most direct way to achieve that outcome.
Additional or different qualifications to those set forth in the discussion paper
The discussion paper refers to some qualifications as I noted. In terms of
possible additional or different qualifications, I will just list some considerations
that have been raised. There are of course important issues of regional
representation and diversity that Meg addressed and that I will omit here.
Some business and lawyer groups have advocated for an express requirement of
knowledge of international investment law. They considered that investment
law is complex and requires specialised knowledge.
Some governments like the US and Japan have emphasised their interest in the
ability to choose an ad hoc arbitrator with expertise or knowledge relevant to
the matter in dispute. They have seen this as important in providing a tribunal
with competence to address the issues in the particular case or in explaining
their success in defending cases.
It has been suggested that public law expertise – on issues like constitutional
and administrative law – is an important qualification. Investment treaty cases
are often described as a form of international judicial review involving a claim
that a government has engaged in misconduct towards a private actor. These
issues are similar to the types of issues regularly considered by public law
judges in domestic law or by human rights court judges.
Some see selective access for foreign investors as the main problem with ISDS.
They consider that different or broader qualifications for adjudicators would not
address the key issue.
Since we are at preliminary stage that involves a degree of issue spotting, I
would raise a fairly novel point. ISDS involves very large awards of legal costs
against losing parties - frequently running into the millions of dollars or euros.
This is unusual in public international law - each side usually bears its own
There is little in the way of institutions in ISDS to evaluate the reasonableness
of costs claims. Domestic systems that shift litigation costs have such systems
which typically involve costs specialists rather than judges. It is not seen as a
role that judges can easily or efficently carry out. So consideration could be
given to providing for people with qualifications in that area, again both in a
possible investment court or in the current system.
Generalist vs. specialist adjudicators
Another issue is the choice between relatively generalist judges or experts in a
particular field. Many international and domestic courts are staffed by
generalist judges. Such courts generally address technical issues through the
use of experts. It was suggested that the broader perspective of generalist judges
can important in the balancing of interests. For example, the WTO decision to
select as initial WTO Appellate Body members distinguished individuals who
were not closely associated with the international trade community has been
described as important in the development of the AB as a separate institution
with an ability to guide panels. As noted, others see a need for specialised
There are important issues of diagnosis for governments. There is also a need
for more information about the current system.
There are probably limits on the power of general criteria in a treaty although
they serve a signalling function. The nomination systems, as described by Meg,
may play a greater role in practice in defining the adjudicator pool.
Independence and neutrality
This is another sensitive and important topic. We have a lot of expertise in the
room today and can benefit from it. I am going to try to provide some
background on some of the major institutional issues that are presented by the
discussion paper in comparison with investor-state arbitration.
Public and parliamentary concerns about investor-state arbitration in a number
of jurisdictions have focused to a considerable degree on criticism about private
"for-profit" arbitrators. It is an important question for governments to consider
as part of their diagnosis. It goes primarily to issues of legitimacy.
The issue of the economic incentives of ISDS arbitrators was raised by some
governments early in our inter-governmental discussion of ISDS. It is discussed
in our 2012 scoping paper on ISDS
and the 2012 progress report
investment Roundtable and the Roundtable has kept it under consideration since
I am sure many of you are familiar with some of the concerns in this area from
the public debate. The ad hoc nature of arbitral appointments is seen by some as
generating an interest in future appointments. The payment of arbitrators by
parties on a per hour basis is seen by some as creating an incentive for
arbitrators to, for example, accept jurisdiction and thus to continue to earn
substantial fees. This view is not limited to NGOs. In an article that many of
you may have seen, The Economist magazine pointed to this issue in light of
arbitral fees running to 600-700 dollars an hour. There is also criticism of the
possible multiple roles of the same person as arbitrator, legal counsel and
expert, including concurrently in different ISDS cases.
Sundaresh Menon, who was then the Attorny-General in Singapore and is now
the Chief Justice, addressed these issues in a well-known 2012 speech. He
underlined both the excessive nature of some criticism and the existence of
issues for consideration. Here is part of what he said:
Unbridled criticisms of how arbitrators are invariably
profit‐driven and biased, or that they always act
strategically so as to be repeat players, are
undoubtedly overstated. However, it is undeniable that
the typical conditions that assure impartiality in the
judicial sphere are lacking in arbitration.1
So I think it is fair to say there are public concerns. That raises issues of
diagnosis for governments.
Some participants yesterday referred to the importance of preserving the
strengths of the current system. One key element is the protection of investor-
state arbitration tribunals from being influenced by host state governments. So
consideration of any changes needs to bear that in mind.
It is useful to start from where we are. I will outline the current situation in five
First, applicable principles in this area are again quite general at present.
Generally, they are limited to requiring independence and/or impartiality
without further detail. They are outlined in our 2012 ISDS scoping paper.
Second, alongside the applicable principles, the arbitration bar has developed
more detailed principles, notably the IBA Principles.
Third, there have been a considerable number of challenges to individual
arbitrators in which the applicable general principles are applied.
Fourth, Codes of conduct have emerged as an important issue. A range of
treaties calls for governments to develop a code of conduct. Rules developed by
the bar are not seen as fully adequate.
Fifth, in the context of challenges to individual arbitrators and rules within the
arbitration system, the issues of ad hoc appointment or compensation by the
hour are not at issue. They are taken as given.
The same is likely to be true of codes of conduct to be developed under treaties
that call for investor-state arbitration.
See Sundaresh Menon, International Arbitration: The Coming of a New Age for Asia (and
ICCA Congress 2012, Opening Plenary Session, §§ 39-40.
Discussion paper issues
Let me turn now to the discussion paper. The EU and Canada have emphasised
the importance of investment treaty dispute settlement meeting the standard of
well-established domestic courts and international courts. Investor-state
arbitration has not often been subject to evaluation from this perspective.
As I mentioned yesterday, we compared ISDS to the WTO and the ECHR in
our initial work in 2012. In our public consultation, we asked for views about a
number of issues relating to conflicts of interest. But we did not ask questions
about the differences between investor-state arbitration and well-established
domestic courts. We did not for example ask about the ad hoc system of
nomination or per hour compensation as such.
Today, there is growing interest in this comparison. In a number of
jurisdictions, domestic judges and academics are examining investor-state
arbitration. That scrutiny is likely to continue as cases continue to emerge. And
the discussion paper squarely presents the issues.
Although more research is necessary, it appears that few if any national
judiciaries today provide for payment of judges on a per case or per hour basis.
Strong national judiciaries generally are paid by a fixed salary and have security
of tenure. These characteristics are generally seen as important to judicial
At the same time, most jurisdictions strongly support commercial arbitration
which operates using ad hoc appointment and payment by the hour.
In beginning to analyse this issue, we have started to look at the historical
development of the principles applicable in domestic courts - this is preliminary
work in progress.
Salaries for judges were adopted in the late 18th and early 19th century in
countries like England, the US and France. Article III of the US Constitution of
1789 has been seen as requiring fixed salaries for federal judges. Fixed salaries
replaced systems in which judges were paid at least in part by fees paid by
Remuneration for judges through litigant fees was seen by prominent critics and
law reformers as having some defects similar to those seen by critics of
investor-state arbitration such as: (i) encouraging expansion of cases; or (ii)
promoting competition for cases. The critics included prominent philosophers
and law reformers such as Jeremy Bentham, notably in a major work edited by
John Stuart Mill.2 There are some interesting parallels between the situations,
but of course also many differences.
As noted, Article III of the US Constitution of 1789 has been seen as requiring
fixed salaries for federal judges. Judicial remuneration was limited to salaries by
1825 in England – any litigant fees were henceforth payable to the State, not to
In 1852, not long after this reform, the House of Lords decided what remains
the leading case on pecuniary interests of judges in the UK (Dimes v
Proprietors of Grand Junction Canal
(1852) 3 HL Cas 759). The case
established what has been termed the rule of “automatic disqualification” where
a judge has a pecuniary interest in a case. There is no inquiry into whether the
judge was actually influenced by the pecuniary interest as a matter of fact –
what matters is the appearance of bias.4
A similarly strict rule was articulated as a matter of US constitutional law by the
US Supreme Court in 1927. (Tumey v. Ohio
, 273 U.S. 510 (1927)). It is
reflected in a 1974 statute that applies to the federal judiciary. (28 U.S.C. §
The focus in this context is thus on the appearance of bias rather than on
whether there is actual bias. This may be an important consideration for
governments in considering issues of diagnosis and legitimacy. Yesterday, a
participant pointed to an important distinction between perceptions of problems
in ISDS and actual problems as a matter of fact. In the field of bias and in
particular bias based on financial interest, perceptions are facts. In the well-
known words of Lord Hewart, justice must not only be done, but must be seen
to be done.
Security of tenure is also generally seen as a key element of the independence of
national judges. A key advance occurred in 1701 when the UK parliament
eliminated the power of the monarch to remove judges and established judicial
tenure during good behaviour. (Act of Settlement, 1701) Parliament claimed the
power to remove judges for itself. Today it is seen as vital that the term of office
Jeremy Bentham, Rationale for Judicial Evidence, Part 2, Book VIII, chs. II-XIX
See Daniel Klerman, Jurisdictional Competition and the Evolution of the Common Law, 74 U. Chi.
L. Rev. 1179, 1204 n.52 (citing Statute of 6 Geo IV, ch 82, § 11 (1825) (taking fees away from the
Chief Justice of King's Bench); ch 83 § 10 (1825) (taking fees away from Chief Justice of Common
Pleas); Statute of 39 Geo III, ch, 90, §§ 1-3 (1799) (taking fees away from puisne judges of King's
Bench and Common Pleas and all judges of Exchequer)).
See, e.g., Grant Hammond, Judicial Recusal: Principles, Process and Problems (2009) ch. 4
(Automatic Disqualification in the British Commonwealth). A well-known English administrative
law treatise summarises the applicable law. Paul Craig, Administrative Law (7th ed. 2012) § 14-002.
of a judge and his or her remuneration can only be terminated in narrow and
defined circumstances. Security of tenure is notably seen as providing
protection for a judge from external pressures and for allowing him or her to
make difficult or unpopular decisions.
As I noted, these principles for judges do not preclude parties to contracts
agreeing to commercial arbitration mechanisms that involve ad hoc nominations
and fee-based compensation for arbitrators. Indeed, as [Art. 4.1(b)
yesterday, commercial arbitration is a vitally important part of the dispute
settlement landscape both for domestic and international commercial disputes.
It is important to recall this importance and the value of commercial arbitration
in the discussion about ad hoc appointment and remuneration.
Perceptions about investment arbitration, however, may differ from commercial
arbitration for several reasons:
investment treaty cases are brought only by investors. The alignment of
economic incentives on certain issues is thus consistent – for example the
investor will always be the party seeking broad findings on jurisdiction.
In commercial arbitration, the parties to a contract do not know
beforehand who will be the claimant and who may be seeking, for
example, broad interpretations of jurisdiction.
unlike commercial arbitration, ISDS frequently involves issues of public
interest and taxpayer liability for any damages.
the repetitive nature of a relatively small number of investment law issues
and the greater use of decisions as precedent compared to commercial
arbitration where cases generally turn on specific contracts.
the role of some ISDS tribunals, unlike commercial arbitration tribunals,
in judging issues of due process in host States.
These factors may generate higher expectations about the due process in ISDS.
In light of the specific purpose of this meeting as a without-prejudice discussion
of a possible multilateral dispute settlement based on the discussion paper, I
have focused on some of the key institutional changes proposed by the
discussion paper and the domestic and international court standard used in the
There are many other more detailed issues about independence that would need
to be addressed. Here governments can look to the substantial body of
comparative law work of standards of judicial independence by bodies like the
Council of Europe and other regional bodies as well as to their own standards of
judicial independence and impartiality. Many of these principles developed by
regional organisations are designed for application both in domestic and
As I noted, as governments consider whether or how certain issues should be
addressed, it is important to preserve the strengths of the current system. A key
consideration is the preservation of the independence of the adjudication system
from the influence of the host government. In national and international courts,
fixed salaries and security of tenure are generally seen as vital aspects of
independence in those areas as well, but additional safeguards are needed.
Participants have also noted the importance of the broader investment treaty
context, including reform to investor-state arbitration and to substantive law.
These factors also need to be borne in mind in considering dispute settlement