La Revue Squire

International approaches to the independence and impartiality of arbitrators

Rédigé par Christian HAUSMANN et Anna-Leigh BARKER le 30 Décembre 2011

The IBA Guidelines on Conflicts of Interest in International Arbitration (the “IBA Guidelines”) were developed to respond to a need for more uniformity in the standards applied in arbitration issues. In particular, the Guidelines provide guidance on when arbitrators should disclose issues which may cast doubt upon their impartiality and/or independence. However, the IBA Guidelines will not lead to uniformity if they are not implemented universally and, as the cases explored in this article show, the application of the IBA Guidelines has not, as yet, been as wide spread as the International Bar Association may have hoped for.

England has been singled out by commentators as an example of a country in which the application of the Guidelines could be problematic given that, in England, a test for determining whether or not bias exists has already been “crystallised” and it is this test which would be used by the courts to determine any issues arising in relation to independence and/or impartiality. Indeed, there are only two reported UK cases which have made reference to the IBA Guidelines and in both of these cases the submissions which were made in reliance upon the IBA Guidelines were not accepted by the judges.

In order to understand why the English courts are so seemingly reluctant to refer to the Guidelines, it is important to understand how the test for apparent bias has evolved under the English precedent case system.

The evolution of the concept of impartiality and independence under UK case law

The first point to note is that there is, in theory, a marked difference between independence and impartiality. In the context of arbitration, the presence of the former, which is an objective concept, would be used to indicate that there was no conflict of interest while the latter would denote the absence of prejudice and is inherently more subjective. However, in practice, this distinction has become almost academic as the terms are generally used interchangeably. Therefore, for the purposes of this article, the terms will also be used interchangeably.

Although it is a long established principle of English law that “justice should not only be done but should manifestly and undoubtedly be seen to be done” , it was not until the case of R v Gough that a test to be used in order to determine the existence of apparent bias was put into place. In handing down judgment in that case, Lord Goff endorsed a two-step test for assessing bias: firstly, the court should ascertain all of the relevant circumstances of the case and, secondly, the court should see if there is any real danger of bias as a result. That is to say whether there is a real possibility of bias.

Laker Airways Inc v FLS Aerospace Limited then extended the test set out in Gough to arbitrators . In this case, the arbitrator appointed by the respondent and its barrister both came from the same chambers. The court held that this was not enough to satisfy the test of bias. In making his decision, the judge paid particular regard to the fact that although barristers in the UK may work from the same chambers and therefore share the same resources, they are essentially self-employed and work independently of one and other, something he described as “one of the peculiarities of English law”. He stated that a reasonable Englishman would know this and would therefore not suspect bias. The test of apparent bias set out in Gough was therefore not satisfied.

However, the decision in Laker Airways was subsequently criticised by commentators on the grounds that the court applied the test in Gough too subjectively and was wrong to instil in the reasonable man, who is supposed to be neutral, an enhanced knowledge of the English legal system. This was particularly inappropriate in the Laker Airways case as the challenge was made by an American company which would not, therefore, have been aware of the intricacies of the UK system.

In 2000, Locabail (UK) Limited v Bayfield Properties Limited confirmed that, when personifying the reasonable man, he should not be given too much special knowledge, thus confirming that the decision in Laker Airways should have been made by reference to more objective criteria.

This reasoning was soon confirmed in the 2001 decision in Porter v Magill, which is now the leading authority on the test for apparent bias. Porter v Magill established that the correct test is whether all of the circumstances of the case, as ascertained by the court, would lead a “fair minded and informed observer” to conclude that there was a “real possibility” of bias. This test has been used by the English courts for a decade and, as it comes from a decision of the House of Lords, it will remain binding upon the lower courts until it is overturned by a decision of the new Supreme Court.

Given that the English courts have such a well-established body of case law to refer to in relation to the issue of apparent bias and given that they are actually bound by the ratio in Porter v Magill, it is perhaps unsurprising that the IBA guidelines have not played a central role in English decisions. However, it would be unfair to say that the English courts have disregarded the Guidelines completely: they have considered them but simply decided not to rely on them in the decision-making process.

The approach of the UK Courts to impartiality and the IBA guidelines

Until earlier this year, ASM Shipping Ltd v TMMI Ltd of England was the only English case to refer to the IBA guidelines . In that case, the arbitrator had previously been instructed by the respondent’s solicitors as a QC in an arbitration between different parties but with the same principal witness. In the earlier case, the party for whom the arbitrator had been acting as counsel had made serious allegations against the witness. The appellant argued that this would have affected the arbitrator’s ability to act impartially and applied for the arbitral award made by him to be set aside as a result.

In assessing the case, both parties agreed that the test for apparent bias would be that stated in Porter v Magill, i.e. what a fair minded and informed observer would conclude having considered the facts . The Honourable Mr Justice Morison concluded that, in his view, “the independent observer would share the feeling of discomfort expressed by [the principal witness] and would have concluded that there was a real possibility that the tribunal was biased” as a result of the arbitrator having previously acted as counsel against the appellant’s main witness . Thus the test in Porter v Magill was satisfied.

Mr Justice Morison then went on to explain why he had rejected the respondent’s argument that there could not have been a conflict of interest because the facts alleged did not fall within the IBA Guidelines’ Red List. Justice Morison confirmed that the issue in question was not “whether what happened fell within the Red List or not”, rather it was whether the test in Porter v Magill was satisfied. Justice Morison then highlighted that the IBA guidelines were indeed only guidelines, which ought to be “applied with robust common sense and without pedantic and unduly formulaic interpretation”.

One can glean from this that where there is a question mark over impartiality or independence, the English courts will apply the test for apparent bias, follow the precedent cases that have developed through the case history leading to the decision in Porter v Magill and will be hesitant to put too much emphasis on international guidelines.

This approach was confirmed earlier this year in the case of A & Ors v B and X . In this case, the appellant was applying to have the arbitrator removed and his award set aside because shortly before completing and issuing the award, the arbitrator had disclosed that he had been instructed as counsel for the respondent’s solicitors in an ongoing matter. This matter was wholly unconnected with the arbitration.

The application was ultimately refused on the basis that the fair-minded and informed observer would not consider that there was a real possibility of bias merely because an arbitrator had acted as counsel for one of the parties in the past.

In making his decision, the judge, his honourable Mr Justice Flaux, also referred to the test for impartiality set out in Porter v Magill and, just like Justice Morison in ASM Shipping, he was quick to dismiss submissions which were made in reliance on the IBA guidelines as opposed to established English case law. Indeed, Justice Flaux pointed out that the Introduction to the Guidelines “makes it clear that the Guidelines are not intended to override national law”. Accordingly, he explained that if there was no apparent or unconscious bias once the common law test for bias had been applied (the test set out in Porter v Magill), nothing written in the Guidelines could alter that conclusion. Accordingly, it seems that the Guidelines will only be used in the English court if what they say happens to compliment, as opposed to contradict, the precedent cases.

Other approaches to the IBA Guidelines

In contrast to the English courts, the courts of other jurisdictions, as well as the International Centre for Settlement of Investment Disputes (the “ICSID”), provide examples of cases in which strong reliance has been placed on the guidance provided in the IBA Guidelines.

The 2009 ICSID decision in Perenco Ecuador Ltd v Republic of Ecuador and Empresa Estatal Petróleos del Ecuador a challenge to an arbitration award was upheld on the basis of Standards 1 and 2 of the IBA Guidelines alone . However, the parties had expressly agreed that any challenge to arbitrators would be resolved according to the Guidelines. It was, therefore, entirely appropriate for the tribunal to base its decision of the provisions set out therein.

Further support for the IBA Guidelines comes from another decision of the ICSID, Alpha Projecktholding GmbH v Ukraine . Indeed, such close regard was paid to the IBA Guidelines in that case that critics have said that the case established the Guidelines as “an international law standard for disclosure by arbitrators”.

The case involved an application for the disqualification of an arbitrator on the grounds that he had studied at Harvard Law School at the same time as counsel for the respondent and the pair had remained friends. It was ultimately decided that this fact did not cast doubt over the impartiality and independence of the arbitrator to the extent that he should be disqualified.

In making its decision, the Tribunal referred to the IBA guidelines and noted that attending university with counsel for one of the parties in an arbitration was not listed on the IBA red or orange list and was not, therefore, something which ought to have been disclosed by the arbitrator.

The Tribunal explained that it had placed reliance on the Guidelines on the basis that the Guidelines applied the UNCITRAL justifiable doubts test which was the same as the test set out in the ICSID’s own Arbitration Rules . This begs the question as to whether or not the ICSID would have been as willing to apply the Guidelines if the tests and theories set out therein were contradictory to the provisions of the ICSID Arbitration Rules. Indeed, in other ICSID cases, such as Participaciones Inversiones Portuarias SARL v Gabonese Republic , the ICSID has not been persuaded by the Guidelines and its approach has been more similar to that of the English courts.

In Gabonese Republic case, the applicant challenged the appointment of an arbitrator on the basis that he had been the president of the Tribunal in an earlier case in which an award had been made against him. In its submissions, the applicant argued that this was something which fell within the Orange List of the IBA Guidelines and it would therefore be justifiable to disqualify the arbitrator as a result. The Tribunal was unconvinced by this argument and in handing down its judgment reminded the parties that “the IBA Guidelines are of indicative value only, even though they may sometimes provide useful indications” .

The same view was taken in Tidewater Inc & Ors v Venezuela . In this case, the claimant applied to have the arbitrator disqualified on the grounds that she had failed to disclose that she had been appointed by the respondent on a number of occasions in the past. The application was ultimately unsuccessful as the judge held that a failure to disclose previous appointments was not in itself sufficiently serious to warrant the disqualification of an arbitrator, especially as this non-disclosure was based on an honest exercise of judgment.

Interestingly, the arbitrator argued that her appointments by Venezuela were in the public domain and that the claimant should, therefore, have been aware of them, regardless of whether or not specific reference had been made to them during the disclosure process. The tribunal did not find favour in this argument and stated that the arbitrator was the person best-placed to provide information about her past appointments and should, therefore, have disclosed all of her appointments, regardless of the fact that they were in the public domain. To have concluded otherwise would put to great a burden on the parties to investigate arbitrators.

Comments such as these seem to suggest that there is a heavier burden on the arbitrator than the parties in the disclosure process. However, the fact remains that the non-disclosure was not sufficient to merit disqualification which has lead critics to argue that, despite the ICSID’s comments, parties should, in practice, continue to carry out their own due diligence into the previous appointments of arbitrators . This is wise, given that in the Ukraine case the ICSID took an entirely different approach and stated that the parties should be assumed to have carried out at least basic internet research into the opposing parties, their counsel and the arbitrators at and early stage in proceedings.

Going back to the relevance of the IBA Guidelines, the Tribunal in the Venezuela case noted that the IBA Guidelines had been “referred to extensively” by both parties in their submissions : Venezuela had argued that the Guidelines were not relevant to ICSID arbitrations and should not, therefore, be considered at all whereas Tidewater claimed that they were universally applicable to all arbitrations, including investment arbitrations.

In response, the Tribunal considered the decisions in the Gabonese Republic case and the Ukraine case and found more favour in the approach taken in the former . The Tribunal concluded that, although it was “useful” to have the guidance provided by the IBA Guidelines, such guidance could “be no more than a rule of thumb” and the correct approach was to follow the legal standard laid down in the Convention itself.

Thus, it would be unfair to suggest that the ICSID will always apply the IBA Guidelines. This is manifestly not the case: the ICSID will use the Guidelines as an indicative tool but will not give them undue authority where there are established legal provisions in place.

A similar approach has also been taken in other national courts. Indeed, in a recent case heard in the Court of Appeal of Madrid (case number 506/2011), the Court was unwilling to pay too close regard to the IBA Guidelines where the national law was sufficiently clear in relation to the concept of impartiality and independence.

In that case, the Court annulled an award rendered under the Spanish Court of Arbitration on the grounds that the arbitrator was not independent and impartial. Both parties alluded to the IBA Guidelines in their submissions. Commenting on the relevance of the Guidelines to the case, the Court noted that the Guidelines could be of use but that Spain’s own legal system, the doctrines of its Constitutional Court and its own national Arbitration rules were so clear on the topic of independence and impartiality that they had adequate means to determine the issues in question without having to refer to the IBA Guidelines. The criteria set out in the Guidelines did not, therefore, influence the Court’s final decision.

More striking than the Court’s disregard of the IBA Guidelines is perhaps the uncompromising approach which it took to the arbitrator’s lack of disclosure. The challenge centred upon the fact that the managing partner of the law firm acting for one of the parties had once acted as an intern for the arbitrator over 30 years earlier. The arbitrator had subsequently dedicated a book to the managing partner and had developed friendships with a number of employees from the same law firm. None of these facts were disclosed until the arbitrator was specifically questioned about them, something which the Court considered cast doubt over his independence and impartiality. Indeed, the Court held that “the mere existence of reasons that can cast doubts on the impartiality of the arbitrator should be enough for him to resign” and found the fact that he had not disclosed . This is a much stricter approach than has been taken by other courts in relation to disclosure and certainly suggests that the responsibility for disclosure lies firmly with the arbitrator as opposed to the parties.

The reason that this decision is so striking is that, in the same year, the Court of Appeal of Madrid suggested that it is the parties, and not the arbitrator, who should take responsibility if there is a lack of transparency in the disclosure process (case number 338/2011).

In that case, the Court of Appeal refused to annul an arbitration award which was challenged on the basis that the arbitrator was a director and proxy in a number of companies which had participated in joint ventures and had commercial relationships with one of the parties to the arbitration. The arbitrator had disclosed the fact that he was a director and proxy to certain companies but had failed to explain the links that those companies had to one of the parties. The Court refused to accept the challenge and based its decision, amongst other things, on the fact that although the arbitrator had not disclosed the link between the companies he was a director of and the party to the arbitration, this information was readily available on the internet. As the undisclosed information was in the public domain, the other party to the litigation could have found it if it had made further enquiries. Accordingly, there was no reason why the other party to the arbitration should not have known that the links in question existed.

This decision shows the Court of Appeal taking a different stance than it did in case number 506/2011 by putting a heavier burden on the parties than the arbitrator. It also stands in contrast to the Venezuela case which indicated that an arbitrator has a responsibility to disclose even if the information in questions was in the public domain and therefore available to the parties.

Lessons to be learned

Commentators have said that, in spite of the introduction of the IBA Guidelines, there “is still much discussion over what constitutes full disclosure” and indeed these very different decisions illustrate that this is most definitely the case. The divergent approaches which are being taken by courts all over the world evidence the fact that decisions are being made on a case by case basis. Indeed, it is not even clear whether a heavier burden will be placed on the parties or the arbitrators. The lesson to be taken from this is surely that all parties in arbitrations should ensure that they protect their own position as much as possible; the parties should carry out a high level of due diligence and arbitrators should ensure that they make full and frank disclosure.

In relation to the relevance of the IBA Guidelines, it is clear that courts recognise that the Guidelines are of useful indicative value. However, it is equally clear that the Guidelines will not play a part in the decision-making process unless the relevant national legislation or Arbitral Rules are unclear or the parties to a dispute have expressly agreed to be bound by them. This seems like the correct approach to take; indeed, as Justice Flaux explained, it is stated in the Guidelines themselves that they were never intended to override national law.

Nouveau commentaire :

Vous souhaitez recevoir nos articles par mail, saisissez ci-dessous votre adresse mail :