Sunday 14 April 2024

What Is Arbitrator Bias? (or: No Good Deed Goes Unpunished)

In February 2024 the High Court in London handed down a decision to remove an arbitrator in a film industry dispute, which, in addition to illustrating the importance of insurance in the production sector, illuminated a paradoxical issue in commercial arbitration: if the parties appoint a subject-matter expert to decide their dispute, does the arbitrator risk being removed for bias if, on the basis of his experience, he expresses a conclusion about the evidence before it has been tested at a hearing? 

The case, H1 and H2 v W, D and F [2024] EWHC 382 (Comm), concerned liability under film production insurance. The insurer contended that the arbitrator, an industry professional without legal training, had given grounds for “justifiable doubts as to his impartiality” and sought an order to remove him under section 24 of the English Arbitration Act 1996.  Mr. Justice Calver granted the order. 

The Dispute

The claimants were producing a television series in Sweden and had obtained insurance to protect them in case a specified actor suffered an injury that delayed production.  The actor duly incurred a serious injury during a stunt and the producers claimed under the policy.

The insurance policy incorporated an arbitration agreement, under which any dispute was to be resolved by “an experienced practitioner in film or television programme production”.  The parties being unable to agree on an arbitrator, the British Film Institute was asked to do so.  The BFI selected one “W”, an experienced industry professional and not a lawyer.

In subsequent terms of reference, the parties agreed to waive “any objection to the appointment of the Arbitrator on the grounds of potential conflict of interest and/or lack of independence or impartiality in respect of any matters known to them at the date of the agreement.”

The insurer pleaded, in essence, that they had excluded liability for any “hazardous activity” and that it was also a condition of the policy that the producers, their agents and their employees should use all reasonable care to avoid accidents.  The producers contended that they had relied on an independent stunt coordinator who had approved the stunt and that any hazard arose from unauthorized changes to the stunt made during shooting by third parties, for which the producers could not be held responsible.  It was said that there was “a central issue in the arbitration” as to industry practice in Sweden: was the stunt coordinator responsible for safety on the set or was it the responsibility of the producers?

Both sides proposed to call expert witnesses on this issue.

The evidence of bias

“I know all three [of the insured’s experts] very well, … what [the witness] said is what I will believe.. I absolutely believe what he says... if that’s what he said I would accept that… I know the professionals. I can say now what I think”.

At a procedural hearing, the arbitrator said that he knew the claimant’s experts, that one in particular was “one of the top Norwegian producers” and that he would accept what he said about the issue of on-set responsibility in Sweden, even though it differed from W’s own views as to industry practice.  He did not know the insurer’s witnesses, arguably depreciating their authority.  He indicated that he did not think the experts needed cross-examination.  In response, the insurer’s counsel submitted that W should listen the evidence before reaching a conclusion and W said that he would, albeit he had formed a view. 

The judge held that these comments gave rise to justifiable doubt as to W’s impartiality, on the basis that he appeared to have prejudged the allegedly central issue.  He revoked W’s appointment.

According to the judge, “the parties might expect the arbitrator to use his special knowledge of the film industry to understand the evidence that is given at the arbitration and any usages of the trade. But they would not expect him to use his special knowledge to pre-determine that he would prefer the evidence of an industry insider whom he knew (both personally and by reputation) over one whom he did not know” [para. 83].  The judge treated the arbitrator’s strong view about the witnesses as evidence of bias.

But what is “bias”?

According to England’s highest court, to be affected by bias is to be “motivated by a desire unfairly to favour one side or to disfavour the other.” (R v Gough (Robert) [1993] UKHL 1, at 4, per Lord Goff.)  Lower courts have said, more vaguely: “Bias can come in many forms. It may consist of irrational prejudice or it may arise from particular circumstances which, for logical reasons, predispose a judge towards a particular view of the evidence or issues before him” (Director General of Fair Trading v Proprietary Association of Great Britain [2000] EWCA Civ 350, at para. 37).  However, what unites all instances of bias is the influence on the tribunal of a collateral factor irrelevant to the proper disposal of the proceedings.

If the parties select a subject-matter expert as arbitrator, let alone where they expressly waive any objection of partiality based on his existing knowledge, they accept that this knowledge can be put to use by him.  The authorities are clear that an arbitrator, at least if an industry professional rather than a lawyer, is entitled to use his personal knowledge (for England, see: Mediterranean and Eastern Export Co Ltd v Fortress Fabrics (Manchester) Ltd [1948] 2 All E.R. 186; for the US: Sapp v. Barenfeld, 34 Cal. 2d 515, 212 P.2d 233 (1949)). 

When an adjudicator expresses a preference about the evidence, that is not in itself any indication of a lack of impartiality with regard to a party.  “The mere fact that the case of one party is regarded with disfavour is manifestly not enough to raise a danger of bias. It is necessary to concentrate on the assertion that there is a danger that this regarding with disfavour was the result of some unfairness” (Porter v Magill [2000] 2 W.L.R. 1420 (C.A.), at 1457).  W clearly wanted to get to the truth of the matter, having asked the parties to make 12 individuals available to answer questions.  However, the parties provided witness statements for only two.  He actually disagreed with the expert witness as to industry practice, but was willing to accept his evidence.

The problem in this case was one of due process, not bias.  The limitation on the arbitrator is not that he may not use his existing knowledge, but that he must ensure that it is deployed with procedural fairness.  In particular, he must enable the parties to challenge his personal view in case it is not already being contended for by one of the parties (Fox v. Wellfair Ltd [1981] 2 Lloyd's Rep. 515).   “It has to be emphasized, however, that the duty to act fairly is quite distinct from the autonomous power of the arbitrators to make findings of fact.” (Bulfracht (Cyprus) Ltd. v Boneset Shipping Company Ltd. "MV Pamphilos" [2002] EWHC 2292, at page 10, per Colman J.)

What happened here?  The arbitrator shocked the insurer’s counsel by indicating that he would accept the evidence of the producers’ witness irrespective of any cross-examination or submissions.  This was because he preferred that witness as a witness. This was evidence of W’s incompetence, as a layman, in the conduct of an arbitration, not of bias in favour of the producers.  It was precisely his function to form a view about the persuasiveness of the parties’ witnesses.  There was no reason to suppose he would have said anything different about him if the “top Norwegian producer” had been called by the insurer, rather than the producers.

Had W not accepted the admonishment of the insurer’s counsel that he should allow the process to unroll in the normal way and in fact disallowed or ignored cross-examination, he could have been accused of failing “properly to conduct the proceedings”, a distinct ground of challenge under section 24(1)(d) of the 1996 Act.  However, it remained to be seen whether, educated by the lawyers appearing before him, he did in fact take account of cross-examination and the advocates’ arguments when the hearing took place.  W’s expressed preference for the producers’ expert was certainly potential misconduct, but it was a category error to treat it as evidence of bias.

Substantial injustice

Did it matter?  Arguably, yes: to remove an arbitrator for improper conduct of the proceedings under section 24(1)(d), the applicant has to show that “that substantial injustice has been or will be caused to the applicant”.  This depends on what would have happened at the end of the arbitration:

“The test for substantial injustice focuses on the issue of whether the arbitrator has come by inappropriate means to one conclusion whereas had appropriate means been adopted, he might realistically have reached a conclusion favourable to the applicant. It does not require the court to try the issue so as to determine, based on the outcome, whether substantial injustice had been caused” (London Underground v Citylink [2007] EWHC 1749 (TCC) at para. 47, per Ramsay J. (in the context of a challenge to an award under section 68). 

One possibility is that W would have ignored cross-examination and submissions and kept to his fixed opinion.  If so, that would have been a failure to conduct the proceedings properly which would probably have affected the outcome.  W would still have been removed.  However, taught by the advocates as to how an arbitration is supposed to proceed, he might have listened in good faith to the evidence and applied an open mind.  We shall never know.

This would have played out differently in the United States.  Although institutional rules, such as rule R19(c) of the Commercial Arbitration Rules of the American Arbitration Association, may empower an arbitral institution to remove an arbitrator for bias, the courts will not interfere during the course of the arbitration to remove an arbitrator for lack of impartiality.  The parties have to wait for the award and then apply to vacate it on grounds of “evident partiality" (Federal Arbitration Act, 9 U.S.C.A. § 10(a)). 

This “narrow  standard of review acts as a bulwark against legal ingenuity. Lawyers can easily find one thing or another in almost any proceeding to which they wish to take exception. There are benefits to such legal creativity in many contexts but not in the one before us. Allowing procedural challenges to every award would force parties into court to argue their dispute a second time, incurring the litigation costs and delays they intended to avoid by agreeing to arbitration in the first place.” (UBS Fin. Servs., Inc. v. Padussis, 842 F.3d 336, 339 (4th Cir. 2016), per Wilkinson J). 

England is less concerned, it seems, about delay created by legal ingenuity.


Lay arbitrators are a valuable option for parties litigating in a specialised industry; and it is easier to teach a layman about the basic rules of arbitration than to turn a lawyer into an industry expert.  Evidence of a lack of impartiality removes the arbitrator whether there is a substantial risk of injustice or not, so it is important to be clear about the difference between holding a relevant view about the evidence and holding an irrelevant view affecting a party.

Finally, one wonders what W thinks about this.  Under section 23(5) of the 1996 Act, the arbitrator is entitled to be heard on the application to remove him, but, understandably, W does not seem to have taken the opportunity.  Although the judge decided to anonymise the judgment, everyone in the case must have known who the “biased” arbitrator was and so too, no doubt, people connected with the production.  As a senior member of the industry, W accepted an invitation by the national film agency to use his experience in the resolution of a dispute, albeit for a fee.  Will he ever accept, or recommend a professional colleague to accept, a similar invitation?

It could surely have been avoided.  Someone – perhaps the Chartered Institute of Arbitrators – could produce a short leaflet explaining to lay arbitrators how the process is supposed to unroll.  The basic principles of arbitration are hardly complex.

However, full marks to the BFI for supporting arbitration in the film industry.  It is to be hoped that this experience does not put them off a sequel.  

As a footnote, it is puzzling that everyone agreed that the practice in Sweden was central to the case.  The insurance policy was governed by English law and clearly placed responsibility on the producers for negligently-caused accidents, whether the accident was caused by an agent (e.g., a freelance stunt coordinator) or an employee. One therefore might wonder how central to the arbitration industry practice in Sweden could be.  But those involved in the case no doubt had their reasons…

Friday 22 December 2023

What is a “European work”? A hole in the AVMS Directive and a footnote on arbitration

On 29 November 2023 the European Audiovisual Observatory organised a conference in Brussels, “The promotion of European works according to the AVMSD: where do we stand?” The conference was intended to launch a new Observatory publication, “The Promotion of European Works”.  Video of the proceedings is accessible here.  As a result of the discussion, I noticed something odd about the definition of “European works” in the Audiovisual Media Services Directive (AVMS) (Directive 2010/13/EU). 

The AVMS Directive requires EU Member States to impose a cultural quota on their television broadcasters.  They must include “European works” in a majority of their transmission time.  The European Convention on Transfrontier Television (ECTT), a Council of Europe (CoE) treaty entered into slightly before the passing of the AVMS Directive, makes similar provision.  Many EU Member States are party to the ECTT.  Articles 25 and 26 ECTT provide for conciliation, then arbitration in case of dispute between states party.

Once the departure of the United Kingdom from the European Union became imminent, the question whether UK works should continue to be regarded as European for the purposes of the EU’s television quota under the AVMS Directive became a persistent, if sotto voce, theme in European audiovisual policy discussions.  As Observatory senior legal analyst, Francisco Cabrera observed while moderating the panel discussion (at 42.57), the “elephant in the room” was the question whether UK works should continue to be “European works”.

In the negotiation of the terms of the British exit from the EU, the EU’s position was always that “Audiovisual services should be excluded from the scope of the economic partnership”.  This position was conceded by the UK, so that the field remained open post-Brexit for attempts to exclude British producers from the quota.  However, for various reasons of international law explored in an article that I published last year, this would be very difficult to achieve.  The main obstacle is that many EU Member States are bound by the ECTT to treat all European works as “European works” for the purposes of the cultural quota (except as between EU Member States, for which EU law takes priority over the ECTT).

What is a “European work”?

There is a discontinuity between the AVMS definition of “European works” and the ECTT conception.  Simplifying slightly, under the Directive European works are (i) works originating in EU Member States; (ii) works originating in European states which are not EU Member States but are party to the ECTT; and (iii) works produced under co-production agreements between the EU and non-EU Member States (Article 1(1)(n), AVMS).  It will be noted that under this definition (leaving aside co-productions), programming from non-EU countries will not count towards the quota unless those states are both (i) European and (ii) party to the ECTT.  In other words, works from European countries which are not party to the ECTT or the EU are excluded from the benefit of the quota under the Directive.

The ECTT formulates the cultural quota more broadly as “creative works, the production or co-production of which is controlled by European natural or legal persons” (Article 2(e), ECTT).  The ECTT means by “European” simply European.  The designation must at least extend to all the Members of the CoE and also covers parties to CoE treaties, such as the European Cultural Convention (ECC), even if, like Belarus or Kazakhstan, they have never been members of the CoE.  This is clear from the fact that proposed amendments to the ECTT must be circulated to, among others, non-CoE members who are party to the ECC (Article 23(2), ECTT).  In short, “European” here simply refers to countries contained by or extending to the European continent.  It follows that programming from the Netherlands is (obviously) European for the purposes of the ECTT, even though the Netherlands is not party to the ECTT.  So to is programming from Norway or Turkey.

The ECTT is mainly concerned with the freedom to broadcast television across borders. According to its recitals, it is intended “to increase the production and circulation of high-quality programmes” with a view to “enhancing Europe's heritage”.  It represents a broad notion of European culture, like the European Cultural Convention, to which it refers, and clearly has the object of ensuring that European works in a broad sense can travel across borders.

Under the AVMS, however, works which come from non-EU, non-ECTT European countries such as Armenia, Monaco or Russia are not “European works”.  Of course, these non-parties cannot enforce the ECTT, but it is a different question whether their works are European.  The states party to the ECTT owe their duty in international law to one another, but that is not to say that a failure by one ECTT party to implement the cultural quota in respect of European works with a non-ECTT origin could not give rise to a breach of the treaty with respect to another ECTT party. 

What does this mean for EU Member States party to the ECTT? 

The many EU Member States party to the ECTT are in an awkward position.  Although as between themselves the AVMS Directive governs, in their relations with non-EU ECTT states they must implement the cultural quota as laid down in the ECTT.  To be precise, their obligation is to “ensure, where practicable and by appropriate means, that a broadcaster within its jurisdiction reserves for European works a majority proportion of its transmission time” (Article 10(1), ECTT). 

Interpreted against that background, are EU Member States party to the ECTT free to implement the cultural quota by limiting themselves to EU works, to the exclusion of some or all ECTT members?  Are they free, in other words, to determine what is or is not a European country for this purpose, notwithstanding their obligations under the ECTT?

A fundamental rule of international law is that “[a] treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose” and that every treaty “must be performed… in good faith” (Articles 31 and 26, Vienna Convention on the Law of Treaties 1969 (VCLT)).  Just as between signing and ratification a state "is obliged to refrain from acts which would defeat the object and purpose of a treaty" (Article 18, VCLT), the object and purpose of the treaty must shape the obligation to implement the treaty, once ratified, in good faith.  In the case of the ECTT, the treaty's object and purpose is to foster the circulation of European works, broadly defined, across borders.

When an ECTT state undertakes to apply the cultural quota, that obligation is undertaken with regard to all the other parties to the treaty.  It would seem that it would not be compliant with the ECTT for a European state to impose a requirement on its broadcasters of transmitting solely national programming, even though in a simplistic sense a majority - indeed, all - of the works broadcast would have their origin in a European state.  By the same token, the limitation of the quota by EU Member States to works originating in EU (or even ECTT) states would not implement the ECTT’s quota for European works (in the broadest sense).  The ECTT requires application of a European quota as such, not a cherry-picking of favoured European states of origin.

It follows that the exclusion of UK works from the television quota under a revised AVMS Directive would compel EU states party to the ECTT to violate the cultural quota of the ECTT.  Any non-EU ECTT state affected would be entitled to pursue a claim in arbitration against the EU Member State concerned.  It is worth noting that arbitration under Article 26 of the ECTT is mandatory once six months have expired from a first request for conciliation.

A final point on arbitration

Under bilateral investment treaties (BITs), various EU Member States have entered into obligations in international law to protect the investments of UK (and often US) companies.  These “investments” invariably are defined to include intellectual property, so any production made in an EU state, such as the recent Poor Things, shot in Hungary in 2021, could qualify.  Such treaties may contain terms which would be violated by differential treatment of EU and non-EU European works in breach of the ECTT. For example, in the UK-Hungary BIT each state party covenanted to “observe any obligation it may have entered into with regard to investors of the other Contracting Party” (Article 2(2), UK-Hungary BIT). It could be argued that the ECTT cultural quota is such an obligation.  The distinctive element in these BITs is that the parties irrevocably consent to arbitration at the request of the investor, who can then bring a claim for damages. 

Even though many of the UK-EU BITs have been terminated in the wake of the CJEU’s debated 2018 decision in Slovak Republic v Achmea (the UK-Hungary BIT was terminated in August 2022), BITs usually have long sunset clauses, typically of 20 years, protecting investments made during the term of the BIT.  No doubt it is an unlikely scenario, but if a distributor found itself losing television sales as a consequence of the exclusion of UK pictures from the quota, a claim in arbitration against the ECTT state concerned would be conceivable. 

All that having been said (and at such regrettable length), UK quality television is so strong in European markets, it may be that the best efforts of policy-makers will not make a lot of difference…


Saturday 9 December 2023

Pre-appointment Disclosure - a debatable amendment to the UK's Arbitration Act 1996

On 6 September 2023, the Law Commission for England and Wales published its Review of the Arbitration Act 1996: Final report and Bill. The Commission notes "the consensus that the Act works well, and that root and branch reform is not needed or wanted", but nonetheless proposes amendments. One would insert a new section 23A on "Impartiality: Duty of Disclosure". This amendment tries to address the position of potential arbitrators, creating a legal duty of disclosure on them before they have been appointed as arbitrators. This approach seems problematic as a matter of principle and questionable in its proposed implementation.


Section 33(1) of the United Kingdom's Arbitration Act 1996 provides that an arbitrator "shall... act fairly and impartially as between the parties." In Halliburton Company v Chubb Bermuda Insurance Ltd [2020] UKSC 48, the UK Supreme Court held that this implied a duty on the arbitrator to disclose to the parties any circumstances that might reasonably give rise to a conclusion by an objective observer that there was a real possibility of bias on the arbitrator's part in relation to the dispute (per Lord Hodge, para. 153). A party to an arbitration may apply to the court for the removal of an arbitrator where "circumstances exist that give rise to justifiable doubts as to his impartiality" (s. 24(1)). According to Halliburton, this is an objective test (paras 52-55).

The new section 23A

The Law Commission's new section would read as follows:

23A Impartiality: duty of disclosure

(1) An individual who has been approached in connection with their possible appointment as an arbitrator must, as soon as reasonably practical, disclose to the person exercising the power of appointment any relevant circumstances of which the individual is, or becomes, aware.

(2) An arbitrator must, as soon as reasonably practical, disclose to the parties to the arbitral proceedings any relevant circumstances of which the arbitrator is, or becomes, aware.

(3) For the purposes of this section— 
(a) “relevant circumstances”, in relation to an individual, are circumstances that might reasonably give rise to justifiable doubts as to the individual’s impartiality in relation to the proceedings, or potential proceedings, concerned, and
(b) an individual is to be treated as being aware of circumstances of which the individual ought reasonably to be aware.

This would be a "mandatory provision" within s. 4(1) of the 1996 Act. It would "have effect notwithstanding any agreement to the contrary". Parties would hence be unable effectively to modify the requirement, whether of their own volition or at the request of any potential arbitrator.

The new s. 23A(1) takes its inspiration from Article 12(1) of the UNCITRAL Model Law on International Commercial Arbitration, which provides:

(1) when a person is approached in connection with his possible appointment as an arbitrator, he shall disclose any circumstances likely to give rise to justifiable doubts as to his impartiality or independence. An arbitrator, from the time of his appointment and throughout the arbitral proceedings, shall without delay disclose any such circumstances to the parties unless they have already been informed of them by him.

However, the Model Law does not in terms impose any "duty" on the potential arbitrator.

The Arbitration (Scotland) Act 2010, which is based on the Model Law, addresses the same issue in Schedule 1, rule 8:

8 (1) This rule applies to—

(a) arbitrators, and

(b) individuals who have been asked to be an arbitrator but who have not yet been appointed.

(2) An individual to whom this rule applies must, without delay disclose—

(a) to the parties, and

(b) in the case of an individual not yet appointed as an arbitrator, to any arbitral appointments referee, other third party or court considering whether to appoint the individual as an arbitrator,

any circumstances known to the individual (or which become known to the individual before the arbitration ends) which might reasonably be considered relevant when considering whether the individual is impartial and independent. [Emphases added.]

Scope of persons bound

The Scottish Act says that the candidate arbitrator "must" (i.e., shall) make the relevant disclosure. It is not clear whether this "must" is anything more than directory. In this context one might recall Lady Arden's concurring opinion in Halliburton, in which she said in relation to the "duty of disclosure" of an arbitrator:

"There is scope for debate as to whether it is a duty at all in the strict sense. The duty [to disclose related arbitral appointments] only arises if the arbitrator wants to take a further appointment in a different arbitration." (para. 161)

It will be noted that whereas the Scottish Act refers to "individuals who have been asked to be an arbitrator", the Law Commission's amendment imposes its "duty" on any "individual who has been approached in connection with their possible appointment as an arbitrator".

The idea is hence that when someone is contacted with a view to possible appointment as an arbitrator, that person comes under a legal duty to disclose potentially sensitive details about his prior dealings and, if he is a partner, for example, those of his law firm. The enquirer is under no obligation to appoint the potential arbitrator from whom it has received the requested disclosure. In contrast with the Scottish Act, the enquirer does not even have to have asked the candidate arbitrator to accept the appointment.

Modification of the duty?

The candidate arbitrator will seek in vain to negotiate the scope of his disclosure, by reason of the status of the new section as "mandatory" under s. 4 of the 1996 Act. Section 23A will "have effect notwithstanding any agreement to the contrary". The parties to the arbitration agreement cannot effectively limit the required disclosure by the candidate arbitrator, even though Lord Hodge said in Halliburton that under the present Act the parties could do so (paras 136-7), no doubt because the duty of disclosure of disclosure of an appointed arbitrator arises simply as an offshoot of the general obligation of fairness under s. 33 of the 1996 Act. This might affect partners in large, international firms for whom the task of checking firm conflicts is burdensome or emergency arbitrators who do not have time to carry out a full conflicts check.

Functional relevance

Is the s. 23A duty a legal duty "in the strict sense"? The general duties of arbitrators in English-seated arbitrations post-appointment, whether deriving from contract or status, can be enforced against them by injunction (see, e.g., Compagnie EuropĂ©ene de Cereals SA v Tradax Export SA [1986] 2 Lloyd's Rep. 301; Cole v Silvermills Estates and Land Ltd [2012] S.C. 1), even if s. 29 gives them immunity from liability for acts done in good faith; and, subject to that immunity, by claims for damages. Some jurisdictions provide less protection against suit than England (see: Born, Gary B., International Commercial Arbitration, Ed. 3, Kluwer Law International, §13.05[A]).  

Although it is difficult to imagine a disappointed party suing a candidate arbitrator for failing to disclose a conflict, it is not impossible.  If an arbitration were delayed for several months by reason of the late disclosure of a conflict by a candidate arbitrator who never in fact was appointed, a trouble-making party might bring a claim.  But the difficulty of identifying a case in which a legal duty "in the strict sense" would have to be enforced encourages doubt as to whether such a duty is actually useful.

If the pre-appointment obligation introduced by s. 23A is not to be enforceable as a strict duty, it is hard to see how it would be enforced. Notably, non-compliance with the s. 23A duty has not been added as a ground for removal of an arbitrator under the 1996 Act.

This brings us back to the supposed function of the new provision. In order to remove an arbitrator for bias it is still necessary to show "that circumstances exist that give rise to justifiable doubts as to his impartiality" (s. 24(1)(a), 1996 Act). To set aside an award on grounds of serious irregularity, it is still necessary to show "failure by the tribunal to comply with section 33 (general duty of tribunal)" (s. 68(2)(a), 1996 Act). Aside from the marginal point that the mere failure to disclose relevant information may in itself support an allegation of partiality, the new section provides no functional link between the hypothetical duty of a potential arbitrator under s. 23A and the objective of preserving the impartiality of the arbitrator once appointed. So the creation of the duty would seem essentially irrelevant. If the objective is to obtain conflict information prior to appointment, why can parties not just ask for it? And add a warranty to the arbitrator's contract, if it concerns them, in the event the party proceeds to appointment?

If one insists on an explicit and effective connection between non-disclosure and the actual terms of the 1996 Act, it could be achieved without creating a duty. One could simply include a presumption that an arbitrator's failure to disclose a relevant fact on being requested to serve or during his appointment will give rise to a presumption that "circumstances exist that give rise to justifiable doubts as to his impartiality" (s. 24) and that there has been "failure by the tribunal to comply with section 33 (general duty of tribunal)" (s. 68). Such a solution was adopted in s. 12(e) of the Revised Uniform Arbitration Act of the United States National Conference of Commissioners on Uniform State Laws (2000), which provides:

An arbitrator appointed as a neutral arbitrator who does not disclose a known, direct, and material interest in the outcome of the arbitration proceeding or a known, existing, and substantial relationship with a party is presumed to act with evident partiality under Section 23(a)(2) [court's power to vacate award].

However, as will be evident from the above, this does not seem to me to be a very pressing reform.


As Lady Arden observed in Halliburton, "the conclusion that as a matter of the law of England and Wales an arbitrator is to be treated as aware of a conflict of interest of which he is not actually aware would on the face of it take English and Wales beyond Scots law, which appears to require actual awareness" (para. 162). Despite citing this observation, the Law Commission comes down in favour of the test of constructive knowledge, driving a wedge between English and Scottish law. Given that the issue is the preservation of impartiality, one might wonder how an arbitrator could be influenced by facts about which he is ignorant... But that is a debate for another day.

Thursday 7 December 2023

Back to London



After nine and a half years at the World Intellectual Property Organization in Geneva, I have come back to London to practise as a lawyer, arbitrator and mediator.  I had lived in Brussels for several years when I worked at the Motion Picture Association, so I was accustomed to a French-speaking environment.  Although I had made many visits to WIPO over the years, it was a new experience, and an educational one, to work in an intergovernmental organization in the United Nations system.

The defects of bureaucracies are well-known, but of all the UN specialised agencies, WIPO seems to be among the most useful.  As a result of some of the international treaties which it administers, WIPO streamlines the registration of patents, trade marks, designs and geographical indications.  It runs other niche programs such as WIPO ALERT, which was my own project to improve information sharing about copyright piracy with the advertising sector.  Now in the hands of my talented former colleagues at WIPO, the concept may grow to encompass cooperation with other actors in the online environment which are in a position to help secure the rule of law in cyberspace.  So much can be achieved by voluntary cooperation by good-faith actors in the online sphere.

No longer subject to the requirement of neutrality applicable to an international civil servant, I plan to make the occasional foray into the public sphere with comments on legal and policy issues which catch my attention.

Saturday 14 September 2013

Microsoft Xbox One: Regulatory puzzles in a converging marketplace

On Thursday, 12 September 2013, the Royal Television Society's Cambridge Conference heard Nancy Tellem, Microsoft's Entertainment & Digital Media President, explain the thinking behind the Microsoft Xbox One console, due for release on 22 November 2013. Xbox One is a "state of the art gaming console, a new generation TV and movie system, and a whole lot more". Gaming, internet browsing, Skype telephony and watching films and television all take place through a single set-top box. Users can rapidly switch between uses or enjoy them simultaneously, using vocal or gestural controls. 

When I asked Nancy Tellem, who started her career as an attorney, where she expected Xbox One to be regulated, and whether as television or video game or both, she engagingly confessed that she "would have to get back to me". 

As the European Commission observed in its April 2013 Green Paper on convergence in audiovisual services, we are experiencing an "on-going transformation of the audiovisual media landscape, characterised by a steady increase in the convergence of media services and the way in which these services are consumed and delivered". Different set-top boxes and interfaces are vying to become the dominant gatekeeper to the digital household. Microsoft's new product seeks to do so by offering an immersive, highly interactive experience, breaking down the border between gaming and filmed entertainment. Microsoft proposes a truly interactive form of television. Understandably, they are beginning with a live action version of a popular game, Halo, created in partnership with Steven Spielberg.

Now, it is not clear precisely how interactivity will be integrated into these live action dramas, but it is obvious that elements of gaming will be part of the user experience. In the UK, the sale of computer games is regulated by the Video Recordings Act 1984 (as amended in 1994 and re-enacted in 2010). However, this does not cover streamed content, as there has to be a supply of a disc or other storage device before the Act bites. Similarly, the Video Recording (Labelling) Regulations 2012 apply only to physical carriers. While PEGI Online, a voluntary scheme for rating online games, is made available by the Interactive Software Federation of Europe, this important category of gaming is essentially unregulated (save by the law of obscenity). Xbox One will make the regulatory gap more troubling to policy-makers and parents alike.

However, the Audiovisual Media Services Directive 2010/13/EU, as implemented by Regulations in 2009 as part 4A of the Communications Act 2003 (further amended in 2010), envisages that EU Member States will regulate on-demand services which are "television-like". As the Xbox One service will involve "the provision of programmes the form and content of which are comparable to the form and content of programmes normally included in television programme services" (section 368(1)(a), 2003 Act), it seems likely (subject to the crucial jurisdictional question) that Microsoft will be subject to regulation by the Authority for Television on Demand and, in relation to advertising associated with its on-demand services, the Advertising Standards Association. Among the many requirements of the 2003 Act is this: "If an on-demand programme service contains material which might seriously impair the physical, mental or moral development of persons under the age of eighteen, the material must be made available in a manner which secures that such persons will not normally see or hear it" (s. 268E(2)). The Directive requires Member States not only to prevent advertising for tobacco products, control product placement, discourage discrimination and so forth, but also to require such services to promote "the production of and access to European works".

So the question is: in what circumstances would the UK have jurisdiction over Microsoft's Xbox One service? To (over)simplify a complex set of provisions, an EU Member State will regulate where the operator is substantially established in that state (either by having its head office there or, where the head office is in another Member State, maintaining a significant part of its relevant workforce there) or uplinks to a satellite from or uses satellite capacity appertaining to that Member State (unlikely to be relevant to Xbox One). 

Although Microsoft has staff working on this service in the UK, they might choose to deliver the service wholly from the US. Would that put them in the clear? Largely. However, the Xbox One service seems to be a "television licensable content service" within section 232 of the 2003 Act. Under section 329 of the 2003 Act, Ofcom can ask the Secretary of State to proscribe a foreign service that is "unacceptable" on grounds that it repeatedly contains matter which offends against taste or decency, or is likely to encourage or to incite the commission of crime, lead to disorder, or be offensive to public feeling. If an order is made, the service may not be included in a cable package - which might be a problem for Microsoft in sub-distributing their service, if that was what they hoped to do. So in practice in the event of concern on Ofcom's part as to the content of the Xbox one service, Microsoft would in fact find itself subject to regulatory pressure.

Be that as it may, the reality is that regulation of streamed services must surely come, even if they emanate from the US. As Professor Joel Reidenberg has argued for some years, it is clear that cyberspace will eventually be subject to regulation, even (or especially) in democratic countries. Recently the British Prime Minister, David Cameron, pushed ISPs to provide better protection for children. As technology and commercial innovation break down the borders between traditional product categories, the process of assimilation of the online to the offline world will accelerate. Xbox One seems to be a new challenge to those product definitions, and if it succeeds, regulatory change must be likely.

Saturday 6 July 2013

Ireland: Data Protection Commissioner loses in Supreme Court over GR

On 3 July 2013 the Irish Supreme Court gave its judgment in the appeal by the Irish Data Protection Commissioner ("DPC") in EMI Records (Ireland) Limited & ors v The Data Protection Commissioner [2013] IESC 34. Although this represents an endorsement of the voluntary graduated response scheme agreed between ISP Eircom and the record labels, the decision was based essentially on technical grounds. It offers an interesting example, however, of the attitude of an authority charged with enforcing data protection laws.

As described in an earlier post, EMI and other record labels had sued Eircom for participating in the infringement of copyright by its Internet access subscribers. The case was settled by a contractual GR scheme, under which an infringing user would on his third notification be suspended from Internet access for one week; after a fourth notification, Eircom would terminate his access agreement. The user was free to find another ISP if he could.

The DPC had taken the position that the conventional process of gathering anonymous data and its transmission and processing by Eircom was in some way an infringement of the rights of internet users under data protection law. The parties to the original action took the matter back to court to get a ruling on the issue, but the DPC refused to take part in it. He had asked the parties to pay his costs, win or lose, or at least not claim costs against him, which the parties had declined to agree. Nonetheless, Charleton J gave a judgment on the issue, robustly holding that there was no valid data protection objection to the GR scheme. 

Nothing daunted, the DPC issued an enforcement notice against Eircom under the Data Protection Acts 1988-2003, ordering it to stop operating the scheme. Eircom sought to appeal using the statutory procedure and the labels sought to be joined in that appeal. In an impressive display of fairmindedness, the DPC opposed their joinder, demanding in any event that the labels agree that they would receive no costs if allowed to take part. The labels, who had no automatic right to participate in the appeal, applied for judicial review of the enforcement notice, alleging that the DPC was wrong in law and, in any case, had failed to state any reasons in his notice (a requirement under the Data Protection Acts).

On 27 June 2012 Charleton J ruled on the judicial review application ([2012] IEHC 264), holding that the enforcement notice was bad in law, confirming his earlier analysis that peer-to-peer enforcement involved no breach of privacy, and held that the notice was bad in any event for lack of reasons.

The Supreme Court has now confirmed his decision, affirming that the lack of reasons was fatal. Given this procedural point, however, the court does not reach the substance of data protection law, beyond saying, en passant: 

"it appears to be accepted that the method by which the Protocol works is that all Eircom does is to receive a series of IP addresses from the record companies, write the appropriate letter to the customer corresponding with that IP address, and invoke the suspension or termination provisions of the Protocol as appropriate. On that basis it is not inherently obvious as to why such activity necessarily involves a breach of data protection law." 

No doubt the DPC will return to the fray in due course. 

What Is Arbitrator Bias? (or: No Good Deed Goes Unpunished)

In February 2024 the High Court in London handed down a decision to remove an arbitrator in a film industry dispute, which, in addition to i...