The facts in Getma v. Guinea case seem familiar enough, but the facts leading to annulment of the award involve a wholly unexpected plot twist—a showdown between an African arbitral institution and the arbitral tribunal over the tribunal’s fees.

When the annulment decision in Getma v. Guinea first came out, it received considerable attention, including in this blog and two letters authored by the tribunal. Most reports were critical of the Cour Commune de Justice et d’Arbitrage (“CCJA”) of l’Organisation pour l’Harmonisation en Afrique du Droit des Affaires (“OHADA”), and raised questions about the legitimacy of its annulment decision.

Those reports, however, did not examine the extent to which the tribunal’s conduct was contrary to CCJA rules and express rulings. Taking this tension as a starting point, the thesis of this post is that, even faced with a terrible, horrible, no-good, very-bad decision on an issue that is within an arbitral institution’s exclusive competence and discretion, a tribunal has only two choices: comply or resign.

Instead, in Getma v. Guinea, the tribunal substituted its own preferences for the institution’s unambiguous rulings despite the institution’s forewarning of the possibility of annulment.

The CCJA and Its Powers

For those who are unfamiliar, the CCJA operates as a traditional arbitral institution (drafting arbitral rules and administering arbitrations), as a multi-national court that is the final expositor of OHADA law, and as the authority competent to annul arbitral awards rendered pursuant to CCJA rules.

Among its powers, the CCJA sets arbitrator fees according to a fee schedule. In a 1999 decision by the CCJA, it affirmed that arbitrators’ fees and expenses are set exclusively by the Court, in accordance with the provisions of the Rules of Arbitration, and that “Any separate arrangement between the parties and the arbitrators concerning their fees is null and void.”

Low Fees and Efforts to Augment

Back in our case between Getma and Guinea, in accordance with its published fee schedule, the CCJA set the arbitrator fees exceptionally low, some might even say breathtakingly low—just €60,000 for an entire tribunal in a case with approximately $50,000,000 in dispute and in which the tribunal reportedly spent a combined total of nearly 1000 hours.

During the appointment process or soon thereafter, the arbitrators apparently realized that the €60,000 amount in the CCJA fee schedule was inordinately low and the President had several communications with the case manager at the CCJA to that effect. The exact nature of these communications is not entirely clear from the record that was available in drafting this post. Among other reassurances in these communications, however, the CCJA eventually consented to the tribunal raising the issue of fees with the parties.

To that end, nearly two years after commencement of the case and shortly before the first evidentiary hearing, in April 2013 the tribunal proposed to the parties that €450,000 was a more reasonable fee for the arbitrators, and it requested that the parties both agree to the increased fees.

When the parties’ formal assent was not forthcoming, the President of the tribunal commenced the May 2013 evidentiary hearing with a request for specific assent by the parties. In making this request, the President stated specifically, “Of course, if the Parties were not to agree, the Tribunal would be forced to reconsider its involvement in this matter[.]” In that statement, however, the President also asserted that party assent was required for “OHADA to make a fully informed decision” and to “enable OHADA to make its decision.” Under its rules, the CCJA retains the power to adjust fees from its schedule “where the circumstances of the case make it exceptionally necessary.”

Getma agreed to the tribunal’s proposal almost immediately. Guinea, apparently with some reluctance, also agreed to the increase, though later stepped away from that agreement.

Notwithstanding the parties’ apparent agreement, and repeated efforts by the tribunal, the CCJA eventually decided in writing on August 2013, and confirmed again on October of 2013, that the tribunal’s fees would remain €60,000. The last confirmation came after Getma, at the behest of the tribunal, directly lobbied the CCJA to increase the fees in September 2013.

The Award and Fee Augmentation

Fast forward several months. Notwithstanding the CCJA’s decision, and another confirmation of that decision by the CCJA on April 14, 2014, the tribunal proceeded to hear the dispute. The arbitral proceedings were completed and tribunal signed the award on April 29, 2014.

The day after signing the award, the tribunal President wrote the parties indicating that the award had been signed. In the letter, however, he stated that “for the sake of good order, the arbitrators – who … are to be paid by the Parties – should receive their emoluments [totaling €450,000] before the award is sent.” The letter also stated that “If one of the Parties pays more than its share, the award will reserve that Party’s right to claim the surplus from the other Party.”

The parties did not immediately pay the €450,000, but counsel for Getma wrote to the CCJA on May 14 indicating its intention to pay the additional fees.

On May 19, the CCJA wrote to the tribunal confirming that the tribunal’s requests for fee augmentation had been “rejected by the Court, by means of several administrative decisions of which [the tribunal was] properly notified.” In this letter, the Court cited the 1999 decision referenced at the beginning of this post and concluded that “Any separate arrangement between the parties and the arbitrators concerning their fees is null and void.”

The letter went on to explain:

The CCJA system of arbitration categorically prohibits any arrangement between the parties and the arbitrators relating to fees, the determination of which falls within the discretionary power of the Court. The arbitrators may not ignore these mandatory rules of CCJA-OHADA arbitration.
As a result, you are formally prohibited from seeking payment of fees directly from the parties, who have already paid in full the amounts due.

The CCJA also wrote a response to Getma’s May 14 letter, reiterating much of the content of its May 19 letter to the tribunal and warning that that if the final award “includes the payment of the amount of €450,000 to the arbitrators, … the award will potentially be subject to invalidation by our the [sic] community’s high court.”

On May 22, 2014, over 3 weeks after signing the award, the Tribunal transmitted the Award directly to the parties. We do not know why the tribunal delayed transmitting the award, but it might be reasonable to assume, in light of its April 30 letter, that the tribunal was waiting for payment of the €450,000, either from one or both parties. As previewed in the tribunal’s April 30 letter, the award explicitly provided for a “right to claim the surplus from the other Party” of any unpaid fees to the tribunal.

In August 2014, Getma paid to the tribunal €225,000; Guinea did not. The tribunal wrote to Guinea and threatened legal action if it did not remit €225,000. Ultimately, the tribunal did not sue Guinea, but it apparently did sue Getma for Guinea’s share of the unpaid fees.

Post-Award Proceedings

Meanwhile, back at the CCJA, Guinea challenged the award. At a public session on November 19, 2015, the CCJA announced its unanimous decision to annul the award. In a later published explanation of the ruling, the Court justified its decision primarily because of the tribunal’s direct contravention of CCJA rules and rulings regarding its fees.

Enforcement proceedings on the annulled award are currently pending in Washington DC. Getma urges the US court to reject the CCJA’s annulment decision and enforce the award because, it argues, the CCJA’s decision on fees was “unfair and biased” and because the parties’ agreement to pay additional fees should be honored as party autonomy.

Notably, among the exhibits in support of Getma’s papers are two letters written by the tribunal, one an “open letter” to the arbitration community and another to Jeuene Afrique, a French publication that focuses on Africa. In the letters, the tribunal criticized the annulment decision as a “legal heresy,” but based its analysis, in part, on somewhat puzzling assertions, like “arbitrators’ compensation has nothing to do with their task” and “[t]he role of an international arbitration institution is to protect the arbitrators that work under its aegis[.]”

Arbitrators and Institutions

While the tale of Getma and Guinea raises many interesting issues, the focus of this post is on arbitrators’ duties when they disagree with arbitral institutions’ rules and rulings.

Arbitral institutions carefully calibrate the balance in their rules between party autonomy and mandatory provisions. Increasingly, institutions also serve as gatekeepers of arbitrator conduct. For example, institutions have increased arbitrator disclosure obligations, police arbitrator caseloads, monitor the time for rendering awards, and even limit arbitrator fees in light of sub-standard arbitrator conduct.

Some such institutional decisions have drawn strong criticisms from arbitrators and been reconsidered as a result.

In all this back-and-forth, however, in no other known case has a tribunal directly rejected an arbitral institution’s express rulings. Indeed, one of the most fundamental features of arbitrators’ mandate is an obligation to apply the arbitral rules agreed to by the parties, including those rules that allocate decisionmaking authority to the institution. If serious disagreement arises, arbitrators may avoid potential liability for resigning if done in a timely manner and on a well-founded basis. Substituting their personal preferences for the institution’s decision, however, is simply not an option.

Moreover, arbitrator criticisms of arbitral institutions may perfectly valid, but those critiques are ordinarily proper only outside the context of individual, pending cases (unless sought through procedures in official enforcement proceedings that some jurisdictions permit in exceptional circumstances). The obvious reason is that arbitrators should not been seen as attempting to influence outcomes in cases in which they presided, especially not when the enforcement proceedings involve not only questions about the tribunal’s conduct, but its conduct in obtaining its own compensation.


The international arbitration community has just finished the ICCA Congress in Mauritius, where the future of arbitration in Africa was discussed, including the need to build strong, legitimate African arbitral institutions that can attract parties and arbitrators. In this aim, Africa faces both real obstacles and unfair perceptions. Some of the issues in the Getma v. Guinea case, including some not discussed in this post, underscore those challenges ahead.

But one challenge that no African arbitral institution should face is arbitrators who deliberately reject that institution’s mandatory rules and rulings, and then seek to blame the institution for the post-award consequences of their own actions.

DISCLAIMER: The opinions expressed are solely those of the author. This post was drafted using documents, and certified translations of those documents, filed by both parties in the annulment proceedings in Washington DC.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration
This page as PDF


  1. Arguably, as the Institution’s Rules are usually applicable only because the Parties have so agreed, the parties are entitled to agree to vary their one agreement. If the Institution than decides to wash its hands of the arbitration, so be it.

    The Parties – and maybe the Arbitrators – may be in breach of any agreement with the Institution, but it is hard to see why that should invalidate the substantive decision of the Arbitrators. Annulment in the circumstances seems, on the face of it, to have been the result of a fit of pique, never the most judicious basis for a purported ruling.

    It is not altogether clear whether CCJA is a public court or a private institution but there seems to be a worldwide trend for Arbitral Institutions to seek to arrogate to themselves power akin to that of a sovereign Court.

    To this layman, that appears contrary to the purpose of private arbitration, created as an alternative to a Court of Law.

  2. I disagree strongly with Layman, and even more strongly with the sentiments expressed by a tribunal that fails to recognize how parties would view their conduct as an abuse of their authority over them. If an arbitrator is unhappy with the fees assessed by the institution, the remedy is don’t accept the case or don’t take further appointments from the institution. But to accept an appointment and then use your authority extract more fees from the parties than the institution assessed is unthinkable. That you would then attempt to publicly shame the institution for enforcing its own rules (especially while there are challenge proceedings pending) is disgraceful.

    If arbitrators could bypass an institution’s authority whenever they are unhappy or any other aspect of the applicable rules, then why bother having institutions at all? What happens to the predictability of arbitration clauses included in parties’ contracts? The rules or just sort of “guideposts” that arbitrators follow only when it suits them?

    I recommend reading the actual “Open Letter”. It is a quintessential “it’s all about me” from an arbitrator’s point of view. It will be a good discussion starter at future conferences where the topic is risks of rogue tribunals and their abuse of authority.

  3. I would say to Mr. McIlwrath only that some Arbitration laws (see, for example, England and Wales Arbitration Act 1996 s.28(2)) provide for excessive fees to be reviewed by the Court of the Seat. In the instant case, the step of annulment without relevant reason seemed inappropriate.
    The operative words of the report – in a paragraph of their own – are:
    “Getma agreed to the tribunal’s proposal almost immediately. Guinea, apparently with some reluctance, also agreed to the increase, though later stepped away from that agreement.”
    For most of us, an agreement is an agreement. Pacta sunt servanda -nothing about stepping away.
    I didn’t think an administrative issue – however serious – should interfere with the substance of an Award (unless, of course in made the process unfair – but that’s another matter.)

  4. By the way, Professor Rogers says: “Taking this tension as a starting point, the thesis of this post is that, even faced with a terrible, horrible, no-good, very-bad decision on an issue that is within an arbitral institution’s exclusive competence and discretion, a tribunal has only two choices: comply or resign.”
    May I suggest that the Parties have a third option – to take their arbitration outwith the control of the Institution and to ask the Arbitrators to continue ad hoc. A new agreement may be necessary but that is a way of dealing with an over-mighty institution. Perhaps Mr. Mcilwrath would like to include ‘rogue’ institutions on his agenda.

  5. Layman, if you were talking about mediation, then maybe I could follow you, because mediation is ultimately non-binding so there is no potentially fatal consequence for a party who refuses a mediator’s demand for some extra cash.

    But this is arbitration where the tribunal, once appointed, holds judicial authority over the parties, subject to only limited future review by a court. So when you say “an agreement is an agreement,” what I hear is a failure to appreciate how parties always feel put under pressure (the legal concept is duress) if a tribunal uses its authority to extract additional compensation from them. Similarly, how should a party react when the chair announces at a procedural hearing, “we would like the parties to authorize the tribunal to take as long as it wants to deliver an award” or “we would like you to approve and pay this young lawyer happens to be sitting next to us here today” (both real examples I have faced multiple times). Consider your answer carefully, because if you disagree, the tribunal isn’t going away.

    Avoiding the risk of arrogance, abuse, and even extortionate behavior from tribunals is one of the reasons that parties choose institutional arbitration over ad hoc. If the only protection from arbitrators running amuck is to have recourse to the courts, then why not just avoid arbitration and use the courts in the first instance?

  6. Thanks for your thoughtful comments Mr. Hartwell.

    Just to clarify, when said a “terrible-horrible, no-good, very bad” decision by an institution, I was not suggesting specifically that the fee decision in this case necessarily fit that description. Certainly the arbitrators thought it did, and others might agree, but there is also room for disagreement on that point. Institutions have complex reasons why they set fees. My characterization in the post was meant to highlight that it is not acceptable for arbitrators to purport to overrule those reasons, particularly when the arbitrators have an unvarnished self-interest in that decision.

    I would agree, however, that if–instead of their own compensation–arbitrators had a serious, but non-self-interested, concern about the effect of an institution’s rulings, suggesting that the parties change their agreement to submit to an ad hoc arbitration might be a legally legitimate way to resolve the tension between arbitral rules and arbitrators’ mandate and protect the award.

    Michael’s comments also capture an important aspect of this case–the timing of efforts to extract an agreement from the parties raises real questions about the voluntariness of their apparent agreement and, relatedly, the parties’ incentives. It appears from the facts that certain important events (including lobbying of the CCJA for a fee increase) occurred while the tribunal was deliberating.

    In this case, timing seems to make worse an already bad set of facts.


  7. I’m not really satisfied with ‘comply or resign’ as the moral lesson of this controversy. There’s a rule which permits the Court to fix the arbitrators’ costs at an amount more or less than the normal one in ‘exceptionally necessary’ cases (R24.3). It’s a no-recourse administrative decision for which ‘no reasons shall be given’ (R1.1). If the circumstances were exceptionally necessary in this case, the arbitrators had a right to exceptionally necessary rather than normal fees. *If the CCJA’s administrative decision was wrong*, the arbitrators’ rights were infringed and they were placed in an unreasonable situation, not one in which ‘comply or resign’ were better options than protest.

  8. While I am very much a supporter of both party autonomy and arbitrators’ authority in general, and not a fan of ever increasing institutional powers restricting them, in this case the tribunal’s right to ask for anything and the parties’ decision to pay more than the institute ordered comes down to a simple hierarchy of the relevant sources of norms: if the applicable rules are mandatory, then party autonomy can only take parties as far as choosing whether to arbitrate under those rules or not; similarly, arbitrators can choose to either take the appointment or not, but cannot amend mandatory rules not left to their discretion.

    What puzzles me in this case, however, is that the annulment seems to ultimately punish the parties for the arbitrators’ excess – and the case manager’s early miscommunication, if true. Without having read the rules regulating the challenge, I cannot help but wonder whether annulment only of that part of the award that ordered payment of the allegedly ‘outstanding’ arbitrators’ fees would not have been a better outcome – unless, of course, such separation was impossible. Otherwise, the Claimant having to potentially initiate another proceeding anew seems to be a misdirected solution. Or am I missing something..?

  9. James,

    Thanks for your post, and for citing R24.3 of the CCJA Rules. This provision is effectively a “finality clause” that is common in many institutional rules. Under the interpretation of and effect given to such clauses in most legal systems, an institution can make a “bad” decision, but since the institution preserves to itself the final say, its cannot make a “wrong” decision.

    To the extent arbitrators do not trust an institution to make good decisions, or are concerned that an institution will put them in an “unreasonable” situation, the only third option they have, in my view, is not to accept the appointment in the first place. There is simply no legal basis for a tribunal to purport to “overrule” an institution’s final decision. As indicated in the post, arbitrators are bound by their mandate to apply the arbitral rules agreed to by the parties, including those rules’ allocation of authority and finality clauses.


    1. Catherine,

      I’m grateful for your blog post and I found your reply to my comment informative. However, from a more concrete point of view, the case still raises questions for me. Among these: What were the exact words of the assurance given by the CCJA case manager? Did the assurance override any of the CCJA’s decision-making power under R24.3 (under applicable administrative law)?


  10. Dalma,

    I agree that, in many respects, annulment is an unfortunate outcome of arbitrator misconduct of any sort–it wastes the parties’ time and money, and can have implications for the legitimacy of the system. These concerns are precisely why arbitrators are often said to have a duty to render an enforceable award.

    While decisions to annul or refusal to enforcement an award should not be taken lightly, I would not characterize the violations in this case as minor, or limited to the narrow issue of arbitrator compensation. Many procedural anomalies in how the tribunal managed the case seem to have become wrapped up in this feud over fees. Meanwhile, in opposing enforcement, Guinea also raises what seem to be credible concerns about the fundamental fairness of the proceedings and substantive outcome given that much of the fee issue was being hashed out while the tribunal was deliberating.


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.