Can Institutional Rules Supersede National Arbitration Law? The English High Court on Party Autonomy and LCIA Rules in Genel Energy v KRG

Pritam Dumbré

The beginning of May 2026 has brought with it another significant judgment from the English Courts reaffirming the strength and autonomy of institutional arbitration frameworks. In Genel Energy Miran Bina Bawi Limited v The Kurdistan Regional Government of Iraq[2026] EWHC 1003 (Comm), Mrs Justice Dias of the King’s Bench Division, High Court of Justice of England and Wales, was required to examine a procedural conflict now increasingly familiar within modern international arbitration practice.

Before the Court lay the question of whether the Tribunal’s costs assessment was to be governed by section 63 of the English Arbitration Act 1996 or Articles 28.1, 28.2, and 28.3 of the London Court of International Arbitration (“LCIA”) Rules 2020. Although it was formally framed as a challenge to a costs award under section 68 of the Arbitration Act 1996, the judgment developed into something far broader. The prima-facie issue (evidentiary) appeared to be that of the insufficiently itemised costs schedules. However, the case ultimately tested the extent to which English Jurisdiction’s willingness to preserve institutional provisions chosen by parties even where concerns surrounding fairness and proportionality visibly arise. The verdict that followed is likely to serve as a precedent in guiding where parties expressly adopt institutional rules and to what extent may those rules displace the “default” provisions of national arbitration legislation?

The Arbitration and What the Case Is Actually About

The underlying Arbitration arose from the termination of Production Sharing Contracts concerning oil and gas reserves within the Kurdistan Region of Iraq. The substantive dispute itself concerned whether the Kurdistan Regional Government (“KRG”) had validly terminated those contracts and whether the respondent, Genel Energy Miran Bina Bawi Limited (“GEMBBL”), was entitled to damages.

The arbitration was seated in London and conducted under the LCIA Rules 2020. By adopting those Rules, the parties also excluded any appeal on a question of law under section 69 of the Arbitration Act 1996. The procedural architecture chosen by the parties therefore became central to the dispute which later emerged before the English Courts. In December 2024, the Tribunal ruled substantially in favour of KRG. The Contracts were held to have been validly terminated and GEMBBL’s counterclaims failed. Following its success, KRG sought recovery of legal and expert costs exceeding US$35.5 million incurred over approximately two and a half years of proceedings.

It is important, however, not to mistake this case for a judgment on the substantive merits of the underlying oil and gas dispute, nor even as a direct judicial review of whether the costs claimed were reasonable in absolute terms. The Court repeatedly emphasised that the matter before it was confined to the legality of the Tribunal’s exercise of its powers under the Arbitration Act 1996 and the LCIA Rules. The judgment instead concerns four interrelated procedural questions:

1. whether the LCIA Rules displaced section 63 of the Arbitration Act 1996;

2. whether the Tribunal sufficiently specified recoverable costs;

3. whether inadequate itemisation amounted to an excess of power under section 68(2)(b);

4. and whether the Court should intervene at all in circumstances where parties had consciously chosen institutional arbitration with limited rights of appeal.

In that sense, the decision belongs within the broader English jurisprudence consistently favouring arbitral finality, party autonomy, and minimal judicial interference.

The Costs Controversy

The sheer magnitude of the costs claim immediately drew the Court’s attention. Mrs Justice Dias described the figures as “truly eye-watering.” KRG’s costs submissions sought recovery of more than US$35.5 million in legal and expert fees alone.

What complicated matters further was the manner in which those costs were presented. The schedules provided by KRG contained:

• aggregate figures grouped broadly by categories of fee earners;

• overall monthly billing totals;

• broad descriptions such as “legal advice regarding issues in dispute”;

• and limited allocation of hours, workstreams, or fee-earner involvement.

There was, as Mrs. Justice Dias expressly observed, “no attempt to allocate costs to particular workstreams and no information to show how much time was spent on any particular item of work by any individual fee earner.”

The Tribunal itself appeared uneasy with the material before and acknowledged that it lacked sufficient information to determine whether certain tasks required partner-level fee earners or whether the volume of hours spent on specific tasks was reasonable. Nonetheless, relying upon its familiarity with the proceedings, the Tribunal proceeded to assess the claim.

In doing so, it reduced legal fees by 20%, reduced one expert’s fees by 50%, and ultimately awarded roughly US$26 million together with additional disbursements and arbitration costs.

It was this procedural handling of costs, rather than the underlying merits dispute, which formed the basis of GEMBBL’s challenge before the English Courts.

The Relevant Legal Framework: Deciding between Section 63 and Article 28.

The challenge centred primarily around section 63 of the Arbitration Act 1996 and Article 28 of the LCIA Rules 2020.

Section 63(3) of the Arbitration Act provides:

“The tribunal may determine by award the recoverable costs of the arbitration on such basis as it thinks fit. If it does so, it shall specify —
(a) the basis on which it has acted, and
(b) the items of recoverable costs and the amount referable to each.”

GEMBBL relied heavily upon subsection (b), arguing that the Tribunal had failed to sufficiently “specify” the recoverable costs awarded.

The Tribunal, however, had proceeded under Article 28.3 of the LCIA Rules 2020, which grants tribunals the power to determine legal costs “on such reasonable basis as it thinks appropriate” and expressly provides that tribunals are not required to apply state court assessment procedures.

The central dilemma therefore became straightforward in formulation, though difficult in consequence: Did Article 28.3 constitute a complete and self-contained mechanism for determining legal costs, thereby displacing section 63 of the Arbitration Act 1996?

GEMBBL’s Case: Specificity as a Mandatory Obligation

GEMBBL’s challenge was brought under section 68(2)(b) of the Arbitration Act on the basis of “serious irregularity with Its argument resting upon several connected propositions.

First, GEMBBL argued that section 63(3)(a) and (b) imposed mandatory obligations upon any tribunal exercising the power to determine recoverable costs. In its submission, the Tribunal was required to sufficiently identify the basis of the award together with the specific recoverable items attributed to each component of work.

Second, GEMBBL argued that KRG’s sparse schedules made scrutiny difficult. Thus, without proper itemisation, there could be no adequate assessment of:

• proportionality;

• reasonableness;

• allocation of workstreams;

• or whether partner-level billing was justified.

Third, GEMBBL argued that the Tribunal’s failure to comply with section 63 amounted to an excess of power under section 68(2)(b), rather than a mere procedural error.

Finally, GEMBBL maintained that the matter should instead have been referred to the Court for assessment under section 63(4).

KRG’s Response: A “Complete Package”

KRG’s response focused heavily upon party autonomy and the contractual nature of institutional arbitration.

KRG argued that by adopting the LCIA Rules, the parties had already agreed upon a complete mechanism governing costs assessment. Article 28.3 therefore displaced the non-mandatory default provisions contained within section 63 of the Arbitration Act.

The argument rested upon three central propositions:

• section 63 was not a mandatory provision of the Arbitration Act;

• parties were free to contract out of it;

• and Article 28.3 constituted precisely such an agreement.

KRG further argued that any complaint regarding itemisation concerned only the exercise of the Tribunal’s powers, not the existence of jurisdiction itself. Even assuming the Tribunal had inadequately assessed costs, that would amount at most to an erroneous exercise of power rather than an excess of power capable of engaging section 68.

That distinction proved to be instrumental in making their case.

The Court’s Decision: Institutional Rules Are Not a “Leaky Sieve”

Mrs Justice Dias ultimately dismissed the challenge.

The judgment repeatedly emphasised the principle of party autonomy emphasized within sections 1 (especially b and c) and 4 of the Arbitration Act. The Court accepted that parties were entitled to adopt institutional rules capable of operating as complete procedural systems. It was further held that Article 28.3 of the LCIA Rules constituted a complete and self-contained mechanism governing legal costs. By adopting the LCIA Rules, the parties had displaced the non-mandatory default provisions contained within section 63 of the Arbitration Act 1996.

The Court rejected the suggestion that institutional rules should be treated merely as partial supplements requiring constant reinforcement from domestic legislation. Such an approach, the Court held, would undermine the commercial expectations of parties selecting institutional arbitration.

One of the judgment’s most memorable observations emerged in this context. The Court cautioned against reducing institutional rules into a “leaky sieve” requiring patchwork supplementation from national legislation.

The phrase is likely to survive well beyond this dispute and summarises the verdict itself.

Excess of Power v Erroneous Exercise of Power

The judgment also revisited a foundational distinction within English arbitration law: the difference between an “excess of power” and an “erroneous exercise of power.”

Relying heavily upon Lesotho Highlands Development Authority v Impregilo SpA [2005] UKHL 43, the Court reaffirmed that section 68 is not intended to function as a mechanism disguised for appeal.

The Tribunal unquestionably possessed jurisdiction to determine costs. The complaint instead concerned how that jurisdiction had been exercised. Even assuming the Tribunal had inadequately itemised costs, the Court held that this would amount only to an erroneous exercise of power rather than a jurisdictional excess. Section 68 therefore remained unavailable.

Justice Dias also noted KRG’s concern that accepting GEMBBL’s argument could effectively open the door to routine post-award costs challenges under section 68 whenever dissatisfied parties alleged insufficient itemisation.

A Judgment Marked by Visible Judicial Unease

What renders the judgment genuinely interesting is the tension visible throughout it. Justice Dias repeatedly acknowledged the weaknesses in KRG’s costs presentation. The judgment recognised that the costs figures were “truly eye-watering” and the schedules lacked detail. Additionally, it was observed how the material furnished by KRG was insufficient for proper interrogation of reasonableness.

Indeed, the Court expressly stated that “best practice” would undoubtedly have involved more rigorous presentation of costs schedules.

Despite these concerns, the judgment ultimately reinforced:

• party autonomy,

• arbitral finality,

• and the intentionally narrow threshold for judicial intervention under the Arbitration Act 1996.

That balance perhaps explains why the decision is likely to attract considerable attention within international arbitration circles. The Court appeared uncomfortable with aspects of the costs presentation, yet nonetheless refused to permit judicial intervention beyond the tightly confined framework established under English arbitration law.

Concluding Remarks

The significance of GEMBBL v KRG lies less in the underlying costs dispute itself and more in what the judgment reveals about the modern English supervisory approach toward institutional arbitration.

The decision reinforces several themes now salient to English arbitration jurisprudence:

• institutional rules are capable of operating as comprehensive procedural systems;

• non-mandatory statutory provisions may readily be displaced by party agreement;

• section 68 remains an exceptional remedy;

• and English Courts will continue resisting attempts to convert procedural dissatisfaction into disguised merits appeals.

For arbitration practitioners, the judgment could perhaps serve as a reminder to adhere by the best practices. Even where institutional autonomy prevails legally, tribunals and parties alike may still face judicial criticism where costs submissions lack transparency and proper substantiation.

Ultimately, the English Courts will preserve arbitral autonomy, however, they may not necessarily conceal their discomfort whilst doing so.