The unappealable finality of arbitral awards underpins its widespread acceptance as a dispute resolution mechanism, yet courts retain a narrow gatekeeping role to prevent outcomes that violate fundamental principles of justice.
The 2024 Kenyan High Court decision in Ongata Works Limited v. Tatu City Limited prompts a debate on whether courts can control arbitral cost awards. In particular, how absolute is tribunal discretion under Article 38 of the ICC Rules or other similar institutional rules?
The decision raises fundamental questions: Did the tribunal exceed its authority by allocating costs for a party-appointed expert witness under Article 38(1)? Did the Kenyan High Court justifiably invoke public policy to set aside the award, or did it undermine arbitral autonomy? This post explores the tension between tribunal discretion and judicial oversight in cost allocation.
Background
The dispute arose from an ICC arbitration where Ongata Works Limited (Claimant) secured damages of Kshs. 21,541,823.40 against Tatu City Limited (Respondent). However, the tribunal also ordered the Claimant to pay the Respondent’s legal and expert witness costs totalling Kshs. 109,479,853 (USD 842,152.72) at an exchange rate of 1 USD = 130 Kshs. This was over five times the damages awarded. The Claimant challenged the cost award before the Kenyan High Court, arguing that it was disproportionate, punitive and violated public policy.
The High Court set aside the award, citing two key grounds:
- Misapplication of the ICC Rules: The tribunal included costs for the Respondent’s party-appointed expert, which the Court deemed impermissible under Article 38(1) of the ICC Rules.
- Disproportionality of the Cost Award: The Court held that the disproportionate cost burden allegedly restricted access to justice. Particularly, the court observed that while tribunals have discretion to allocate costs under Section 32B (1) of Kenya’s Arbitration Act, the discretion must be exercised reasonably. This culminated in a violation of public policy.
Article 38 ICC Rules: Tribunal Discretion and the Scope of “Other Costs”
Article 38 of the ICC Rules outlines the framework for cost allocation under the ICC, granting tribunals broad discretion while imposing certain boundaries:
- Costs Defined: Article 38(1) of the Rules includes the reasonable legal costs, which envisage the ad valorem method, and other costs incurred by the parties for the arbitration. Though it provides that the costs of the tribunal include the costs for tribunal-appointed experts, the provision does not explicitly exclude the costs of party-appointed experts. Therefore, the tribunal in Tatu City had the liberty to interpret “other costs” to encompass the Respondent’s expert witness fees, provided they were reasonable and necessary for the case.
- Reasonableness and Proportionality: Article 38(5) of the Rules provides that tribunals are to take into account factors they consider relevant in awarding costs. This includes the parties’ conduct during the arbitration. As such, the tribunal has the discretion to justify the allocation of expert costs by assessing their relevance to rebutting the Claimant’s claims, among other considerations.
- Final Allocation Authority: Under Article 38(4) of the Rules, the tribunal retains ultimate authority to fix the costs of the arbitration and decide the proportion each party is to bear. The tribunal in Tatu City allocated 100 per cent of the Respondent’s costs to the Claimant, possibly due to considerations guaranteed under Article 38(5) of the ICC Rules.
Through examination of these Rules, the tribunal’s decision regarding the types of reimbursable costs aligns with ICC practice. Article 38(1)’s silence on party-appointed expert fees implies a non-exhaustive list of reimbursable costs, granting tribunals interpretive discretion to include such expenses when reasonably incurred. However, a separate issue arises in the allocation of those costs. Even if reimbursable in principle, the decision to assign 100% of the Respondent’s legal and expert costs to the Claimant, despite the Claimant partially succeeding in the arbitration, raises legitimate concerns about proportionality. The ICC Rules (Article 38(5)) require tribunals to consider relevant factors, including party conduct and the outcome of the case. Thus, while the High Court may have erred in excluding expert fees outright, its discomfort with the disproportionate allocation of costs relative to the damages awarded and each party’s success rate is more defensible. This distinction is crucial: tribunals enjoy broad discretion under the ICC Rules, but such discretion must be exercised in line with principles of equity and fairness embedded in the arbitral framework.
Judicial Intervention: Public Policy or Overreach?
Public Policy in arbitration as defined in the Kenyan decision in Christ for All Nations v. Apollo Insurance Co. Limited (2002) EA 366, “… must be confined to matters that fundamentally offend the country’s sense of justice, morality or national interest. Mere procedural irregularities or erroneous decisions by arbitrators cannot be elevated to public policy violations.” This is reflected under the New York Convention Article V(2)(b), where public policy defenses are reserved for breaches of a state’s most basic notions of morality and justice. Comparative jurisprudence from Singapore as seen in VV v. VW, where the Singapore High Court declined to review a cost award, affirming the place of tribunal discretion in the allocation of costs. Similarly, the Swiss Federal Tribunal (2006) ruled that courts could only intervene if costs were “wholly disproportionate to the necessary costs of defense”, reinforcing the view that tribunals have significant autonomy in determining costs.
The Kenyan High Court’s rationale that excessive costs violate public policy marks a significant departure from both national and international arbitration norms. The Court’s reasoning reflects a misapplication of Article 38(1) of the ICC Rules, adopting a narrow construction that excludes party-appointed expert costs — a position at odds with the ICC’s flexible cost framework, which affords tribunals discretion in cost allocation. More importantly, however, the court appears to have mischaracterized the nature of the cost issue. While it relied on the “disproportionality” of the award, its analysis conflates two distinct issues: the amount of costs and their allocation.
Contrary to the court’s apparent reasoning, the total costs (Kshs. 109 million / ~USD 842,000) may not have been disproportionate to the overall scale and complexity of the dispute. The more legitimate concern — and one that goes largely unexamined by the court — is that 100% of those costs were allocated to the Claimant, even though it partially succeeded in the arbitration. Given that the Respondent successfully defended the bulk of the claim, it is arguably unjust for the Claimant to shoulder the entire cost burden. If the court had instead focused on this issue of cost allocation fairness, its intervention would have been more coherent and defensible.
That said, the court’s conclusion — that such disproportionality amounts to a violation of Kenyan public policy — stretches the doctrine beyond accepted limits. Under both Kenyan case law (Christ for All Nations v. Apollo Insurance) and the New York Convention (Article V(2)(b)), public policy is meant to address fundamental breaches of justice or morality, not errors of cost assessment absent fraud, bad faith, or procedural abuse. Courts in Singapore, as discussed above, have held that disproportionate cost awards alone do not rise to the level of public policy violations unless they are wholly arbitrary or unjustifiable. The Kenyan High Court’s reliance on abstract proportionality, without identifying a clear violation of fundamental principles, risks setting a precedent for overly broad judicial review, undermining the autonomy central to international arbitration.
Conclusion
The Tatu City case presents both opportunities and challenges for arbitration practice in Kenya and a source of reflection for international arbitration in general. On one hand, the ruling promotes greater accountability in cost allocation, ensuring that tribunals exercise their discretion within reasonable limits. On the other hand, the broad interpretation of public policy risks encouraging more intrusive judicial oversight, with undue judicial interference in the arbitral discretion in cost awards.
While the cost award in Tatu City was severe, the High Court’s use of public policy to address disproportionality was misplaced. Disproportionate costs, without fraud or due process violations, do not justify non-enforcement under public policy. Concerns about cost fairness should be resolved through procedural safeguards, such as reasoned cost decisions and party submissions, not post-award court intervention.
Arbitration already offers mechanisms to address cost-related concerns. Parties can request reasons for allocations, make targeted submissions, and challenge surprise decisions issued without prior notice. Where tribunals fail to engage with such submissions, due process risks may arise — but these are procedural, not public policy issues. Tatu City may involve such overlooked dynamics.
As arbitration under ad valorem models grows, courts must strike a balance: upholding fairness without eroding tribunal autonomy or finality. That balance is key to maintaining the legitimacy of the arbitral system.
________________________
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.
