This is the second post in ICCA’s series of posts focused on international arbitration in Africa in the lead up to the ICCA-KIAC joint conference “Africa & International Arbitration: Untold Stories”, taking place in Kigali on 5 June 2025.
Introduction
The concept of immunity operates in distinct legal contexts. Under international law, it functions as a protection of state sovereignty, exempting diplomats from certain laws and taxes by host states. Domestically, it constitutes a formal exemption from legal liability. In arbitration, immunity manifests differently across legal traditions. In common law jurisdictions, arbitrators are typically shielded from liability for acts performed in their arbitral capacity, akin to judicial immunity. Conversely, in civil law systems, arbitrators are professionals subject to contractual and tortuous liability.
The central debate surrounding arbitrator immunity concerns whether arbitrators deserve immunity and, if so, the extent of such protection. While scholarly consensus acknowledges that immunity is essential for safeguarding arbitral neutrality and impartiality, its practical application varies significantly across jurisdictions. Some legal systems, like the United States of America, grant absolute immunity, treating arbitrators similarly to judges to preserve independence. Others, like France and England, adopt qualified immunity, limiting liability based on the quasi-judicial and contractual nature of the arbitrator’s status and functions. A third category (including Spain) imposes no immunity, subjecting arbitrators to full civil liability under contract and tort law.
These divergent approaches stem from competing theoretical justifications such as jurisdictional theory (supporting absolute immunity), contractual theory (supporting liability); and hybrid theory (supporting qualified immunity). While these theories help us to understand the basis for the liability or non-liability of arbitrators, practical realities of arbitration seem to suggest that focusing solely on limiting liability is insufficient. The true challenge lies not merely in liability exposure but in post-award litigation, where arbitrators remain vulnerable to judicial scrutiny over their conduct, despite formal immunity.
This form of immunity, contingent on good faith or bad faith, is similar to a cloak of cotton candy that dissolves quickly under judicial scrutiny. Consider this scenario: an arbitrator in Kenya renders a final award, confident in statutory immunity for acts performed in good faith. Weeks later, they are summoned to defend allegations of bad faith in Court. Does qualified immunity genuinely protect arbitrators, or does it merely defer liability? In this blog post, I critically examine arbitrator immunity in East Africa, arguing that the region’s legal frameworks ostensibly protect arbitrators yet fail to provide meaningful immunity due to carve-outs that expose them to post-award litigation, challenging qualified immunity as a viable alternative to full arbitrator immunity. The discussion focuses on the arbitration legal regimes in Uganda, Kenya, Tanzania, and Rwanda, highlighting similarities and differences in their laws, as well as the gaps and challenges in achieving true immunity for arbitrators.
Arbitrator Immunity in East Africa
Among the surveyed East African countries, only Kenya and Tanzania have explicit legal provisions for arbitrator immunity.
In Uganda, the Arbitration and Conciliation Act lacks an explicit arbitrator immunity provision, despite recent amendments in 2008 and 2024 and manifestations in favour of reforming the law to include immunity by the Uganda Law Reform Commission (ULRC). Rwanda’s Law on Arbitration and Conciliation in Commercial Matters is likewise silent, and the country’s position on the matter remains unclear. The 2022 Alternative Dispute Resolution Policy in Rwanda recommends repealing the entire Act to establish separate statutes for arbitration and mediation, but, so far, only the Kigali International Arbitration Centre (KIAC) Rules under Art. 47 provides a form of protection through the exclusion of liability.
Even in Kenya and Tanzania, arbitrator immunity is a relatively recent phenomenon, with amendments to the Kenyan Arbitration Act relating to immunity being made in 2009 and in Tanzania in 2020, repealing the Tanzania Arbitration Act 1931. Consequently, the practical application of these provisions has yet to be extensively tested in courts.
In Kenya, the Arbitration Act 1995, under section 16B(1), provides immunity from liability for “anything done or omitted to be done in good faith in the discharge or purported discharge of [their] functions [as arbitrators].” This immunity extends to a servant or agent duly authorised by the arbitrator but does not cover arbitral institutions and arbitrators who resign or withdraw (section 16B (2) & (3)).
In Tanzania, under the Arbitration Act 2020, section 31(1) protects arbitrators from liability for anything done or omitted in the discharge or purported discharge of their functions unless proven to be in bad faith or due to professional negligence. This immunity also extends to the employee or agent of the arbitrators (section 31(2)), and arbitral institutions and other appointing authorities (section 76(1)).
While both countries take a qualified approach to an arbitrator’s immunity, the language used is different. While the Kenyan provision protects arbitrators for “acts or omissions done in good faith,” the Tanzanian provision removes the protection in case of bad faith or professional negligence. Despite this difference in language, the provisions have the same impact. Both language choices seem to suggest that the immunity is relevant at the end of the proceedings since the actions and omissions can only be reviewed post-facto.
The illusion of immunity
Neither the Kenyan nor the Tanzanian Arbitration statutes define good faith or bad faith. However, in Bellevue Development Company Ltd v Gikonyo & 3 Others, the Supreme Court of Kenya, while examining Article 160(5) of the Kenyan Constitution (which grants immunity to judicial officers and coached in language similar to section 16B (1)) adopted Black’s Law Dictionary’s definition of good faith as a “state of mind consisting in honesty in belief or purpose, faithfulness to one’s duty or obligations, observance of reasonable commercial standards of fair dealing in a given trade or business, or absence of intent to defraud or to seek unconscionable advantage.” Further, it defined bad faith as an antithesis to acting in good faith with a wilful intent to act dishonestly or unfaithfully in the performance of judicial acts.
The Court cautioned against treating the term “good faith” as a limitation rather than a presumption that acting in a judicial capacity within jurisdiction implies acting lawfully and in good faith. Moreover, the court highlighted that deviating from this understanding of good faith “would lead to an absurd position of good faith bases of judges’ actions being debatable points and open to an intolerable deluge of litigation, each unhappy litigant suing judges left, right, and centre as wounded pride dictates.” Yet, this reasoning has not fully insulated arbitrators from post-award challenges.
This kind of qualified immunity exposes arbitrators, undermining their ability to make robust and impartial decisions. Hodges notes that the “downside of a carve-out for bad faith or negligence is that these […] claims inevitably require considerable factual enquiry, which may prevent claims from being dismissed in a summary fashion.”
This is illustrated by the Kenyan case of Junction Apartments Limited v CM Construction (E.A) Ltd & Another, where the High Court dismissed an arbitrator’s immunity claim under section 16B against being added as a party to an application for setting aside. The court held that “although the arbitrator enjoyed immunity […], allegations made against him [were] subject to proof of whether the Arbitrator acted in good faith or not and to buttress the setting aside of the arbitral award. This [called] for evidence and [could not] be curtailed at the preliminary objection.” The case proceeded to trial, demonstrating how such carve-outs often lead to protracted litigation.
Those carve-outs function alongside challenge and removal provisions to provide avenues against an arbitrator’s performance during and after the proceedings (sections 13-15, Kenyan Arbitration Act; sections 25-26, Tanzanian Arbitration Act). For instance, the Tanzanian Act permits a party to apply to the arbitration centre to remove an arbitrator on grounds of refusal or failure to conduct the proceedings properly, to use all reasonable dispatch in conducting the proceedings, to make an award, and where substantial injustice has been or will be caused to the applicant (section 26(1)(d)). If parties already possess the tools to address arbitrator misconduct, why allow post-award suits?
If arbitrators are willing to accept the parties’ power to challenge them during the proceedings, immunity, then, is a necessary concession to protect them after the proceedings are done. Granting immunity based on proof of good faith and bad faith casts doubts over the protection of arbitrators from suit, especially during the practical application of the provisions at the end of the arbitral proceedings. The consequences of those carve-outs suggest that qualified immunity functions as a liability deferral rather than true protection.
Conclusion
East Africa’s arbitration regimes ostensibly protect arbitrators while exposing them to post-award litigation. Reforms to those regimes should amend the current provisions or introduce provisions that grant arbitrators real protection post-award. The good faith/bad faith carve-outs undermine immunity’s purpose through their vague definitions and evidential requirements, forcing arbitrators into costly litigation. New reforms should clearly define good faith to prevent frivolous claims and include provisions requiring parties who fail to challenge an arbitrator’s conduct during the proceedings to waive their right to do so before courts after the proceedings. Until reforms address the flaws of the good faith standard, qualified immunity remains a theoretical safeguard. Arbitrators deserve immunity from suit, not just limited liability.
Follow along and see all of Kluwer Arbitration Blog’s coverage of ICCA Kigali 2025 here.
________________________
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.
