Appraising The New English Arbitration Act 2025 And Some Lessons For Nigeria

The first, and arguably the most discussed, innovation introduced by the AA is found in section 1, which amends section 6 of the 1996 Act on the law applicable to an arbitration agreement where the agreement itself is silent on the point.

As noted elsewhere¹ , when a dispute arises under a cross-border contract containing an arbitration clause, the parties typically indicate (i) the governing law of the substance of their dispute and (ii) the law of the seat of arbitration, or such choice(s) may be implied from the circumstances of the case. More often than not, the parties fail to specify the law governing the arbitration agreement itself. This omission raises a fundamental question: in the absence of express stipulation, should the law governing the arbitration agreement follow the law of the substance of the dispute or the law of the seat?

This question has long troubled the international arbitration community. Under English law, it was historically one of the few areas of uncertainty, with two opposing schools of thought supported by conflicting decisions of the English Court of Appeal.² Far from being a merely academic debate, the issue carried significant practical consequences, often leading to costly litigation solely to determine the applicable law of the arbitration agreement.

Prior to the enactment of the AA, the UK Supreme Court confronted the matter in Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb ("Enka v. Chubb").³ In a closely split 3:2 decision, the Court held that where the seat of arbitration and the law governing the contract had been expressly or impliedly chosen but the parties failed to specify the law to govern their arbitration agreement, then the governing law of the contract would ordinarily apply to the arbitration agreement (being a part of the contract), absent good reason to the contrary. If no governing law of the contract was chosen, then the "closest connection" applied, which would almost invariably point to the law of the seat.

While this decision was welcomed in some quarters for providing much-needed clarity, it also attracted criticism.⁴ Unsurprisingly, the AA addressed the issue as a matter of priority. In clear terms, it departs from Enka v. Chubb by providing that the law applicable to an arbitration agreement is either (i) the law expressly agreed to by the parties to govern the arbitration agreement; or (ii) in the absence of such agreement, the law of the seat of arbitration.

Given the prominence of the debate and the attention Enka v. Chubb received, it is somewhat surprising that the drafters of the AMA did not take the opportunity to clarify the approach Nigerian courts should adopt when faced with this question. There remains a notable scarcity of Nigerian jurisprudence undertaking a detailed analysis of the competing laws that may apply in arbitration disputes, particularly in determining the governing law of the arbitration agreement.⁵ In practice, Nigerian courts have often looked to English decisions, including those of the UK Supreme Court, where there is a gap in Nigerian law. On that basis, it would have been reasonable to expect that Enka v. Chubb would be followed. However, the enactment of the AA has now complicated matters: it is uncertain whether Nigerian courts will still consider Enka v. Chubb persuasive in light of section 1 of the AA.

What this means is that only an amendment of the AMA can categorically bring this controversy to an end as a matter of Nigerian law. This legal certainty is necessary if Nigeria is to continue its quest to be seen as an arbitration-friendly jurisdiction.

A further significant innovation of the AA is its codification of the arbitrator's duty of disclosure. Section 2 of the AA introduces a new section 23A, which provides that an individual approached with a view to possible appointment as an arbitrator must, as soon as reasonably practicable, disclose any relevant circumstances of which they are, or become, aware. Importantly, the AA defines "relevant circumstances" as those that might reasonably give rise to justifiable doubts as to the individual's impartiality, in relation to the proceedings in question or other potential proceedings. The AA further provides that an individual is deemed aware of circumstances of which they ought reasonably to be aware.

This provision effectively codifies the reasoning of the UK Supreme Court in Halliburton v Chubb, ⁶ a landmark decision delivered on 27 November 2020. In that case, the Court clarified that even where no real possibility of bias is ultimately established, an arbitrator nonetheless owes a duty to disclose circumstances which might "reasonably" give rise to doubts about their impartiality.

The Nigerian position is contained in section 8 of the AMA, which similarly requires a person approached for possible appointment as arbitrator to disclose to the parties any relevant circumstances not within the parties' knowledge. However, unlike the AA, the AMA neither defines "relevant circumstances" for this purpose nor expressly requires disclosure to be made "as soon as reasonably practicable." While the formulation under the AA is undoubtedly clearer, a Nigerian court interpreting section 8 of the AMA would likely adopt a similar approach, both in construing "relevant circumstances" and in requiring prompt disclosure. This view is reinforced by the fact that section 8 of the AMA imposes a continuing duty of disclosure from appointment and throughout the arbitral proceedings.

Section 29 of the 1996 Act, just like section 13 of the AMA, contained the provision on the immunity of an arbitrator, appointing authority or an arbitral institution for anything done or omitted to be done in the discharge of their functions unless such action or omission is shown to have been in bad faith.

Section 4 of the AA has now amended sections 25, 28 and 29 of the 1996 Act by expanding the scope of such immunity by explicitly protecting arbitrators from liability for resignations and applications for their removal, unless such resignation is shown to be unreasonable. This is a welcome addition as it reinforces the independence of arbitrators.

A key area of divergence between the positions of English and Nigerian law is in respect of the power of a court to hear and determine a question as to the substantive jurisdiction of the tribunal. A recurring difficulty for parties is how best to proceed where it appears manifestly clear that the tribunal lacks jurisdiction. Should the party nonetheless participate in the arbitration, raise the objection before the tribunal in line with the competence-competence principle, and incur costs in the process? Or may the party approach the court directly to challenge the tribunal's jurisdiction?

Under section 32 of the 1996 Act, the court could, on the application of a party, determine any question as to the tribunal's substantive jurisdiction. However, such an application was only admissible if either (i) all the other parties to the arbitration consented in writing, or (ii) the tribunal itself permitted the application, and the court was satisfied that the determination was likely to produce substantial cost savings, that the application was made promptly, and that there was good reason why the court should decide the matter.

The AA has now refined this position. By virtue of section 5, a new subsection (1A) has been inserted into section 32 of the 1996 Act, which provides that the court must not entertain an application on a jurisdictional question where the tribunal has already ruled on that issue. In such circumstances, the only recourse available to a party is to seek to set aside the arbitral award on the ground of lack of substantive jurisdiction.

In contrast, the AMA is silent on whether a Nigerian court may entertain a jurisdictional challenge that has not first been submitted to the tribunal. Section 14(6) of the AMA merely provides that, where a tribunal has ruled on its jurisdiction, a party may, within 30 days, apply to the court to decide the matter. It therefore remains uncertain, in light of the competence-competence principle, whether Nigerian courts will permit a party to bypass the tribunal altogether and approach the court directly on jurisdiction.

While competence-competence is a well-settled principle, and the preferable course is generally to raise jurisdictional objections before the tribunal, the more nuanced approach under English law is, in our view, the better model. There are indeed cases where an early recourse to the court is justified, for example, where the tribunal manifestly lacks jurisdiction and costs can be avoided by resolving the question upfront. At the very least, Nigerian courts should adopt a cautious approach, imposing a high burden of proof on any party seeking to bypass the tribunal and demonstrate why the court, rather than the tribunal, ought to hear and determine the issue.

The AA, in section 6, also makes clear that a tribunal shall have powers to impose costs even in instances where the tribunal rules that it has no substantive jurisdiction or has exceeded its substantive jurisdiction. Section 61 of the 1996 Act has been amended accordingly. This provision flows from the well-established rule under the English court that costs follow the event and is consistent with the rule that parties with frivolous cases should compensate respondents for loss suffered in defending such proceedings. The AMA will benefit from a similar provision.

Another cost saving measure introduced by the AA is found in section 7 of the AA which amends section 39 of the 1996 Act by empowering arbitral tribunals to make an award on a summary basis in relation to a claim or a particular issue where the arbitral tribunal considers that a party has no real prospect of succeeding on the claim or a party has no real prospect of succeeding in the defence of issue. This is akin to the special procedure under Rule 41 of the ICSID Arbitration Rules 2023, which empowers ICSID Tribunals in deserving cases to render an early award and find that a claim is manifestly without legal merit.⁷

There is no equivalent provision under the AMA, and at a time when users of the arbitration system are beginning to question the costs associated with arbitration proceedings, this is another area that merits the attention of the lawmakers in Nigeria when considering the next round of amendments to the AMA.

Further, section 9 of the AA amends the 1996 Act to strengthen the powers of English courts to intervene in support of arbitration proceedings. Section 44 of the 1996 Act empowered English courts to make orders in relation to the taking of evidence of witnesses, the preservation of evidence, orders relating to properties subject to the proceedings, etc. Section 9 of the AA has now made it clear that those powers may be exercised by the court both in relation to parties to arbitration proceedings and also in relation to third parties. This provision aligns with Section 43 of the AMA, which empowers Nigerian courts to compel the attendance before an arbitral tribunal of a witness wherever they may be within Nigeria. The AMA is not limited to parties and appears wide enough to cover third parties.

As noted previously, where an arbitral tribunal has ruled on its jurisdiction, the only option available to a party under the AA is to apply to set aside the arbitral award relying on the ground of lack of substantive jurisdiction. To give effect to this, section 10 of the AA has amended section 67 of the 1996 Act by narrowing the scope of challenging such awards on jurisdictional grounds by preventing losing parties from introducing new evidence or arguments in an attempt to obtain a full rehearing by the court. This has the effect of reducing delays and unnecessary costs associated with the post-award process.⁸

Lastly, section 12 of the AA has amended section 70 of the 1996 Act by clarifying the start date for the 28-day limit for challenging an arbitral award under the different applicable scenarios.