For nearly 30 years, the Arbitration Act 1996 (AA 1996) has provided an effective and popular framework for arbitrations seated within England, Wales and Northern Ireland.¹ By 2021, however, calls for modernisation of the AA 1996 had grown louder, as many of London’s competitors, including Singapore, Switzerland and Germany, took steps to modernise their own arbitration laws.

In November 2021, the UK Government asked the Law Commission to ensure that the UK’s arbitration legislation remained “state of the art,” both for domestic and international commercial arbitration.² The Law Commission conducted a wide-ranging consultation with stakeholders, which led to the publication of its recommendations for amendments to the AA 1996 on 5 September 2023. These recommendations included, in the words of the Law Commission, some “major initiatives, and a very small number of minor corrections.”³ 

These recommendations have now been implemented by the Arbitration Act 2025 (AA 2025), which received Royal Assent on 24 February 2025. The date on which the AA 2025 will take effect remains to be set out in forthcoming regulation, but the Government has indicated that this will be “as soon as practicable.”⁴

The AA 2025 amends the provisions of the AA 1996, rather than replacing the existing legislation. These changes to the AA 1996 were accurately described by Lord Ponsonby during the passage of the Bill as “evolution not revolution.”⁵  Nevertheless, the amendments significantly update the prior legislative framework for arbitration in the UK and have potentially far-reaching ramifications. This article considers the key reforms and their practical implications.

1. New Clarity Regarding the Applicable Law for Arbitration Agreements

The law governing an arbitration agreement addresses critical matters such as the scope and validity of an arbitration agreement. The AA 1996 was silent as to how the law governing the arbitration agreement should be determined. Parties therefore had to look to common law for the answer. 

Until 2020, the leading authority was Sulamérica Cia Nacional de Seguros SA v Enesa Engenharia SA and others [2012] EWCA Civ 638, which set out a three-stage test for determining the law applicable to an arbitration agreement: (i) have the parties made an express choice of law?; (ii) if not, have they made an implied choice of law?; (iii) if not, with which system of law does the arbitration agreement have its closest and most real connection?  However, the English courts’ approach to the application of this test had been somewhat inconsistent.⁶

The Supreme Court’s 2020 decision in Enka v Chubb [2020] UKSC 38 significantly reformulated the test for determining the law applicable to an arbitration agreement. The Supreme Court in Enka held that, where there is an express or implied choice of law governing the underlying contract, this choice will generally apply to the arbitration agreement,⁷ or, where there is no express or implied choice of law, the law of the seat of the arbitration will generally apply.⁸

In recent years, the Enka approach has been strongly criticised for the significance it places on the law of the underlying contract in determining the law of the arbitration agreement. Under the rule in Enka, if a London-seated arbitration agreement is contained within a foreign law contract (and does not expressly specify a governing law applicable to the arbitration agreement itself), then the arbitration agreement would typically be deemed to be governed by that foreign law. This is significant because, where an arbitration agreement is governed by foreign law, key provisions of the AA 1996 do not apply, with the result that parties are deprived of the full protection and support of the English courts for arbitrations seated in London.⁹

The AA 2025 now provides welcome clarity to the choice of law applicable to arbitration agreements. It provides that:¹⁰

• the law applicable to arbitration agreements is the law expressly chosen by the parties;
• otherwise, it will be the law of the seat of the arbitration; and
• an express choice of law to govern the main contract will not constitute an express choice of law to govern the agreement to arbitrate.

These changes implement a straightforward statutory rule that the arbitration agreement is governed by the law of the seat, absent an express choice of law. As a consequence, parties who choose London as an arbitral seat – absent an express choice of law applicable to the arbitration agreement – now know that any dispute as to the scope or validity of the arbitration agreement will benefit from an arbitration-friendly jurisdiction, offering a robust legal framework that supports and facilitates the arbitration process, and where arbitration agreements are respected and enforced. This reform should further enhance the attractiveness of London as an arbitral seat.¹¹

2. Strengthening of Arbitrator Powers and Immunity

The AA 2025 also strengthens arbitrators’ ability to render robust and impartial decisions in three broad respects.

First, the AA 2025 confirms that arbitrators can make awards on a summary basis on issues that have no real prospect of success. Whether UK-based arbitral tribunals have an inherent power to use summary dismissal has been a heated topic of discussion in recent years. In its Final Report, the Law Commission ultimately concluded that arbitral tribunals seated in the UK “probably” do have “an implicit power to use summary disposal” pursuant to the tribunal’s procedural discretion.¹²  However, in the absence of any express provision under the AA 1996 to this effect, there had been some “uncertainty” on this point.¹³

The AA 2025 dispels any such uncertainty by expressly conferring upon the tribunal the power (unless the parties agree otherwise) to render an award on a summary basis upon an application made by a party to the proceedings, if the other party has no real prospect of succeeding.¹⁴  This is the same threshold as is applied in UK Court proceedings for summary dismissal. Parties and London-based tribunals should therefore benefit from the guidance provided by an extensive preexisting body of case law that interprets and applies the “no real prospect of success” standard.

The AA 2025 further clarifies that: (i) application for summary dismissal must be made on notice to the other parties; and (ii) the tribunal must afford the parties a reasonable opportunity to make representations to the tribunal.

This change sets the UK apart from other arbitration-friendly jurisdictions where arbitrators do not have statutory express summary dismissal powers (for example, France and New York) and should further enhance the appeal of London as a seat. In particular, an effective summary dismissal mechanism has the potential to save substantial time and costs where one party raises frivolous claims or defences for tactical reasons. 

Second, the AA 2025 strengthens arbitrator immunity. The AA 1996 already provides a degree of immunity for arbitrators. Under Section 29 of the AA 1996, arbitrators will not be liable for anything done, or omitted to be done, in the discharge or purported discharge of their functions as arbitrator unless the act or omission is shown to have been in bad faith.

However, Section 29 has been subject to criticism because arbitrator immunity could be lost in the following circumstances: (i)  if an arbitrator resigned; or (ii) if a party applied to Court to have an arbitrator removed, in which case the arbitrator could incur liability for the costs of the application, even if unsuccessful.¹⁵

The AA 2025 addresses, and significantly narrows, these immunity exceptions. It provides that: (i) an arbitrator’s resignation will not give rise to any liability for the arbitrator, unless it is shown that the resignation was, in all the circumstances, unreasonable;¹⁶ and (ii) a Court may not order the arbitrator to pay the costs of an application for their removal, unless any act or omission by the arbitrator in connection with the proceedings is shown to have been in bad faith.¹⁷

This is a welcome amendment. In particular, the changes to the regime regarding arbitrators’ personal liability will further support arbitrators in making robust and impartial decisions without fear of tactical litigation against them.

Third, the AA 2025 empowers emergency arbitrators to issue peremptory orders (i.e., a final order or direction that specifies a time to comply with the emergency arbitrator’s original order) and make relevant applications for court orders in support of peremptory orders.¹⁸

The AA 1996 did not include provisions addressing emergency arbitrators and was therefore silent as to what should happen if an emergency arbitrator’s order was simply ignored by one or both parties. The AA 1996’s silence on this issue is understandable, since the emergency arbitrator procedural innovation was introduced into major arbitral institution rules long after the AA 1996 was first enacted.¹⁹ However, in recent years, this lacuna created uncertainty as to the legal effect of an emergency arbitrator order under UK law and led to some criticism of the UK arbitration framework’s approach to emergency arbitration.²⁰ This criticism took on increased force in the context of the growing prevalence and popularity of emergency arbitration: twenty-eight emergency arbitrator applications were filed under the International Chamber of Commerce (ICC) Rules in 2023, compared to just six in 2013.²¹

The AA 2025 addresses this issue directly and removes any uncertainty as to whether an emergency arbitrator order can be enforced. In practice, this should further encourage the increased use of peremptory orders, where appropriate, by emergency arbitrators seated in London. Ultimately, this should enhance the attractiveness of London as an arbitral seat as emergency arbitration becomes an increasingly valued tool for arbitration users. 

3. New Rules for Jurisdiction Challenges

The AA 2025 also revises the framework for challenges to an award for lack of substantive jurisdiction under Section 67 of the AA 1996. In particular, the AA 2025 envisages the following:²²

• A ground for objection that was not raised before the arbitral tribunal must not be raised with the Court, unless the applicant did not know and could not with reasonable diligence have discovered that ground during the arbitral proceedings;

• evidence that was not put before the arbitral tribunal must not be considered by the Court unless the applicant shows that it could not, with reasonable diligence, have put the evidence before the arbitral tribunal; and

• evidence that was heard by the arbitral tribunal must not be reheard by the Court.

All three of these restrictions are subject to a statutory exception: the AA 2025 recognises that the restrictions do not apply if a Court rules otherwise in the “interests of justice.”²³  It remains to be seen how willing the Court will be to apply this exception.

These changes introduce significant new restrictions on how jurisdictional challenges under Section 67 of the AA 1996 can be pleaded. Under the unamended AA 1996, a jurisdictional challenge could lead to a de novo Court review of the tribunal’s jurisdiction. This could potentially include the Court considering new arguments and evidence and/or rehearing arguments and evidence that had already been put to the tribunal.

Now, under the AA 2025, a party is restricted from raising arguments and evidence that they did not put before the tribunal, unless the arguments and evidence could not have been reasonably discovered during the arbitration. Parties will also be restricted from seeking to relitigate evidence that was put before the arbitrator or tribunal by way of a rehearing. In practice, this effectively reduces the scope of potential Section 67 challenges to those: (i) arising from new evidence that only reasonably came to light after the arbitral proceedings concluded; or (ii) relating to the Court’s review of a tribunal’s decision on jurisdiction, based solely on a reconsideration of the grounds for objection which were put before the tribunal without rehearing any evidence.

This reform is likely to reduce the cost – and increase the efficiency of – Section 67 challenges. It is also likely to further reduce the number of successful Section 67 challenges and to discourage applicants from making unmeritorious challenges in the first place. 

4. Codification of Arbitrators’ Duty of Disclosure

The AA 2025 codifies the existing common law duty on arbitrators to disclose any circumstances that might reasonably give rise to justifiable doubts as to their impartiality in relation to proceedings (as previously set out in Halliburton Co v Chubb Bermuda Insurance Ltd [2020] UKSC 48).²⁴

The AA 2025 further clarifies that such disclosure(s) must be made “as soon as reasonably practical” once an individual is approached for a possible appointment as an arbitrator, and the duty to disclose continues to apply if an arbitrator later becomes aware or ought reasonably to be aware of relevant circumstances.²⁵ 

It is anticipated that there will be little practical change as a result of this amendment. Typically, arbitrators already make relevant disclosures in a timely way, and awards can be successfully challenged where they fail to do so.²⁶  Indeed, the High Court recently reaffirmed the Courts’ willingness to set aside awards on the basis of failures to make proper disclosures in Aiteo Eastern E & P Company Limited v Shell Western Supply and Trading Limited and others [2024] EWHC 1993 (Comm).

While this change therefore mainly reaffirms the disclosure requirements on arbitrators that already exist at common law, it brings the UK’s arbitration legislation into line with international best practice, as adopted, for example, in the UNCITRAL Model Law, the ICC Rules and the LCIA Rules. 

5. Concluding Remarks

The key reforms to the AA 1996 described above ensure that the UK’s arbitration legislation remains state-of-the-art well into the 21st century. The AA 2025 offers clarity and certainty on the law applicable to arbitration agreements. This resolves definitively, and in an arbitration-friendly manner, the problems created by Enka v Chubb for London-seated arbitrations: this is, in the authors’ view, the most significant reform introduced by the AA 2025.

The new, streamlined procedure to challenge jurisdiction and the statutory recognition of arbitrators’ power to order the summary dismissal of claims and grant peremptory orders are important reforms that promote the efficiency of the arbitral process and enhance the attractiveness of London as a choice of seat. Ultimately, whilst it is correct that the reforms represent a steady evolution of UK arbitration law rather than a revolution, that evolution in the authors’ view should secure London’s position as one of the leading centres of the world for international arbitration for many years to come.