Arbitration in Saudi Arabia – Part one of a series
Users of Saudi-seated arbitration often ask: whatever the Saudi Arbitration Law says, how will the courts actually treat an award when it is challenged? Thanks to a growing body of published appellate case law, that question can increasingly be answered from data rather than mere impression. The SCCA Country Report (July 2026)[1] is the most current and comprehensive contribution yet. It examines 967 arbitration-related decisions of the Saudi Courts of Appeal from January 2023 to June 2025, provided by the Ministry of Justice and translated into English with links to the full texts.[2] Prepared at the invitation of UNCITRAL’s Board of Editors for the Digest Revision, it will become part of the forthcoming UNCITRAL Digest of Case Law on the Model Law, placing the Saudi material alongside jurisprudence already familiar to international practitioners.[3]
For many parties, the instinctive question is whether the Saudi courts are “pro-arbitration”. The more revealing question, and the one this data finally lets us answer, is narrower: on precisely what grounds do the Saudi courts set awards aside, and can a particular award survive them? That is what this update addresses, with an eye to what it means for anyone drafting or running a Saudi-linked arbitration.
1. What the dataset does, and does not, cover
Before examining the numbers, two limits are worth keeping in mind. The dataset is highly relevant to commercial arbitration but is not confined to it, as it captures every category of arbitration that came before the Saudi courts.[4] Moreover, the dataset is not a study of how foreign awards are recognized or enforced. Rather, it examines Saudi-seated arbitration under the Saudi Arbitration Law, viewed through the appellate courts that supervise it. By so revealing how the supervisory courts behave, that narrowing is what gives the report its value to a party choosing Saudi Arabia as a seat.
2. Annulment by the numbers
Of the 967 judgments, 194 were applications to annul (set aside) an award, representing 20.1% of the total. The courts refused 174 (89.7%) and granted only 20, producing a success rate of 10.3%, barely one in ten. Of those 20, 12 awards were set aside in full and 8 in part.[5]
Read alongside four earlier SCCA studies covering 2017 to 2023, the combined picture runs to more than 3,300 Court of Appeal judgments across 2017 to 2025. Of the 565 annulment applications in that larger set, 518 were refused (91.7%) and 47 granted (8.3%), with 31 awards set aside in full and 16 in part.[6] A success rate such as this, which holds in the same narrow band for the better part of a decade, is far harder to wave away than a single good year.
3. A narrow remedy, not an appeal
What the courts do with the grounds for annulment is more informative than the success rate. The Courts of Appeal, the report finds, treat the Article 50 grounds as a closed list, not to be stretched by analogy, and will not reopen the facts, revisit the merits, or go behind the tribunal’s view of the evidence. A mistake in how the law was applied, or how the facts were found, is not by itself a reason to set an award aside.[7]
As the report notes, the Article 50 grounds map closely onto Article 34 of the Model Law and Article V of the New York Convention. The dataset measures annulment by the Saudi courts, not enforcement of foreign awards abroad, but the overlap tells its own story. A court that declines to reopen the merits is keeping to the bargain the parties struck: the tribunal decides, and the court steps in only where the statute allows.
4. Where the real risk lies
A low success rate is not the same as immunity. The awards that did fall show clearly that the risk of having an award annulled has not vanished, though it has become specific.
Awards that were fully annulled succumbed to a handful of familiar problems. A few rested on public policy: the retroactive application of the Civil Transactions Law to a limitation question, and the recognition of a lease the law required to be registered. Several concerned how the tribunal was composed, including a truncated panel, an arbitrator who sat out the deliberations on the merits, a change in the panel without a party’s consent, and a unilateral appointment in breach of procedural fairness. Others were due-process failures, such as service to a WhatsApp number other than the party’s registered one, or a party left unable to make its defense. The rest went to authority or form: a non-arbitrable dispute, an award against someone never bound by the agreement, an award issued after the arbitration period had expired without a valid extension, and an award that fell short of the formal requirements.[8]
The partial annulments tell a similar story. Most of those annulments arose due to a tribunal straying past its mandate. Awards were partially annulled where they taxed VAT belonging to the tax authorities, where a “correction” did rather more than correct, where attorneys’ fees were awarded outside the tribunal’s remit, where the tribunal ruled on questions no one had put to it, and where it decided a contract the arbitration agreement did not reach. Two partial annulments were procedural: arbitration fees granted with no request and no real reasoning, which was also contrary to Sharia and public policy, and an additional award made without all the arbitrators taking part.[9]
Annulments based on Sharia and public policy remained limited. Sharia was the basis for annulment in a single case (0.52% of applications) and public policy in only three (1.55%).[10] They are mandatory limits worth understanding, but nothing in the data suggests the courts use them as a way into the merits. A later post in this series will look at both in their own right.
5. What this means in practice
The question is no longer whether the Saudi courts are “pro-arbitration”, but whether a given award can withstand the grounds they apply. Encouragingly, most of the defects that brought awards down can be headed off at the outset.
(a) Get the agreement and the parties right. Name the parties, the disputes in scope, and the appointment mechanism with care, since several annulments turned on precisely these points.
(b) Treat service as a step to be observed, not assumed. Use the agreed channel and registered details, and keep proof of delivery; a notice sent to the wrong WhatsApp number was enough to undo an award.
(c) Protect the tribunal and its deliberations. See that the panel is properly constituted and that every member shares in deciding the merits. A truncated tribunal, an absent arbitrator, or an unconsented change each proved fatal.
(d) Stay inside the mandate and the clock. Keep jurisdiction, the time limit, and any extension under close watch. Awards were lost for deciding what was never referred, ruling outside the clause, or landing after the period had expired.
(e) Watch arbitrability and the mandatory rules. Tax, including VAT, sits outside the tribunal’s hands, and fees awarded without a request or proper reasoning were struck on Sharia and public-policy grounds. Check the relief against those limits before issuing the award.
This is the first piece in a series. The subsequent articles will each take up a specific issue the study brings into view, and draw out what it means in practice for parties arbitrating in Saudi Arabia.
Ontier’s Capabilities
Ontier’s International Arbitration practice advises clients on arbitrations seated in Saudi Arabia and across the GCC, and on the recognition and enforcement of awards. Supported by the firm’s Middle East Desk and by lawyers in Saudi Arabia, the UK, Spain, Italy, the US, and Latin America, we combine regional knowledge with international arbitration experience.
We are positioned to assist with:
– drafting and reviewing arbitration agreements for Saudi-seated and GCC-related contracts;
– tribunal constitution, procedure, and the conduct of Saudi-seated arbitrations;
– annulment (set-aside) proceedings and the defense of awards before the Saudi courts;
– recognition and enforcement of arbitral awards; and
– emergency and interim relief and applications before arbitral institutions, including the SCCA, ICC, DIAC, and LCIA.
Key Contacts
Kamal Khashoggi
Partner, Saudi Arabia
kkhashoggi@ontier.net | +1 617 4357691 / +966 530066212
Lojayne Shaheen
Partner, Saudi Arabia
lshaheen@ontier.net | +966 55 770 9752
Mercedes Romero
Partner, Spain
mromero@ontier.net | +34 648 81 62 36
Claire Morel de Westgaver
Partner, United Kingdom
claire.morel@ontier.co.uk | +44 7867 800879
[1]Saudi Center for Commercial Arbitration, Arbitration in Saudi Arabia: Case Law and Legislative Analysis in Light of the UNCITRAL Model Law and Saudi Arbitration Framework, Country Report (July 2026) (the “SCCA Country Report”), Executive Summary, pp. 5–7. Page references are to the report as reviewed and should be confirmed against the final published version.
[2]SCCA Country Report, pp. 6–7, 31, 35. The report describes 967 Court of Appeal decisions issued between January 2023 and June 2025, provided by the Ministry of Justice, translated into English and made available through direct links to the full texts.
[3]SCCA Country Report, Executive Summary, p. 6; Part I, pp. 30–31. The report was prepared in response to an official invitation from the Board of Editors for the Digest Revision of UNCITRAL and is to form part of the forthcoming UNCITRAL Digest of Case Law on the Model Law on International Commercial Arbitration.
[4]SCCA Country Report, p. 35. The report states that the study encompasses all categories of arbitration brought before the Saudi courts and is not limited to commercial arbitration, and that in every case examined the seat was in the Kingdom and the applicable law was the Saudi Arbitration Law.
[5]SCCA Country Report, pp. 10–11, 110–111.
[6]SCCA Country Report, Executive Summary, p. 11; see also p. 35. The cumulative figures are drawn from the present study together with four earlier SCCA studies (2021, 2022, 2023, and 2025) of Court of Appeal judgments issued between 2017 and 2023; the report notes that the present study is methodologically independent.
[7]SCCA Country Report, pp. 10, 95–96, 110.
[8]SCCA Country Report, pp. 111–113.
[9]SCCA Country Report, p. 113.
[10]SCCA Country Report, pp. 11, 111.
