Introduction
GAFTA arbitration sits at the heart of the international grain and feed trade. It is fast, specialist, and built around standard-form contracts that those traders use to seek commercial certainty and expert decision-making. Yet the fact that a dispute is referred to GAFTA does not remove it from the supervisory framework of English arbitration law, where the seat is London (which is the case by default where GAFTA arbitration is chosen). Once a London-seated GAFTA award is issued, any recourse to the High Court is governed by the Arbitration Act 1996, not by any free-standing doctrine of merits review.
GAFTA itself operates a two-tier arbitral structure, ordinarily involving a first-tier tribunal and a Board of Appeal. That internal appeal is part of the arbitral process. Recourse to the court is something different. In the High Court, the available routes are statutory and limited. They are principally: a challenge to the tribunal’s substantive jurisdiction under section 67; a challenge for serious irregularity under section 68; and, unless excluded by agreement, an appeal on a point of English law under section 69. Those provisions are then supplemented by section 70, which imposes important requirements on exhaustion of arbitral remedies, timing and procedure.
The distinction is not merely conceptual. It is often outcome-determinative. A party that dresses up a factual complaint as a point of law, or who tries to turn a merits grievance into a section 68 irregularity argument, is likely to fail. Equally, a party that misunderstands the relationship between the GAFTA appeal process and the court deadlines may find that even a potentially good point is lost on limitation. The recent authorities show that the Commercial Court continues to support the finality of specialist commodity arbitration while remaining willing to intervene where the Act permits it and where the relevant statutory threshold is genuinely met.
The post-2025 landscape also matters. The Arbitration Act 2025 came into force on 1 August 2025. It made important changes to the procedural architecture governing challenges. Most notably, it changed the handling of section 67 jurisdiction challenges, clarified the time computation rules in section 70, and tightened the relationship between section 32 and section 67. Those reforms matter directly to GAFTA users because commodity arbitrations frequently raise jurisdictional objections, time-sensitive challenges, and questions about whether an issue should first be taken to the tribunal, to an arbitral appeal board, or to the court. This article examines the principal routes by which a GAFTA award may be challenged or appealed in the High Court; the governing principles and procedural requirements under each route; the principal grounds relied upon in practice; and the leading authorities, especially recent GAFTA and commodity cases. It then explains what changed with the coming into force of the Arbitration Act 2025 and concludes with practical observations for drafting, case management, and challenge strategy.
The GAFTA structure and the supervisory role of the High Court
GAFTA’s arbitration system is itself an important part of the legal context. GAFTA explains publicly that its system is a two-stage arbitration process and that its standard contracts incorporate its arbitration rules. That matters because, in many cases, the award that comes before the High Court is not the first-tier award but the Board of Appeal award. The party considering court recourse must therefore begin by asking a preliminary question: what is the operative award, and has the available arbitral process of appeal or review been exhausted?
That question leads immediately to section 70(2) of the 1996 Act. The statute provides that an application or appeal under sections 67 to 69 may not be brought unless the applicant has first exhausted any available arbitral process of appeal or review and any available recourse under section 57. In GAFTA practice, that often means that the Board of Appeal award is the relevant award. But it is not always that simple. As PEC Ltd v Asia Golden Rice Co Ltd shows, the answer depends on the structure of the relevant arbitral rules and on the nature of the challenge being pursued.
The second structural point is that the High Court is not a second commercial merits tribunal. That proposition is basic, but it is worth stating clearly because it explains why the statutory routes are framed as they are. The Arbitration Act 1996 is built around party autonomy, finality, and limited court intervention. Section 1(c) expressly states that the court should not intervene except as provided by the Act. The leading authorities on sections 68 and 69 repeatedly emphasise that the purpose of the court is not to re-try the arbitration or to supervise every detail of the arbitral reasoning. It is to intervene only where the statutory conditions are met.
The modern GAFTA authorities illustrate that supervisory approach well. In Sharp Corp Ltd v Viterra BV, the Supreme Court entertained a true section 69 appeal from GAFTA awards, but it did so by insisting on fidelity to the tribunal’s factual findings and by rejecting an appellate approach that had gone beyond the issue actually before the arbitrators. In CAFI–Commodity and Freight Integrators DMCC v GTCS Trading DMCC, the court intervened because the GAFTA Appeal Board had gone wrong on jurisdiction, had failed properly to resolve a live issue central to liability, and had also erred on a point of law. In Trans Trade RK SA v State Food and Grain Corporation of Ukraine, Andrew Baker J allowed a section 69 appeal because the Board had erred on section 49(2) of the Sale of Goods Act 1979, but he did not use section 69 as a licence to invent an unargued alternative case.
What follows is an analysis of s.67, s.68 and s.69 challenges as illustrated by recent Commercial Court cases.
Section 67: challenge for lack of substantive jurisdiction
The legal framework
Section 67 is the route where the complaint is that the tribunal lacked substantive jurisdiction. Section 30 explains what “substantive jurisdiction” means: whether there is a valid arbitration agreement; whether the tribunal is properly constituted; and what matters have been submitted to arbitration in accordance with that agreement. In GAFTA disputes, section 67 issues commonly arise in four types of case: incorporation disputes, authority disputes, notice-of-arbitration disputes, and scope disputes.
The critical point is that section 67 is about authority, not correctness. The court asks whether the tribunal had the power to decide the dispute or the issue in question. It does not ask whether the tribunal was merely wrong on the merits.
Key section 67 principles illustrated by case law
The first principle is that the existence of a GAFTA arbitration agreement remains a matter of proof and construction. Even in a trade in which GAFTA forms are commonplace, the court will not lightly infer a valid arbitration agreement without a proper contractual basis.
That is shown most clearly by Black Sea Commodities Ltd v Lemarc Agromond Pte Ltd. The dispute arose out of negotiations in March 2018 concerning the sale of Ukrainian corn FOB Odesa. The seller challenged a GAFTA award under section 67, arguing that no binding arbitration agreement had ever been made. The buyer argued, first, that later exchanges of draft conditions had created an arbitration agreement and, second, by a late amendment, that a GAFTA arbitration clause was implied by trade custom into an earlier contract allegedly concluded on 9 March. Sir Michael Burton accepted the seller’s argument. The court held that the issue before it was whether there was a binding agreement for arbitration, not merely whether there was a sale contract. On the custom point, the court stressed that any alleged trade custom had to be “binding” in the sense of being invariable and not merely a practice sometimes followed in the market. The case is a warning against assuming that widespread market practice is enough to incorporate a GAFTA clause without a proper contractual hook.
The second principle is that a notice of arbitration will be construed commercially and in substance, not with sterile technicality. That principle is illustrated by LLC Agronefteprodukt v Ameropa AG. There, two wheat sale contracts each contained separate GAFTA arbitration clauses. The buyer’s notice of arbitration referred to both contracts, appointed an arbitrator, and then asked, as a separate matter, whether the seller would accept that the disputes be adjudicated under a single arbitration for efficiency. The seller later argued under section 67 that the notice was defective because it purported to commence a single arbitration in respect of two separate contracts. Sir William Blair rejected the challenge. The court held that, properly construed, the notice was effective to commence arbitrations under both contracts. The fact that the notice later raised the possibility of a single consolidated arbitration did not alter the essential effect of the notice itself. Ameropa is therefore a valuable authority on substance over form, and on the court’s reluctance to invalidate an arbitration for an artificial procedural objection where the commercial meaning of the notice is clear.
The third principle is that, in a two-tier GAFTA structure, the court deadlines for a jurisdiction challenge do not necessarily wait for the arbitral appeal on the merits. That is the significance of PEC Ltd v Asia Golden Rice Co Ltd. PEC concerned an alleged rice purchase contract, issues of authority, and whether the claimant was party to a GAFTA arbitration agreement at all. Andrew Smith J treated the section 67 application as a new hearing of the jurisdiction issue. More importantly for present purposes, the court held that where the GAFTA 125 Rules provide for a merits appeal but not an appeal on jurisdiction, the 28-day period for challenging the first-tier tribunal’s jurisdictional ruling runs from the first-tier award, not from the later Board of Appeal award on other matters. That point is a trap for the unwary and remains one of the most important procedural lessons in commodity arbitration.
The fourth principle is that the scope of a GAFTA arbitration clause can extend to disputes about whether a later contract affected rights under an earlier contract, even where both contracts contain arbitration clauses. The leading recent authority is CAFI–Commodity and Freight Integrators DMCC v GTCS Trading DMCC. The parties entered into a first contract for the sale of Russian milling wheat. When the buyer encountered payment difficulties and the seller purported to terminate for anticipatory breach, the parties entered into a second contract for the same cargo at a lower price. The second contract stated that the first contract was “terminated and considered void”. Both contracts contained materially identical GAFTA arbitration clauses. The seller commenced arbitration under the first contract. The GAFTA First-tier Tribunal concluded that by entering the second contract, the seller had waived its first-contract claim, but the Board of Appeal took the view that it had no jurisdiction to interpret the second contract because the arbitration had been commenced only under the first contract.
Henshaw J rejected that analysis. Construing the transaction as a whole, he held that a dispute as to whether rights under the first contract had been extinguished or waived by the second contract was still a dispute “arising out of or under” the first contract. Rational business parties would not ordinarily be taken, merely by entering a second contract with a similar arbitration clause, to have silently carved out from the first clause a class of disputes concerning the continuing effect of the first contract. CAFI is therefore a major authority on scope, construction, and the interaction of multiple arbitration clauses in related commodity contracts.
The fifth principle is that the court itself decides jurisdiction. That broader proposition is classically established by Dallah Real Estate and Tourism Holding Co v Ministry of Religious Affairs, Government of Pakistan. Dallah was not a GAFTA case, but it remains indispensable in any section 67 analysis. The tribunal had ruled that Pakistan was bound by the arbitration agreement; the Supreme Court held otherwise. The court stressed that, when jurisdiction is challenged, it is for the court to determine whether a valid arbitration agreement exists and binds the party said to be subject to it. In the GAFTA context, Dallah underpins the basic idea that the tribunal can rule on, but cannot finally determine its own substantive jurisdiction against a party who denies ever having agreed to arbitrate.
Practical section 67 grounds in GAFTA cases
In practice, the most common section 67 grounds in GAFTA litigation are these. First, there was no valid arbitration agreement because GAFTA terms were never incorporated. Black Sea is the paradigm example. Second, the tribunal lacked jurisdiction because the arbitration was not validly commenced. Ameropa shows how that argument can fail where the notice is commercially sufficient, but it remains a common line of attack. Third, the dispute or issue fell outside the scope of the clause or the reference. CAFI illustrates the modern approach to scope in the context of related contracts. Fourth, the wrong party is said to be bound, or a signatory/authority issue arises. PEC illustrates that pattern. Fifth, in a two-tier scheme, the challenge may be time-barred if it is not brought at the correct stage. PEC again is central.
Section 68: challenge for serious irregularity
The legal framework
Section 68 is not a merits appeal. It is a safety-net provision intended to protect the integrity of the arbitral process in cases of serious procedural failure that cause substantial injustice. The statutory heads include, among other things, failure by the tribunal to comply with its general duty under section 33, excess of powers, failure to conduct the proceedings in accordance with the agreed procedure, failure to deal with all the issues put to the tribunal, uncertainty or ambiguity, fraud, and contrary public policy.
The orthodox starting point is Lesotho Highlands Development Authority v Impregilo SpA. There, the House of Lords stressed that section 68 must be construed as a narrow remedial provision consistent with the finality aims of the 1996 Act. It is not to be used as a back door appeal on fact or law.
Key section 68 principles illustrated by case law
The first principle is that section 68 is concerned with process, not correctness. Lesotho remains the leading authority for that proposition. The mere fact that a tribunal has made an error of law or fact does not establish serious irregularity. Something must have gone wrong in the arbitral process itself.
The second principle is that a failure to deal with an issue can amount to serious irregularity where the issue is central and where substantial injustice results. The strongest recent GAFTA example is CAFI. The Board of Appeal had taken the view that it lacked jurisdiction to interpret the second contract, but it nevertheless went on to hold the buyer liable under the first contract and to reject the waiver defence. Henshaw J held that this involved, at the very least, a serious irregularity. Before the Board, the buyer had argued that the second contract extinguished the seller’s first-contract rights, and the seller had argued that the second contract was irrelevant. That crystallised an issue which the Board had to determine. It was not open to the Board to decide liability under the first contract while declining to interpret the very contract that directly addressed whether the first-contract claim had been waived. For a GAFTA practitioner, CAFI is a model authority on section 68(2)(d), namely failure to deal with all the issues that were put to the tribunal.
The third principle is that section 68 may be engaged where a tribunal decides a case on a basis that the parties have not had a fair opportunity to address. K v A is important here. The case concerned a GAFTA award arising from hacked email payment instructions. The buyer challenged the award under sections 67, 68 and 69. Popplewell J accepted in principle that if a tribunal introduces and relies on a legal route that the parties have not had a fair opportunity to address, section 68 may be the proper route of complaint. On the facts, however, the buyer did not establish a material procedural injustice sufficient to justify setting the award aside on the relevant grounds. K v A is therefore useful both positively and negatively: it shows the type of complaint that may fit section 68, while also illustrating the court’s reluctance to find substantial injustice where the complaint is really about the tribunal’s reasoning rather than the fairness of the process.
The fourth principle is that section 68 will only rarely succeed, but it is not merely theoretical. The modern example outside GAFTA is Federal Republic of Nigeria v Process & Industrial Developments Ltd. Robin Knowles J held that the award had been affected by serious irregularities, including bribery and false evidence, and set it aside. The case is highly exceptional and should not be treated as a template for ordinary commodity disputes, but it remains important because it demonstrates the real function of section 68 as a means of protecting the integrity of the arbitral process where that integrity has been seriously compromised.
The fifth principle is that section 68 is not available merely because a tribunal’s reasoning is sparse, awkward or commercially unpersuasive. The courts continue to read arbitral awards generously and commercially rather than microscopically. That theme can be traced back to Zermalt Holdings SA v Nu-Life Upholstery Repairs Ltd, where Bingham J said that courts do not approach arbitration awards with a meticulous legal eye bent on picking holes. Although Zermalt pre-dates the 1996 Act, its spirit is still reflected in modern section 68 reasoning and in the broader judicial approach to specialist arbitral awards.
Practical section 68 grounds in GAFTA cases
In GAFTA practice, the section 68 grounds most likely to matter are these. First, failure to deal with a central issue, as in CAFI. Second, deciding the case on a basis not argued, which K v A shows may be relevant but hard to establish in a way that satisfies the statutory threshold. Third, procedural unfairness in the conduct of the hearing or the treatment of evidence. Fourth, excess of powers, where the tribunal decides something outside the submission or purports to make an order it had no power to make. Fifth, extreme cases of fraud or public policy, though these are rare in ordinary GAFTA litigation.
The practical point is that section 68 should be pleaded with discipline. It is often tempting to add a section 68 ground whenever a party is unhappy with the award. That is usually unwise. Unless the complaint really concerns the fairness or integrity of the arbitral process and can be shown to have caused substantial injustice, section 68 tends to distract from stronger section 67 or section 69 points.
Section 69: appeal on a point of law
The legal framework
Section 69 allows an appeal to the court on a question of English law arising out of an award, unless the parties have agreed otherwise. It is therefore an opt-out provision, unlike sections 67 and 68, which are mandatory protections. In some institutional contexts, the right is excluded expressly or by adopting rules that waive recourse. GAFTA contracts do not automatically exclude section 69 merely by providing that those awards are final and binding. Clear exclusion language is required.
Section 69 is subject to permission unless all parties consent. The applicant must satisfy the court that the question will substantially affect the rights of one or more parties, that the question is one the tribunal was asked to determine, that based on the findings of fact in the award the decision is either obviously wrong or, if the question is of general public importance, at least open to serious doubt, and that it is just and proper in all the circumstances for the court to determine the question.
Each element matters. Most section 69 applications fail not because the legal point is wholly fanciful, but because the point is not purely legal, was not squarely before the tribunal, or cannot satisfy the “obviously wrong” threshold.
Key section 69 principles illustrated by case law
The first principle is that the appeal must involve a true question of law, not a disguised challenge to factual findings. One of the clearest recent warnings on that distinction came in Laysun Service Co Ltd v Del Monte International GmbH, where the alleged “questions of law” were in truth findings of fact or evaluative conclusions built on the tribunal’s factual assessment. The point is especially important in GAFTA disputes, where the factual matrix can be dense and commercially nuanced. A party dissatisfied with the tribunal’s assessment of communications, market practice or witness credibility cannot repackage that dissatisfaction as a point of law.
The second principle is that the question must have been one the tribunal was asked to determine. Sharp Corp Ltd v Viterra BV is now the leading authority on that requirement. The case arose out of two GAFTA Appeal Board awards concerning damages under the GAFTA Default Clause after the buyers defaulted when the goods had already been discharged and warehoused at Mundra. The Supreme Court held that the Court of Appeal had fallen into error by deciding the appeal on the basis that the contracts had been varied, even though that was not a point fairly and squarely before the arbitral tribunal and required a factual finding not made by the tribunal. The Supreme Court emphasised that a section 69 appeal cannot be used to decide a different legal question from that argued in the arbitration, nor may the court invent new findings of fact unless they inevitably follow from those already made. Sharp is therefore essential not only on GAFTA damages but also on the proper appellate discipline under section 69.
The third principle is that the point must substantially affect the parties’ rights. This threshold is sometimes overlooked, but it is real. In Shaw v MFP Foundation Pilings Ltd, the court was prepared to assume that there might be an arguable point of law, but the amount at stake was so small that the answer would not substantially affect the parties’ rights. More recently, Allseeds Switzerland SA v Intergrain SA shows that the “substantial effect” requirement can still be considered at the substantive hearing, even where permission has already been granted. In Allseeds, Butcher J observed that the court will not lightly revisit the components of the permission threshold at the merits stage, absent highly unusual circumstances, but the “substantially affect” requirement may still sometimes require scrutiny.
The fourth principle is that the “obviously wrong” standard is stringent. The classic modern judicial explanations are found in HMV UK Ltd v Propinvest Friar Ltd Partnership and Merthyr (South Wales) Ltd v Cwmbargoed Estates Ltd. Arden LJ in HMV suggested that “obviously wrong” connotes something unarguable, illogical, or lacking any reasonable explanation. In Merthyr, the court stressed that the obviousness of the error should be demonstrable from the award itself and should not require the reader to plough through masses of evidence and complex factual material. Those formulations are not commodity-specific, but they are regularly cited in section 69 cases and accurately capture why so few appeals succeed.
The fifth principle is that section 69 appeals proceed on the facts found in the award. The court cannot rehear evidence. That principle is an old one, but Sharp re-states it in modern and particularly important terms for GAFTA cases. It is also reflected in the procedural rules, which tightly restrict the arbitration documents that may be put before the court.
Major GAFTA and commodity section 69 authorities
The first major modern GAFTA authority is Sharp Corp Ltd v Viterra BV itself. The dispute concerned the proper measure of damages under the GAFTA Default Clause where the goods had already arrived at destination, been discharged, warehoused and customs-cleared at the time of default. The Supreme Court held that damages were to be assessed by reference to the market in which it would be reasonable for the seller to sell the contract goods at the date of default, applying ordinary compensatory principles and the mitigation principle. It also held that the appellate court must keep strictly within the section 69 framework. Sharp, therefore, matters both substantively and procedurally.
The second major authority is Trans Trade RK SA v State Food and Grain Corporation of Ukraine. The seller sold Ukrainian feed corn under three FOB contracts requiring payment on a cash-against-documents basis and reserving title until payment. The GAFTA Board of Appeal held that the seller could sue for the price under section 49(2) of the Sale of Goods Act 1979 because the price was payable on a day certain, irrespective of delivery. Andrew Baker J disagreed. He held that although the contracts contained a latest date for payment, the buyer’s obligation to pay remained conditional on the seller’s documentary-delivery performance; accordingly, the price was not payable “irrespective of delivery” within section 49(2). The seller’s proper remedy was therefore damages, not an action for the price. Trans Trade is a major recent section 69 authority because it shows the court intervening on a pure statutory-construction point arising out of a GAFTA award.
The third key GAFTA authority is K v A. The case arose out of the fraudulent interception of payment instructions. The seller sold Romanian sunflower meal under a GAFTA contract requiring payment into the seller’s bank within two banking days against documents. The buyer paid into a fraudulent account and argued that the payment to the bank sufficed. Popplewell J rejected that construction, holding that a contractual obligation to pay into the seller’s bank account was not simply an obligation to pay the bank irrespective of the identity of the destination account. The award, therefore, stood on the main section 69 issue. K v A is valuable because it shows the court engaging seriously with a narrow legal point arising out of a highly fact-specific trade dispute.
The fourth authority is Allseeds Switzerland SA v Intergrain SA. Although arising under FOSFA rather than GAFTA, it is plainly relevant to commodity arbitration generally. The dispute concerned a CIF soybean contract, damaged cargo, and whether the sellers had complied with their insurance obligations when cover was procured only after the vessel’s arrival and without proper disclosure. Butcher J’s judgment is significant for two reasons. First, it is a recent commodity-law illustration of a section 69 appeal on a pure legal question concerning CIF obligations and the legal effect of an insurer’s rejection of cover. Second, it contains a useful discussion about the extent to which issues decided at the permission stage should be reopened at the substantive appeal hearing.
The fifth authority, though not GAFTA-specific, is Bunge SA v Nidera BV. The case arose out of a GAFTA default-clause dispute and reached the Supreme Court. Although not framed solely as a first-instance section 69 decision, it is central to any article on court appeals from GAFTA awards because it shows the appellate courts correcting the legal analysis of damages under the standard GAFTA clauses and re-emphasising the compensatory principle. In Bunge, the Court held that where an export prohibition would in any event have prevented lawful performance shortly after the anticipatory repudiation, the recoverable loss was merely nominal. The case, therefore, remains a cornerstone authority on GAFTA damages and the kind of legal issue that can justify appellate correction.
Practical section 69 grounds in GAFTA cases
In GAFTA practice, section 69 points most often arise from contract construction, the legal effect of market-standard clauses, the measure of damages under standard forms, statutory rights and remedies under the Sale of Goods Act 1979, and the legal consequences of later contractual arrangements. Sharp, Trans Trade, K v A and Bunge are each examples of those categories.
However, section 69 is not a general licence to seek “a better answer” from the High Court. The applicant must formulate a clean legal question. The question must come from the award itself. The court must be able to answer it on the facts found by the tribunal. And the error must be sufficiently clear to meet the statutory threshold. Those constraints explain why section 69 survives in English law but remains exceptional in practice.
Section 70: exhaustion, time limits and other procedural requirements
No article on court recourse from GAFTA awards would be complete without a discrete treatment of section 70. In practice, section 70 is where many otherwise arguable challenges fail.
The first requirement is exhaustion of any arbitral process of appeal or review. Section 70(2) requires the applicant to exhaust any available arbitral process of appeal or review and any recourse under section 57 before going to court. In two-tier commodity arbitration, it can be straightforward or subtle depending on the issue. The general rule points to the Board of Appeal award as the operative award. But PEC shows that where the arbitral appeal process does not extend to the jurisdiction issue, a section 67 challenge to the first-tier award may have to be made without waiting for the merits appeal to finish.
The second requirement is timing. Before the 2025 reforms, section 70(3) required an application or appeal to be brought within 28 days of the date of the award or, where there had been an arbitral process of appeal or review, of the date when the appellant was notified of the result of that process. The 2025 Act clarified this by introducing the concept of the “applicable date”, which now expressly includes: notification of the result of an arbitral appeal or review; the date of a material correction or material additional award under section 57; notification of refusal of a material section 57 application; or, in any other case, the date of the award.
The best recent illustration is JSC “Kazan Oil Plant” v Aves Trade DMCC. The claimant sought to challenge a FOSFA Board of Appeal award under section 69. The appeal award existed on one date, but the award was not physically released until the arbitration fees were later paid. Bright J held that where the challenge was to the appeal award itself, and no further arbitral appeal or review existed, time ran from the date of the award and not from the later date on which it was received. The claim, brought 43 days after the award but 28 days after receipt, was therefore out of time. Although Kazan Oil arose under FOSFA rather than GAFTA, it is directly relevant because the reasoning turns on the two-tier commodity arbitration structure and the meaning of section 70.
The third requirement is that a party should consider whether a section 57 route is available and material before launching a court application. Section 57 permits correction of awards and additional awards in certain circumstances. The 2025 amendments matter here because they now make clear that the 28-day clock can run from a material section 57 correction or decision.
The fourth requirement is that the challenge must be procedurally coherent. This includes proper service, proper formulation of the route of challenge, and compliance with CPR Part 62 and the Commercial Court Guide. The courts have shown themselves willing to take a strict approach to procedural failures in arbitration claims, and there is no reason to think GAFTA cases will be treated differently.
The fifth requirement is that parties must preserve objections in time. Sections 31 and 73 can deprive a party of jurisdictional complaints if it takes part in the arbitration without objecting properly. Ameropa is again a good reminder that silence, subsequent participation, and related settlement steps can complicate or undermine later attempts to argue that the arbitration was never validly commenced.
What changed with the Arbitration Act 2025
The Arbitration Act 2025 did not rewrite the whole architecture of recourse from arbitral awards. But it did make several changes of real practical importance. Three matter most for present purposes.
1. Section 69 survived intact in substance
The Law Commission considered whether section 69 should be amended or abolished and concluded that no change was necessary. Parliament adopted that view. The right of appeal on a point of English law, therefore, remains part of English arbitration law unless excluded by the parties. That is important for GAFTA users because standard-form commodity contracts continue to generate recurring legal questions of wider commercial significance. Section 69 remains the mechanism by which those questions can, in limited circumstances, reach the High Court.
2. Section 67 was materially recalibrated
Before the reform, the orthodox position was that a section 67 application, where the tribunal had ruled on jurisdiction, took the form of a full rehearing in court. Dallah and later authorities reflected that position. The 2025 Act changed it. The explanatory notes state that the reform was intended to make section 67 more consistent with the scheme of sections 68 and 69. The Act now contemplates rules of court providing, in appropriate cases, that no new grounds of objection may be raised unless the applicant could not with reasonable diligence have raised them earlier; no new evidence may be adduced unless the applicant could not with reasonable diligence have put it before the tribunal; and evidence heard by the tribunal must not be re-heard by the court.
For GAFTA users, this is a major practical change. Historically, a party could fight jurisdiction before the tribunal and then, if unsuccessful, mount what was effectively a fresh jurisdiction contest in court. That approach is now constrained. The strategy for a jurisdiction objection, therefore, has to be designed earlier and more carefully than before.
3. Sections 32 and 67 are more sharply separated
The 2025 Act also amended section 32 so that the court may not entertain a section 32 application if the tribunal has already ruled on the objection in question. The policy is clear. A party should generally choose whether to seek a court ruling on jurisdiction in advance or whether to allow the tribunal to rule and then, if necessary, challenge under section 67. The law is now less hospitable to parallel or duplicative jurisdiction battles. That matters in GAFTA cases because traders and counsel often face an early strategic choice between challenging jurisdiction immediately and pressing on with the arbitral process.
4. Section 70 now states the 28-day time rules more clearly
The clarificatory amendments to section 70 are not merely cosmetic. In practice, they reduce arguments about when time starts to run in cases involving arbitral appeals and section 57 applications. The problem addressed in Kazan Oil is a good example of the sort of uncertainty that the amended provision seeks to rationalise.
A practical framework for traders and foreign lawyers
For a trader, in-house counsel, or disputes lawyer considering recourse from a GAFTA award, the following sequence is usually the right one.
First, identify the award. Is the operative award a first-tier award or a Board of Appeal award? Second, ask whether the issue is jurisdiction, procedural unfairness, or pure law. Third, check whether any arbitral appeal, review, or section 57 recourse remains to be exhausted. Fourth, calculate the time limit conservatively and immediately. Fifth, formulate the challenge under the correct route and resist the temptation to overload the claim with weak alternative grounds.
If the complaint is that there was no binding GAFTA arbitration agreement, that the clause was never incorporated, that the arbitration was not validly commenced, or that the dispute fell outside the clause, section 67 is usually the correct route. Black Sea, Ameropa, PEC, CAFI and Dallah provide the principal map.
If the complaint is that the tribunal failed to decide a central issue, decided the case on a basis not argued, or otherwise conducted the arbitration unfairly, section 68 may be the correct route. Lesotho, K v A, and CAFI are the principal guideposts.
If the complaint is that the tribunal answered a question of English law wrongly, and the issue was squarely before the tribunal and materially affected the outcome, section 69 may be available. Sharp, Trans Trade, K v A, Allseeds and Bunge are the most useful modern authorities.
In every case, section 70 must be treated as central rather than incidental. Too many arbitration claims are approached as though timing and exhaustion are technicalities that can be sorted out later. The commodity cases show the opposite. Section 70 often determines whether the court will hear the challenge at all.
Conclusion
Appeals and challenges from the GAFTA awards to the High Court are not a single subject but a statutory matrix. The essential discipline is to distinguish clearly between the internal GAFTA appellate process and the court’s supervisory role under the Arbitration Act 1996. Once that distinction is maintained, the doctrinal landscape becomes clearer.
Section 67 is concerned with authority: was there a valid arbitration agreement, was the arbitration properly commenced, and did the tribunal have jurisdiction over the dispute or issue? Black Sea, Ameropa, PEC and CAFI show the kinds of questions that arise in GAFTA practice and the court’s insistence on contractual analysis rather than assumption.
Section 68 is concerned with process: did something go seriously wrong in the conduct of the arbitration, and did it cause substantial injustice? Lesotho remains the governing principle, while K v A and CAFI show what the provision looks like in modern commodity arbitration.
Section 69 is concerned with narrow correction of legal error. It survives because Parliament continues to regard it as a sensible compromise between finality and legal coherence. Sharp, Trans Trade, K v A and Allseeds show that section 69 is alive and important in commodity practice, but only where the point is truly one of law, squarely before the tribunal, and answerable on the tribunal’s own findings of fact.
The Arbitration Act 2025 has not altered those fundamentals. What it has done is sharpen the procedural edges. Section 67 is no longer the same kind of full rehearing mechanism that it used to be. Section 32 now more clearly requires an election. Section 70 states the time rules more clearly. For GAFTA users, the practical message is straightforward. Jurisdiction objections must be considered earlier; challenge strategy must be more disciplined; and the line between arbitral finality and court supervision remains real, but it is still one that the High Court will police where the statute permits, and the case warrants it.
Appendix: selected public links to key authorities and materials
- Arbitration Act 1996 (sections 30, 57, 67, 68, 69 and 70): https://www.legislation.gov.uk/ukpga/1996/23/contents
- Arbitration Act 2025: https://www.legislation.gov.uk/ukpga/2025/4
- Explanatory Notes to the Arbitration Act 2025: https://www.legislation.gov.uk/ukpga/2025/4/pdfs/ukpgaen_20250004_en.pdf
- GAFTA arbitration overview: https://www.gafta.com/Arbitration
- Sharp Corp Ltd v Viterra BV [2024] UKSC 14: https://supremecourt.uk/cases/uksc-2023-0029
- K v A [2019] EWHC 1118 (Comm): https://www.judiciary.uk/wp-content/uploads/2019/05/K-v-A-2019-EWHC-1118-Comm.pdf
- Lesotho Highlands Development Authority v Impregilo SpA [2005] UKHL 43: https://publications.parliament.uk/pa/ld200506/ldjudgmt/jd050630/leso-1.htm
- Dallah Real Estate and Tourism Holding Co v Ministry of Religious Affairs [2010] UKSC 46: https://www.supremecourt.uk/cases/uksc-2009-0165
- Federal Republic of Nigeria v Process & Industrial Developments Ltd [2023] EWHC 2638 (Comm): https://www.judiciary.uk/judgments/the-federal-republic-of-nigeria-v-process-industrial-developments-limited/
