Partner Sydney
"The judgment has immediate implications for airlines, ground handlers and cargo operators globally, particularly those relying on the standard form SGHA without bespoke jurisdiction clauses."
The judgment has immediate implications for airlines, ground handlers and cargo operators globally, particularly those relying on the standard form SGHA without bespoke jurisdiction clauses.
The SGHA and its commercial function
The SGHA has, for decades, served as the aviation industry’s standard-form template governing the relationship between airlines and ground service providers. Its appeal lies in its neutrality and efficiency: it provides a default contractual framework, reducing the need for extensive negotiation. It also enables parties in different jurisdictions to transact on broadly familiar terms.
Article 9 of the SGHA contains an ostensibly flexible dispute resolution clause that encourages the parties to seek an amicable resolution in the first instance and failing that, expressly provides that “the Parties may elect to resolve the dispute through arbitration“. The final fallback is for the national courts specified in Annex B to have jurisdiction over the dispute.
For many users of the SGHA, Article 9 has been assumed to offer a pathway to arbitration, with the flexibility to elect that process where appropriate. The decision in Dnata demonstrates that, at least under Australian law, that assumption is misplaced.
The dispute and the attempted “election” of arbitration
The underlying dispute arose from a workplace injury sustained by a Dnata employee while unloading cargo from a Polar Air-owned Boeing 747 freighter aircraft in Sydney. The injured employee brought proceedings in negligence against the airline, which in turn advanced cross-claims against Dnata under the SGHA. In response, Dnata sought to stay those cross-claims in favour of arbitration, relying on Article 9 of the SGHA, Article 7(2) of the International Arbitration Act 1974 (Cth) and Article 8 of the 1985 UNCITRAL Model Law on International Commercial Arbitration.
The key issue was whether Article 9 constituted an “arbitration agreement” within the meaning of the International Arbitration Act 1974 (Cth). Dnata contended that it did and that its unilateral “election” to arbitrate was sufficient to engage the statutory stay. Polar Air argued that arbitration under Article 9 required the consent of both parties and could not be imposed unilaterally.
Construction of Article 9: the central issue
The NSW Court of Appeal unanimously dismissed Dnata’s appeal, holding that Article 9 of the SGHA does not create a binding arbitration agreement.
The starting point was the language of the clause itself. The phrase “the parties may elect to resolve the dispute through arbitration” was held to require a joint election, not a unilateral choice. The court emphasised that “the parties” means both parties acting together, not either party acting independently. This conclusion was reinforced by the subsequent sentence, which contemplates the possibility that “the parties fail to agree to an arbitration process,” in which case the dispute is to be determined by the courts. That express fallback to litigation, in the absence of agreement, was in the court’s mind inconsistent with any binding or unilateral arbitration mechanism.
The court also considered the absence of essential arbitral mechanics. Article 9 does not specify a seat, governing rules or even the number of arbitrators with any precision. While such omissions are not always fatal and do not automatically result in arbitration agreements being rendered void, the court considered it unlikely that sophisticated commercial parties – including the IATA – intended to create a binding arbitration obligation that would leave these matters entirely open.
Taken together, the court concluded that Article 9 is best understood as an agreement to negotiate or agree on arbitration in the future, rather than a standalone arbitration agreement. As such, the court found that it fell outside the statutory definition required to compel arbitration under the International Arbitration Act 1974.
Dnata has been granted leave to appeal to the NSW Supreme Court. Its Notice of Appeal poses a simple but poignant question: “on its proper construction, does article 9 of the Main Agreement [SGHA] contain an “arbitration agreement” within the meaning of the International Arbitration Act 1974 (Cth)?”. The Supreme Court’s decision is pending as at the date of this article.
Potential misalignment with domestic and comparative case law
"The approach taken by the NSW Court of Appeal in Dnata sits somewhat uneasily with the pro-arbitration stance traditionally adopted in Australian courts and other common law jurisdictions, most notably the English courts."
The approach taken by the NSW Court of Appeal in Dnata sits somewhat uneasily with the pro-arbitration stance traditionally adopted in Australian courts and other common law jurisdictions, most notably the English courts.
For instance, in Comandate Marine Corp v Pan Australia Shipping Pty Ltd [2006] FCAFC 192, the Full Federal Court emphasised that arbitration agreements should be construed in a liberal and commercially sensible manner, even where the drafting is less than precise. Allsop J observed that, in the context of standard form international contracts, “the authorities…are clear that a liberal approach should be taken” and that “this liberal approach is underpinned by the sensible commercial presumption that the parties did not intend the inconvenience of having possible disputes from their transaction heard in two places” (per Allsop J, at 164-165).
That statement reflects a judicial inclination – shared with many jurisdictions across the world – to give effect to the parties’ intention to arbitrate, rather than defeat it by strict contractual interpretation.
A similar divergence can be observed when Dnata is compared with the position under English law. The starting point under English law remains the well-known principle in Fiona Trust & Holding Corporation v Privalov [2007] UKHL 40, under which arbitration agreements are construed broadly on the basis that rational commercial parties are presumed to have intended a ‘one-stop shop’ for the resolution of their disputes. That presumption has frequently led English courts to uphold arbitration clauses despite imprecise or incomplete drafting, on the footing that commercial parties are unlikely to have intended a fragmented resolution of their disputes.
Against that background, the reasoning in Dnata marks a notable divergence. The NSW Court of Appeal declined to infer any binding commitment to arbitrate from language that contemplated an election to arbitrate, notwithstanding that the clause expressly referred to arbitration and formed part of a standard form contract used in global aviation services. In doing so, the court gave decisive weight to the literal requirement that “the parties may elect”, interpreting this as necessitating actual agreement at the point of dispute, rather than recognising the clause as conferring an enforceable procedural option to arbitrate. It is questionable whether such an approach would be followed or applied in the English courts or other common law jurisdictions.
Commercial implications for the aviation sector
For industry participants, the implications are immediate and significant. The SGHA is widely used across jurisdictions, often with minimal amendment.
Parties who have assumed that Article 9 of the SGHA provides a reliable route to arbitration may now find themselves exposed to litigation in national courts, particularly where disputes arise in jurisdictions that are commercially inconvenient – or procedurally incompatible – with one or both parties’ operations, or where, in light of the Dnata decision, there is a nexus to Australia.
For airlines, the decision in Dnata raises the prospect of defending claims in unfamiliar or less favourable forums, potentially undermining expectations of neutrality and procedural efficiency. Unlike arbitration, airlines may also find themselves more exposed to being joined as co-defendants or brought into proceedings as third parties in disputes with which they have no direct contractual nexus. Certain jurisdictions permit the joinder of non-parties more readily in court proceedings than in arbitration, where the ability to involve third parties is typically more constrained under most institutional rules and national arbitration legislation. More broadly, it may dilute the procedural advantages commonly associated with arbitration, including confidentiality, procedural flexibility and the ability to confine disputes to the signatories to the arbitration agreement.
For cargo operators, ground handlers and other stakeholders, the decision increases the risk of fragmented proceedings, particularly where related claims straddle contractual or tortious causes of action. By way of example, questions arise as to how such disputes will now be managed where a claimant in negligence elects to pursue alleged tortfeasors – such as an airline or cargo operator – through court proceedings, while insurers seek to push related indemnity and contractual claims into arbitration. The possibility of parallel proceedings of this kind introduces both procedural complexity and the potential for inconsistent outcomes.
More broadly, the decision highlights the limits of relying on industry-standard drafting in complex, cross-border transactions and sectors such as aviation. The SGHA’s strength – its flexibility and neutrality – becomes a weakness where critical provisions, such as dispute resolution, are left insufficiently defined. In that respect, Dnata serves as a reminder that dispute resolution provisions should not be treated as boilerplate clauses. Airlines, ground handlers, cargo operators and other industry participants may want to review their existing dispute resolution arrangements, particularly where they rely on standard-form agreements such as the SGHA. Parties should consider whether their current clauses accurately reflect their preferred method of dispute resolution and, where arbitration is intended, ensure that the clause creates a clear and binding obligation to arbitrate.
For new contracts, parties may wish to move beyond the standard wording of Article 9 and adopt bespoke dispute resolution provisions that clearly identify whether arbitration is mandatory or optional, together with key procedural features such as the arbitral seat, governing rules and the method of appointing arbitrators. Doing so will help ensure that the dispute resolution mechanism operates as intended and reduce the prospect of courts being required to determine threshold questions of jurisdiction before the substantive dispute can be addressed.
"If arbitration is intended to be mandatory, the contract must say so in equivocal terms and provide a workable mechanism for its application."
Conclusion
Ultimately, the NSW Court of Appeal’s reasoning in Dnata reflects an application of two foundational principles. First, a reminder that arbitration is strictly consensual: a party cannot be compelled to arbitrate absent a clear and binding agreement. Second, contracts must be interpreted according to their plain, express language. Where that language contemplates future agreement and a lack of mutual consensus, the courts will not supply what the parties have omitted.
For practitioners and commercial parties alike, the lesson from Dnata is clear. If arbitration is intended to be mandatory, the contract must say so in equivocal terms and provide a workable mechanism for its application. The NSW Court of Appeal’s view was that the SGHA, in its current form, does not do this. The decision therefore underscores the importance of carefully drafting arbitration agreements at the contracting stage – even if they form part of standard agreements – to ensure that they are applicable to and reflect the parties’ intended commercial objective.
Key contacts
Partner Sydney
Senior Associate Dubai
