Issues Relating to Non-Signatories in International Arbitration: A Comparative Analysis of Three Recent Landmark Cases

Author: Jeffrey, Chein-Yu Long*

I. Introduction

International commercial arbitration fundamentally rests on the principle of consent. The cornerstone doctrine that “arbitration is a creature of contract” emphasizes that parties cannot be compelled to arbitrate disputes if they have not agreed to submit them to arbitration. However, the modern commercial reality presents complex challenges to this foundational principle. In today’s interconnected global economy, commercial transactions frequently involve multiple parties operating through corporate groups, supply chains, and intricate contractual frameworks. These complex structures often lead to situations where entities that never formally signed an arbitration agreement become embroiled in disputes that are subject to arbitration.

The issue of non-signatories in international arbitration has emerged as one of the most pervasive and unsettled debates in the field. The traditional concept of consent for arbitration is difficult to reconcile with the complex commercial reality and the role of non-signatories today. This tension has given rise to various doctrines and legal theories aimed at determining when non-signatories can be bound by or benefit from arbitration agreements. These theories attempt to balance the sanctity of consent with commercial practicality and the needs of efficient dispute resolution. However, different jurisdictions apply these theories with varying degrees of rigor and through distinct legal frameworks, creating a fragmented international landscape.

II. Theoretical Framework

The tension between arbitration’s consent-based foundation and the commercial necessity of binding non-signatories has generated several theoretical frameworks that courts and tribunals apply with varying degrees of acceptance. Understanding these theories is essential for unpacking the reasoning in three focus cases and appreciating the diverse approaches across jurisdictions.

1. Foundational Principles: Consent and Privity

At its core, arbitration derives its legitimacy from party consent. This principle is enshrined in national arbitration laws and international instruments like the New York Convention, which requires an “agreement in writing” for enforcement of arbitral awards. Closely related is the doctrine of privity of contract, which dictates that contractual obligations and rights are limited to the parties who formed the agreement. Together, these principles create a barrier to extending arbitration agreements to non-signatories.

However, strict adherence to these principles can lead to fragmented dispute resolution, inefficient parallel proceedings, and potentially contradictory outcomes—scenarios at odds with arbitration’s promise of effective dispute resolution. This practical reality has prompted courts and tribunals to develop various theories that either find implied consent or provide limited exceptions to the privity doctrine.

2. The Group of Companies Doctrine

Originally developed in French arbitration jurisprudence through landmark cases like Dow Chemical v. Isover Saint Gobain, the Group of Companies doctrine allows extension of arbitration agreements to non-signatory companies within the same corporate group. The doctrine operates on the premise that companies within a group may constitute a single economic reality (une réalité économique unique) despite their separate legal personalities.

Application of this doctrine typically requires establishing: (1) a group structure; (2) active role of the non-signatory in negotiation, performance, or termination of the contract; and (3) mutual intention to bind the non-signatory. The doctrine maintains a theoretical connection to consent by focusing on common intention, but it departs from strict formalism by inferring such intention from conduct and corporate relationships.

3. Equitable Estoppel

Predominantly applied in common law jurisdictions, particularly the United States, equitable estoppel prevents a party from adopting inconsistent positions to the detriment of another. In the arbitration context, it manifests in two main variants: First, signatory estoppel prevents a signatory from avoiding arbitration with a non-signatory when the claims are “inextricably intertwined” with the agreement containing the arbitration clause. Second, non-signatory estoppel allows a non-signatory to compel arbitration when the signatory makes claims that presume the existence of the contract containing the arbitration clause.

Unlike the Group of Companies doctrine, equitable estoppel focuses less on finding implied consent and more on preventing unfairness or contradictory behavior—essentially operating as an exception to the consent requirement rather than a reinterpretation of it.

4. Other Theories: Agency, Alter Ego, and Third-Party Beneficiary

Several additional theories complete the theoretical landscape. Agency principles can bind principals to arbitration agreements signed by authorized agents. The alter ego or veil-piercing doctrine allows arbitration agreements to bind parent companies where corporate separateness has been abused. Third-party beneficiary theory permits intended beneficiaries of contracts to enforce arbitration clauses despite not signing them.

These various theories reflect different approaches to balancing consent with commercial practicality. Some maintain stronger connections to traditional consent (implied agency), while others represent more significant departures (alter ego). This theoretical diversity helps explain the different approaches taken by courts in three focus cases and across jurisdictions more broadly.

III. GE Energy Power Conversion France SAS v. Outokumpu Stainless USA

In 2020, the United States Supreme Court delivered a unanimous decision in GE Energy Power Conversion France SAS v. Outokumpu Stainless USA, LLC that significantly clarified the position of non-signatories in international arbitration under U.S. law. The case addressed a critical question: whether the New York Convention prevents non-signatories from invoking domestic equitable estoppel doctrines to enforce international arbitration agreements.

1. Case Background and Procedural History

The dispute originated from a complex commercial arrangement involving multiple parties. ThyssenKrupp Stainless USA, LLC contracted with F.L. Industries, Inc. (Fives) to construct cold rolling mills at its Alabama plant. Fives subsequently subcontracted with GE Energy to supply motors for the project.

After Outokumpu acquired ThyssenKrupp’s Alabama plant in 2012, the motors allegedly failed, causing significant damage. Outokumpu and its insurers sued GE Energy in Alabama state court, asserting claims for negligence, breach of warranty, and products liability. GE Energy removed the case to federal court and moved to compel arbitration based on the arbitration clauses in the contracts between Outokumpu (as ThyssenKrupp’s successor) and Fives, despite not being a signatory to those contracts.

The District Court granted GE Energy’s motion, finding that GE Energy, as a subcontractor, fell within the contracts’ definition of “Seller,” which included subcontractors. The Eleventh Circuit reversed, holding that the New York Convention requires arbitration agreements to be signed by the parties seeking to enforce them, thereby precluding GE Energy from compelling arbitration through equitable estoppel. This decision conflicted with previous rulings from the First and Fourth Circuits, which had allowed non-signatories to enforce international arbitration agreements through equitable estoppel.

2. The Supreme Court’s Analysis

Justice Thomas, writing for a unanimous Court, reversed the Eleventh Circuit’s decision. The Court focused on the text, structure, and purpose of the New York Convention and its implementation in the Federal Arbitration Act (FAA).

The Court made several key determinations.

First, the New York Convention does not explicitly prohibit the application of domestic doctrines like equitable estoppel. Article II(1) and (2) of the Convention requires contracting states to recognize agreements “in writing” but does not address who may enforce such agreements. Article II(3) directs courts to refer “parties” to arbitration without restricting “parties” to signatories.

Second, the Convention’s silence on non-signatory enforcement permits Chapter 1 of the FAA to fill the gap. Section 208 of the FAA specifies that Chapter 1 applies to actions under Chapter 2 (which implements the Convention) “to the extent that [Chapter 1] is not in conflict with [Chapter 2] or the Convention.” Since there is no conflict with the Convention’s text or purpose, domestic equitable estoppel doctrines remain available in international arbitration.

Third, the Court noted that the Convention’s drafting history and post-ratification understanding support its interpretation. The Court cited examples of other signatory countries that permit non-signatories to enforce arbitration agreements, demonstrating international consistency with the Court’s approach.

3. Implications for International Arbitration Practice in the U.S.

The GE Energy decision has several significant implications.

It resolves a circuit split and brings clarity to the enforcement of international arbitration agreements by non-signatories in the U.S. The decision also aligns U.S. practice with that of other major arbitration jurisdictions, enhancing international consistency and predictability. It reflects the Supreme Court’s consistently pro-arbitration stance, extending it to the international context.

The ruling recognizes that the New York Convention establishes a floor, not a ceiling, for the enforcement of arbitration agreements. Countries may apply domestic doctrines that facilitate arbitration beyond the Convention’s minimum requirements.

However, the Court left unresolved which specific formulation of equitable estoppel applies in this context and whether state or federal law governs. These questions remain for lower courts to address in subsequent cases.

IV. Cox and Kings Ltd. v. SAP India Pvt. Ltd.

The 2023 decision by a five-judge Constitution Bench of the Indian Supreme Court in Cox and Kings Ltd. v. SAP India Pvt. Ltd. represents a watershed moment in Indian arbitration jurisprudence. This landmark ruling definitively settled questions surrounding the Group of Companies Doctrine (GOCD) and its application to non-signatories in arbitration proceedings.

1. Case Background and Procedural History

Cox & Kings Ltd. (C&K), a travel and tourism company, entered into a series of agreements with SAP India Pvt. Ltd. (SAP India) to implement enterprise resource planning (ERP) software. The initial contractual relationship began in 2010 with a Software End User License Agreement, followed by additional agreements in 2015 for implementing the “Hybris Solution” software for an e-commerce platform.

Although the agreements were between C&K and SAP India, the German parent company of SAP India, SAP SE GmbH (SAP SE), became directly involved when technical challenges arose during implementation. SAP SE’s global experts intervened in the project, which nonetheless faced repeated delays. By November 2016, C&K terminated the contract, alleging SAP India’s failure to deliver functional software, and demanded a substantial refund.

When arbitration proceedings commenced, C&K sought to include SAP SE as a party despite it not being a signatory to the agreements containing the arbitration clauses. C&K argued that SAP SE’s direct involvement in the project and its corporate relationship with SAP India justified binding it to the arbitration agreement under the Group of Companies Doctrine.

The legal journey culminated in a reference to a five-judge Constitution Bench to address fundamental questions about the doctrine’s validity under Indian law and the scope of its application.

2. The Articulation of the “Group of Companies” Doctrine

The Constitution Bench affirmed the validity of the Group of Companies Doctrine in Indian law while clarifying its proper scope and application. The Court rooted the doctrine in Sections 2(1)(h) and 7 of the Arbitration and Conciliation Act, 1996, which define “party” and recognize the various forms an arbitration agreement may take.

The Court emphasized that the GOCD is a consent-based doctrine, not a tool to override corporate separateness. Its application requires establishing:

1. The existence of a group of companies’ structure; and
2. The conduct of the signatory and non-signatory parties indicating common intention to bind the non-signatory to the arbitration agreement.

The Court delineated some factors for applying the doctrine:

• Mutual intent of the parties;
• Relationship between the signatory and the non-signatory;
• Commonality of the subject matter;
• Composite nature of the transactions; and
• Performance of the contract.

Most significantly, the Court clarified that participation of the non-signatory in contract performance is the most important factor to consider. This emphasis on actual conduct rather than mere corporate affiliation represented a key refinement of the doctrine.

3. The Indian Supreme Court’s Clarification

The Court’s reasoning reflected a nuanced understanding of arbitration’s consent-based foundation and modern commercial realities. It rejected both extremes: neither a purely formalistic approach that would insist on signatures nor an expansive view that would treat all corporate affiliates as automatically bound.

The judgment emphasized that the GOCD operates independently of other doctrines like alter ego or veil-piercing, which may also bind non-signatories but on different theoretical grounds. This distinction preserved the integrity of corporate separateness as a general principle while allowing flexibility in appropriate cases.

4. Impact on Arbitration Practice in India

The Cox and Kings decision has profound implications for arbitration practice in India.

It aligns Indian arbitration law with its pro-arbitration goals, enhancing India’s attractiveness as an arbitration venue. By establishing clear parameters for binding non-signatories, the judgment provides much-needed predictability for complex commercial relationships.

The ruling empowers arbitral tribunals to make prima facie determinations on joinder issues, reinforcing the competence-competence principle. Courts at the referral stage are directed to exercise restraint, deferring detailed factual analysis to tribunals.

The decision introduces a high evidentiary threshold for invoking the GOCD, requiring specific pleadings and evidence of the non-signatory’s role. This safeguard prevents misuse while maintaining flexibility for appropriate cases.

V. Dallah Real Estate & Tourism Holding Co. v. Pakistan

The case of Dallah Real Estate and Tourism Holding Company v. Ministry of Religious Affairs and Government of Pakistan presents one of the most notable decisions on non-signatories in international arbitration from the United Kingdom. The UK Supreme Court’s 2010 judgment, alongside a contrasting decision by the Paris Court of Appeal on the same facts, illuminates fundamental differences in how jurisdictions approach the extension of arbitration agreements to non-signatories.

1. Case Background and Procedural History

Dallah, a Saudi Arabian conglomerate, negotiated with the Government of Pakistan’s Ministry of Religious Affairs (MORA) to develop housing for Pakistani pilgrims in Mecca. The parties initially signed a Memorandum of Understanding in 1995. Subsequently, the Pakistani government established a statutory trust called the Awami Hajj Trust through a presidential ordinance specifically to execute the project.

In September 1996, Dallah signed the final Agreement with the Trust, which included an ICC arbitration clause requiring disputes to be resolved in Paris. Following a change in government in Pakistan, the Trust’s enabling ordinance lapsed in December 1996, effectively dissolving the Trust. When disputes arose, MORA issued a termination notice on government letterhead, declaring the agreement “null and void”.

Dallah initiated ICC arbitration against the Pakistani Government in 1998. Pakistan objected to the tribunal’s jurisdiction, arguing it was not a signatory to the arbitration agreement. The tribunal, applying French law, issued a jurisdictional award finding that Pakistan was bound by the arbitration clause through “transnational principles of good faith” given its significant involvement in the transaction. The tribunal ultimately awarded Dallah $20.58 million.

When Dallah sought to enforce the award in England, the enforcement proceedings moved through the English court system, with both the High Court and Court of Appeal refusing enforcement. The case ultimately reached the UK Supreme Court, which unanimously upheld the lower courts’ decisions.

2. The UK Approach to Non-Signatories

The UK Supreme Court’s approach reflected a stringent commitment to consent and corporate formalities in arbitration. The Court confirmed that under Article V(1)(a) of the New York Convention, English courts must conduct a full de novo review of an arbitral tribunal’s jurisdictional findings, rather than showing deference to the tribunal’s determination. This approach prioritizes judicial oversight as a safeguard against enforcing awards against parties who never consented to arbitration.

Applying French law (as the law of the seat), the Court employed the “common intention” test to determine whether Pakistan should be bound by the arbitration agreement despite not signing it. This test required objective evidence that both Dallah and Pakistan intended the latter to be bound by the arbitration agreement.

The Court’s analysis emphasized the deliberate corporate structuring and formal documentation of the transaction. The Court reasoned that the Agreement explicitly identified the Trust, not the Government, as the contracting party.  The inclusion of an assignment clause permitting the Trust to assign rights to the Government indicated the Government was not already a party.

While Dallah contended that the Government was actively involved in the negotiation of the Agreement, supervised the performance of the Agreement and issued the termination notice on its own name, the Court held that the Government’s involvement was simply acting with its role as a guarantor under the Agreement. The Court finally concluded that, the Court should respect separate legal personalities and the contract structure that the parties had chosen in the given factual scenario.

Strikingly, the Paris Court of Appeal reached the opposite conclusion applying the same French law principles. The French court gave greater weight to Pakistan’s involvement throughout the contractual relationship, finding that Pakistan “acted as if the contract were its own” and functioned as “the veritable Pakistani party” to the Agreement. This divergence illustrates how differently courts can apply the same legal standard based on their legal traditions and policy priorities.

3. Implications for International Arbitration Practice in the UK

The Dallah decision confirms the UK’s formalistic approach to arbitration agreements, prioritizing express consent and corporate structures over commercial realities. This position contrasts sharply with the more flexible approaches taken in France and some other jurisdictions.

The decision also reinforces the importance of careful transaction structuring. Parties must anticipate enforcement difficulties by securing explicit guarantees or separate arbitration agreements with all relevant entities, rather than relying on theories of implied consent.

The stark contrast between the UK and French decisions on identical facts demonstrates the significance of enforcement forum selection in international arbitration, creating potential for “arbitration tourism” where parties seek enforcement in jurisdictions most favorable to their position.

VI. Comparative Analysis

The three landmark cases from the United States, India, and the United Kingdom represent distinct approaches to the non-signatory conundrum in international arbitration. Their comparison reveals important jurisdictional differences, procedural variations, and policy tensions that shape the global landscape of arbitration practice.

1. Jurisdictional Differences in Applying Non-Signatory Theories

Each jurisdiction applies different theoretical frameworks to determine when non-signatories may be bound by arbitration agreements, reflecting broader legal traditions and policy priorities.

The United States, as exemplified in GE Energy, adopts a pragmatic, pro-arbitration approach. The Supreme Court sanctioned the use of equitable estoppel to bind non-signatories, focusing on preventing unfairness and inconsistent positions rather than strictly enforcing traditional consent requirements.

India, through Cox and Kings, embraces the Group of Companies doctrine while ensuring it remains rooted in consent principles. The Indian Supreme Court carefully delineated the doctrine’s parameters, requiring evidence of mutual intent inferred from conduct and active participation.

The United Kingdom, as demonstrated in Dallah, takes the most formalistic position, strictly adhering to traditional notions of separation of corporate entities, privity of contract and express consent. The UK Supreme Court rejected applying more flexible doctrines of implied consent, prioritizing corporate formalities and explicit agreement over functional relationships.

These divergent approaches create significant implications for parties structuring multi-entity transactions. The same factual scenario—such as in Dallah—can yield opposite outcomes depending on the enforcement jurisdiction, undermining predictability in international commerce.

2. Procedural Variations in Determining Non-Signatory Issues

Beyond substantive differences, the cases reveal important procedural variations in how non-signatory questions are addressed within arbitral processes.

In the U.S., GE Energy confirms that courts decide arbitrability questions, including whether non-signatories may compel arbitration, unless the parties “clearly and unmistakably” delegate this authority to arbitrators. This approach prioritizes judicial oversight of threshold consent questions.

India’s Cox and Kings decision, by contrast, empowers arbitral tribunals to make detailed factual determinations regarding the joinder of non-signatories. The Court directed that at the referral stage, courts should exercise restraint and make only prima facie assessments, deferring comprehensive analysis to tribunals.

The UK’s Dallah judgment establishes a rigorous de novo review standard for arbitral findings on jurisdiction involving non-signatories. The Supreme Court categorically rejected deference to the tribunal’s jurisdictional determination, holding that courts must independently assess whether a valid arbitration agreement exists between the relevant parties.

These procedural differences affect forum selection strategy, timing of objections, and the allocation of decision-making authority between courts and tribunals—all critical considerations for effective advocacy in international disputes.

3. Enforcement Considerations Across Jurisdictions

The contrasting outcomes in the Dallah case—where French courts enforced an award that English courts refused to recognize—highlight the critical issue of enforcement forum selection when non-signatories are involved.

This divergence creates opportunity for strategic enforcement actions, as parties may forum shop for jurisdictions that take more expansive views of consent or apply more flexible non-signatory theories. It also creates risk of inconsistent judgments and undermines the New York Convention’s goal of promoting uniform treatment of arbitral awards.

The GE Energy decision, by explicitly recognizing that other Convention signatories permit non-signatory enforcement under domestic doctrines, potentially reduces this inconsistency by aligning U.S. practice with international norms. Similarly, Cox and Kings explicitly sought to align Indian practice with global standards to enhance India’s status as an arbitration-friendly jurisdiction.

However, significant tensions remain. The fundamental question of whether non-signatories should be bound by arbitration agreements continues to receive different answers across jurisdictions, reflecting deeper philosophical differences about the nature of consent, the importance of corporate separateness, and the proper balance between formalism and functionality in commercial dispute resolution.

4. Drafting Considerations for Arbitration Agreements

Anticipating non-signatory issues during contract formation can prevent costly disputes later. Practitioners should carefully consider the following points:

• Defining “party” broadly in arbitration clauses to explicitly include or exclude affiliates, subsidiaries, parent companies, and other related entities that may become involved in performance. The contracts in GE Energy defined “Seller” to include subcontractors, which formed a basis (though ultimately not the decisive one) for the District Court’s initial decision.
• Including explicit joinder and consolidation provisions that specify the circumstances under which non-signatories may participate in arbitration proceedings. This approach creates express consent for potential non-signatory involvement.
• For particularly critical relationships, securing separate arbitration agreements or guarantees from potential non-signatory entities. Dallah demonstrates the risks of failing to secure direct commitments from the actual decision-makers behind corporate entities.

5. Litigation Strategy When Representing Non-Signatories

As demonstrated in the above cases, when representing non-signatories in disputes, practitioners should bear in mind the following considerations.

First, a practitioner should assess forum options carefully, recognizing that different jurisdictions apply varying standards for binding non-signatories. The contrasting approaches in Dallah illustrate how forum selection can be outcome-determinative.

Second, a practitioner should develop factual evidence of active participation in contract negotiation, performance, or dispute handling. As Cox and Kings emphasized, participation in contract performance is particularly persuasive evidence of implied consent.

Third, a practitioner should consider whether to challenge tribunal jurisdiction immediately or preserve objections for enforcement proceedings, recognizing the different standards of review applied by courts across jurisdictions.

6. Risk Mitigation in Multi-Party Commercial Relationships

Beyond dispute scenarios, practitioners advising on transaction structures should be aware of the above cases.

When advising a corporate group, a practitioner should help clients evaluate corporate structures carefully when entities operate as economic units but maintain legal separation. The divergence between corporate formality and economic reality creates arbitration risk. Thus, practitioners should:

• Consider jurisdiction-specific risks when structuring international transactions, particularly when involving state entities or special purpose vehicles, as in Dallah.
• Document the involvement and consent of all related entities, creating contemporaneous evidence of common intention that may later support or defeat attempts to bind non-signatories.
• Recognize that complex multi-party arrangements may require multiple, complementary dispute resolution mechanisms rather than relying solely on extending a single arbitration clause to all potential parties.

VII. Conclusion

The treatment of non-signatories in international arbitration reveals fundamental tensions between competing values: consent versus efficiency, formalism versus functionality, and predictability versus flexibility. The three cases examined here demonstrate how different legal systems navigate these tensions, sometimes reaching dramatically different conclusions on similar factual scenarios.

Despite their differences, these decisions collectively suggest an emerging middle ground. Looking forward, the non-signatory jurisprudence will likely continue to evolve toward greater harmonization as courts and tribunals engage in transnational dialogue. The GE Energy Court’s explicit recognition of other countries’ approaches signals this trend toward comparative reasoning in developing domestic doctrines.

For practitioners, the diverse approaches to non-signatories create both risks and opportunities. Careful drafting, strategic forum selection, and thorough documentation of consent can mitigate risks and enhance predictability. Ultimately, while perfect uniformity across jurisdictions remains elusive, a nuanced understanding of different approaches allows practitioners to navigate this complex terrain effectively, balancing the fundamental principle of consent with the practical necessities of modern international commerce.

*Jeffrey, Chien-Yu Long is a senior associate of Formosa Transnational, Attorneys at Law, and graduate of Columbia Law School (LL.M. ’25). Jeffrey’s specialization is in international arbitration, shipping/maritime laws and maritime arbitration.