The Power to Bind: The ‘Group of Companies’ Doctrine and Legal Views on Non-Signatories in India During the Change

Introduction : The decision of the Indian Supreme Court in the case of Cox & Kings Ltd. v. SAP India Pvt. Ltd. supports the idea of a group of enterprises. The court ruled that an arbitration clause can be enforced even against a corporate affiliate that did not sign the agreement, as long as there is evidence of shared purpose and activity. Jurisdiction, burden of evidence, and enforcement remain unresolved, while the decision brings India one step closer to a more equitable arbitration consent process from an economic standpoint. This article will examine the decision’s consequences in view of recent developments in Indian statute law and worldwide arbitration practice.  

The Group-of-Companies Doctrine: Concept and evolution 

The Group-of-companies (GOC) doctrine If the “intention common to all companies involved” was enough, a panel in the 1982 ICC arbitration case Dow Chemical v. Isover saint Gobain could hold affiliates that weren’t directly involved in the arbitration agreement responsible for its terms. Common law nations like the US and UK have largely abandoned the theory in favor of more traditional notions of contracts like agency, assignment, or estoppel, but the French legal system fully embraced it.  This highlights the doctrine’s standing as a contentious subject on an international level. The highest court in India has declared that businesses are a “present reality” and has used precedents from international law to support its position (UNCITRAL model law art.7 and New York convention Art.2) to support an expansive interpretation of who qualified as a “party”. By doing this India can shift towards the more liberal French method, with significant qualifications addressed below.

Legal Basis in Indian Law: Consent and the Arbitration Act

It has long been mandated by Indian law that the arbitration agreement serve as written proof of the parties’ consent. Section 7 of the Arbitration Act of 1996, Paragraph 1 defines an arbitration clause as “an agreement by the parties to submit to arbitration all or any disputes…between them”. Section 7(3)-(4) follows up with writing requirements: the agreement must be reduced to writing (contract, email communication, pleadings). It is pertinent to note that the Act does not require every “party” to an arbitration agreement to be a party to the agreement. Section 2(1)(h) defines “party” as “a party to an arbitration agreement” (and persons claiming through or under such a party), but Cox & Kings (Constitution Bench) held that this definition cannot be construed to permit group-binding by assignment alone.

Rather, the Court interpreted Sections 2 and 7 in the context of Article 2 of the New York Convention, holding that the “definition of ‘parties’…may encompass both signatories and non-signatories” if the Court finds intent. What this means, therefore, is that while the formalities of the arbitration clause (written form, signature by at least one of the signatories) are still in place, the law of India has effectively made consent implicit, such that a non- signatory affiliate that actively participated in the transaction may be considered a “party” through its actions The Court’s analysis elevates consent from a mere formal requirement to a substantive assessment based on conduct. Although this represents a step forward for commercial realism, it clouds the referral stage and jurisdictional certainty.

Key Factors and Evidentiary Burden

The Supreme Court has stressed once again that the GOC concept is flexible and dependent on consent. It found a “test” with several factors.  (derived from ONGC Ltd v. Discovery Enterprises, 2022) to 

assess whether a non-signatory has implied consent:

• Mutual intent of the parties : Did the signatories and the non-signatory intend to be bound by the arbitration provision? Engaging in substantial discourse or bargaining is one indicator.
• Relationship of the non-signatory : Did the signatory and the affiliate have a tight corporate connection or coordinate their businesses? Just because two people are members of the same group doesn’t mean they have to work together on that deal.
• Commonality of subject matter: Are the disputes between the signatory and non-signatory : Parties related to the same transaction or series of transactions? An integrated transaction tends to demonstrate that the affiliates should be bound to one dispute resolution forum.
• Composite nature of transaction: Was the transaction a composite of various contracts  between different group companies, which tends to indicate that to prevent fragmented litigation, all contracts should be subject to one arbitral forum?
• Performance of the contract: Is the non-signatory reaping the benefits of the provision in the contract or has it fulfilled its duties under it? One indicator that the affiliate is likely to be bound by the provision is if the affiliate actively performs. Whoever is requesting the joinder must provide evidence of these factors. Because of the nature of the investigation, it may be necessary to look into the inner workings of the company.

Implication (Evidentiary Burden): Tribunals will have to examine the factual matrix extensively, perhaps even considering evidence of what the board or management knew and approved. This might cloud the distinction between “jurisdiction” and merits. The need to “cumulatively consider” all of these factors indicates a fact-intensive inquiry. Non-signatories must be ready for intrusive discovery into their corporate affairs. For parties that are signatories, this implies that any form of cooperation between the parties and their affiliates in the course of an agreement may be reviewed for implicit consent at a later date. In any case, the matter will depend on the facts of the case, emphasizing that the doctrine is not a general principle but a discretionary measure.

Jurisdictional Dynamics: Courts vs. Tribunals : The decision of Cox & Kings (and subsequently the decision in Adavya Projects v. Vishal Structurals in 2025) makes it clear that the challenge to the status of a non-signatory is a matter of jurisdiction under Sections 8/11 of the Act. In the referral phase (judicial gateway), the courts are normally required to conduct only a prima facie assessment of the arbitration clause, leaving the rest of the disputes to the tribunal to decide. The Constitution Bench, however, actually held that the courts should not litigate fully the role of the non-signatory on a Section 8 reference. The courts are only required to assess if an arbitration agreement arguably exists, leaving all doubts to the tribunal to finally decide.

However, the “tribunal-first” approach, according to critics, undermines the rights of non-signatories. As one author puts it, “requiring an affiliate to arbitrate without even a prima facie court determination of its status may be ‘a step too far.’” In India, after the constitution of the tribunal, the non-signatory loses direct access to courts on jurisdictional issues, and it can only contest jurisdiction after the award. The tribunal-first model is restrictive on judicial review at the referral stage. A non-signatory can thus be required to arbitrate and can only challenge jurisdiction effectively after the award has been issued.

Implication (Jurisdictional Risk) : The deferment of complex questions to the tribunal means that the bar to “bind” a non-signatory is actually the risk of surviving an entire arbitration process. This is problematic on several grounds. An affiliate may be reluctant to incur the costs of arbitration, aware that it will have no say on jurisdiction until the end. For claimants, it offers a simplified procedure to seek the joinder of all desired parties. However, for justice, it raises a conflict: must consent to arbitration be assumed so easily that a party waives the usual court protections? The future may hold challenges or legislative guidance on the extent of the referral court’s powers regarding non-signatory claims.

Enforcement Complexity under the New York Convention

The World has Digested the Doctrine of the GOC How open will other courts be towards India treating a non-signatory as part? This applies to both arbitral and enforcement courts. Per Article V, the NYC’s enforcement must be justified by the party in the contest. For these factors to be applicable, they should either be found in an invalid arbitration agreement, or “the said party was not given proper notice… or was otherwise unable to present his case” (Art V(1)(b)). Even if the non-signatory had agreed to arbitration and Indian law governed the contract, the New York or London courts would deal with the matter in accordance with the New York Law or the London Law as the case may be. The claimants would gain by a simplified procedure in which all parties are included. But isn’t this in tension with fairness? Shouldn’t a party just be permitted to waive its usual court protections by consenting to arbitration instead? The competence of referral courts on issues involving non-signatories to the agreement may be limited by future challenges or legislative direction. 

Implication (Enforcement and Interim Relief): This is a catch-22. A bona fide affiliate may have pressing needs (preserving value, halting damage) but be left in limbo until the arbitration is finished. Further, an emergency arbitrator (before the tribunal is constituted) would likely be unable to provide relief to a non-signatory, under current interpretation, since the status of the non-signatory is in doubt. Parties must plan accordingly. If an affiliate may challenge arbitration, it may be useful to include direct interim relief clauses or multi-party notices. Otherwise, however, affiliates experience a temporary “blackout” phase with no remedy available. On the bright side, this provision avoids the issuance of several contradictory injunctions: it provides that only after reaching a consensus can an affiliate seek relief from courts, thus avoiding interference at an early stage.

This is where the problem lies: states are highly disparate in their application of the doctrine. The United Kingdom, as already mentioned, “totally denie[d] the doctrine” in the case of Halliburton v. Chubb. Under article II(1) of the NYC, an arbitral award is binding on a subsidiary, but an English court of enforcement might simply hold that there was no arbitration agreement between the parties. Likewise, most of the American governments would be loath to bind non-signatories to the agreement.  In contrast, France and Brazil have been receptive to group arrangements; they would be more likely to enforce such awards. The lack of However, uniform international recognition leads to enforcement difficulties. An award that is binding on a non-signatory under Indian law may encounter difficulties in enforcement in countries that require strict contractual privity under Article II of the Convention.

Added Analytic Note: Where parties agree that Indian law should govern yet choose an impartial arbitration seat (such as Singapore or London), Indian courts may apply the group-of-companies concept to determine who has jurisdiction. It may be difficult for the international arbitration court to enforce its ruling if it is bound by Indian law and other states do not permit the joinder of non-signatories. If no legislation is named to regulate the arbitration agreement, enforcement jurisdiction courts will make an informed estimate (and sometimes add their own conflict-of-law considerations) in determining which law would apply. Anyone who has read K&K knows that the jurisdictions likely to implement an agreement, the laws that will govern it, and the arbitration seat must now be in harmony. 

Drafting and Commercial Consequences

Contract drafting must now change in response to these new trends

• Identify Parties Precisely: List all entities to be bound by the arbitration agreement. If it is intended to bind affiliates, state this explicitly (or, better, list them). Otherwise, exclude them by name or category.
• Be Wary of Broad Terms: Do not use phrases such as “parties and their subsidiaries” unless so intended. The court will now examine casual references. A “group” clause may inadvertently bind an unwanted affiliate.
• Coordinate Governing Law and Seat: Selecting an arbitration seat or law that is not controlled by Indian law may be appropriate if the underlying contract is subject to Indian law and the arbitration agreement is to bind affiliates. Unless otherwise decided, the arbitrators may follow the law of the contract, which is Indian law in this case. It is critical to ensure that the law authorizes binding affiliates in cases where the legislation of the seat or arbitral tribunal differs; failure to do so may impede enforcement.
• Design Interim Relief Provisions: In view of the uncertainties of non-signatory interim relief, it may be provided in express clauses regarding interim relief (for example, “the arbitral tribunal may provide provisional relief to all group affiliates, regardless of their signatory status”).

Conclusion

The Group-of-Companies idea was explicitly acknowledged in the landmark decision of Cox & Kings, which revolutionized Indian arbitration practice. By shifting the emphasis from mere signature to substantive intent, the theory has brought Indian arbitration law in line with other economically relevant regimes, which is a welcome development. Nevertheless, this extension of arbitral jurisdiction also brings about certain structural problems.

For the Group-of-Companies hypothesis to remain relevant in the future, its use must be careful. Use it sparingly, and arbitration will run more smoothly; use it excessively, and you risk undermining consent, the whole basis of arbitration. The present roadblock in Group-of-Companies theoretical law is control, not recognition. 

Author:- Palak Manik, in case of any queries please contact/write back to us at support@ipandlegalfilings.com or   IP & Legal Filing.