Non Signatories in Arbitration

The doctrine of privity of contract posits that only parties to a contract may claim under or be claimed against the contract. [1] Applying this doctrine, only parties to an arbitration agreement can be bound to arbitrate under it. [2] Moreover, consent is the cornerstone of arbitration [3] and the tribunal derives its jurisdiction from such consent. [4] Thus, deviations from these cardinal legal principles can only be rare.

The award of an arbitral tribunal acting under the Arbitral Rules of the International Chamber of Commerce [5] (hereinafter “ICC Rules”), in the Dow Chemicals [6] arbitration, sparked off the debate around what is called the ‘group of companies’ doctrine. The doctrine, in essence, postulates that an entity within a group of companies can be held bound by an arbitration agreement or an arbitration clause contained in a contract to which another member of the group was a signatory, if given the facts of the case, the two entities within the group could be treated as having shared a ‘single economic reality’. The requirements for the application of the doctrine are less demanding than those for the classical ‘lifting of the corporate veil’. However, in the context of arbitration, both the theories seek to disregard the separate juristic personality of individual corporations and hold a corporation to be bound by an arbitration agreement to which it was not a signatory.

Jurisdictions across the globe have taken differing views on these theories. The most notable is the English decision in Peterson Farms [7] that held that the ‘group of companies’ doctrine forms no part of English law. In India, the question of whether a non-signatory can be bound by an agreement to arbitrate is yet to be answered.

This project intends to determine what the Indian courts are likely to do if the question was to be placed before them. In this attempt, the project will be guided by the pronouncements of the Indian courts on the role of consent, the general jurisprudence in India on disregard of corporate form, formal and substantive requirements for a valid agreement to arbitrate under Indian law and decisions from similarly placed countries and certain other jurisdictions.

Chapter 1

Binding Non-Signatories in Arbitration: Approaches in certain Jurisdictions

Several approaches have been taken by arbitral tribunals and national courts to bind non-signatories to an arbitration agreement. [8] The focus, here, is on two such approaches in select jurisdictions of the world.

The Group of Companies Doctrine:

The ‘Group of Companies’ doctrine allows the extension of an arbitration agreement in a contract entered into by a member of a group of companies to other members of the group. [9] It recognises the common economic reality (‘une réalité économique unique’) [10] of a group of companies irrespective of the distinct juridical identity of each of its members. [11]

However, various jurisdictions have given varied interpretations to the requirement for binding non-signatories thus. A comparative view of the judiciary’s view in a few of these countries will assist in better analyzing Indian law.

France

The French courts were among the first to endorse the “group of companies” doctrine. In Dow Chemical, [12] the Paris Court of Appeal confirmed an ICC arbitral award that was, for the first time, principally based on that doctrine. [13] Under the theory as set out by the tribunal and approved by the Court of Appeal, the third party and the signatory must not only belong to the same group of companies, but the third party must also play a role in the negotiation, execution, performance or termination of the contract. Therefore, they must have a common economic reality and it must be the assumed intention of the signatories that the effect of the arbitration agreement is to be extended to the third party. [14] It is important to note that the tribunal did not disregard the requirement for consent, but merely considered that, in the particular circumstances of the case, consent of all the parties was implied. [15]

This doctrine has since been followed by the French courts in a number of decisions. [16]

England

The doctrine has been generally disapproved in England. The earliest such case was Caparo Group Ltd. v. Fagor Arrastate Sociedad Cooperative, [17] which, although casting doubt on the status of the doctrine, confined the decision to the facts of that case. [18]

Thereafter, a question was raised before the English Commercial Court in a proceeding under section 67 of the U.K. Arbitration Act, 1996 as to whether the entities within the claimant’s group in an arbitral proceeding could claim damages although they were not parties to the arbitration. [19]

The respondent claimed damages for losses accruing to its sister companies as well, applying the “group of companies” doctrine. The tribunal held that the arbitration agreement was separable from the main contract, thus Arkansas law, as designated by the parties, was inapplicable and it applied the common intention of the parties. The Court severely criticised the award and held that Arkansas law did, in fact, apply. [20] The court did not exactly reason why the “group of companies” doctrine was not a part of English law; instead, it chose to treat the issue as “one subject to the chosen proper law of the agreement”. [21] Thus, it seems that the English courts would accept the doctrine if it were part of the law governing the agreement, i.e., if the parties had, for example, chosen French law.

U. S. A.

U.S. courts, although applying several theories for binding non-signatories to an arbitration agreement, [22] have expressly rejected the “group of companies” doctrine in the Court of Appeals for the Second Circuit in Sarhank Group v. Oracle Corp. [23] Under U.S. federal arbitration law, companies of the same group will not be bound to an arbitration agreement by their mere implication in the conclusion, performance or termination of the underlying contract. [24] In this case, ostensibly relying on the French doctrine, the arbitral tribunal affirmed jurisdiction over both the parent and subsidiary companies. [25] In reversing the district court’s order and remanding the same, the Second Circuit emphasized that, under the New York Convention, it was for the court to decide whether a party has consented to arbitrate, and stressed on the importance of consent. [26]

Emphasizing the separate legal personality of the respondent, the Second Circuit held that:

“the principal reason corporations form wholly owned foreign subsidiaries is to insulate themselves from liability for the torts and contracts of the subsidiary and from the jurisdiction of foreign courts [and that] [t]he practice of dealing through a subsidiary is entirely appropriate and essential to [the U.S.]’s conduct of foreign trade.” [27]

Critical evaluation of the doctrine

The “group of companies” doctrine has been accepted by countries such as Switzerland, Spain, Brazil and Turkey, while simultaneously rejected by several others. [28]

Examining the approach and reasoning taken by courts in the preceding paragraphs, it appears that the reason for introduction of the doctrine, although may be sound, is in direct conflict with the principle of privity of contract.

Some authors have also argued that the rules developed under this doctrine are too ambiguous and run counter to the clear intention of parties. [29]

English arbitration law requires the parties’ consent to be evidenced in writing. [30] Some authors [31] have submitted that the Court’s dismissal of the “group of companies” doctrine may be following the general trend of English courts respecting party autonomy in international arbitration and preserving privity of contract, [32] placing significant emphasis on the consent of each of the parties to important issues, such as the joinder of non-signatories to arbitral proceedings [33] and the award of damages to such parties. [34]

Lifting the Corporate Veil:

The lack of universal acceptance of the “group of companies” doctrine does not imply that non-signatories can never be bound to an arbitration agreement. A more common form of joining a third party to an arbitration agreement or to any contract is the theory of piercing or lifting the corporate veil. Although in the “group of companies” doctrine, it may be argued that the corporate veil of the group is being lifted to join all members of the group to arbitration, the practice of courts in countries discussed here clearly indicates that these two theories are treated differently.

France

In Orri v Société des Lubrifiants Elf Aquitaine, [35] the French Court of Cassation, confirming a decision of the Paris Court of Appeal, [36] permitted a piercing of the corporate veil and extension of the arbitration clause since the appellant, a party to the arbitration agreement had fraudulently used the corporate veil of several marionette companies to avoid paying his creditors. The Court’s decision was based on the fact that the appellant was the sole decision-maker in the particular association of companies. [37]

This is not the isolated case of lifting the corporate veil in France – several judgments of French courts have done so, both before and after the Orri judgment. [38]

England

English courts have pierced the corporate veil on several occasions, such as when a subsidiary company is used for an illegitimate purpose, [39] where it is used as a mere façade, [40] or in any other instance of fraud, [41] or even when a contracting party acts as agent for another company. [42]

However, English court practice indicates great reluctance in disregarding the separate legal entities of corporate personalities and courts have lifted the corporate veil only in exceptional circumstances. [43] In Roussel-Uclaf v. G.D. Searle, [44] the English High Court had held that a subsidiary claiming the benefit of an arbitration agreement to which it was not a party was entitled to a stay of court proceedings in favour of arbitration. Both defendants in this case applied for a stay of the claimant’s action before the English court under section 1 of the U.K. Arbitration Act, 1975 [45] – Searle (UK) claiming not as a party to the arbitration agreement, but as a party claiming “through or under” Searle (US). [46]  The stay was granted on the basis that the two defendants and their actions were “so closely related on the facts in this case that it would be right to hold that the subsidiary can establish that it is within the purview of the arbitration clause.” [47]  In November, 2008 in the City of London v. Sancheti, [48] the English Court of Appeal overturned the decision in Roussel-Uclaf. The Sancheti decision affirms the restrictive approach of the English courts to determination of the parties to international arbitration agreements and confirms the divide between English and a number of other jurisdictions on the topic. [49]

Relying on Roussel-Uclaf, the respondent in the Sancheti case argued that the City of London was a person claiming “through or under” the United Kingdom (section 82(2) Arbitration Act 1996). The Court of Appeal ruled that Roussel-Uclaf was “wrongly decided on this point and should not be followed.” [50] It held that “a mere legal or commercial connection” to the relevant agreements “is not sufficient” to bind a non-party, claiming “through or under” a party to an arbitration agreement. [51]

However, the Court left unclear what, beyond “a mere legal or commercial connection,” might qualify as claiming “through or under” a party. Thus, the earlier position in English law remains, and a non-signatory to an arbitration agreement may be bound under several circumstances mentioned above.

It may also be possible for a non-signatory party to be bound by an arbitration agreement where the contract in which the arbitration agreement appears confers a substantive right on that party. In the Nisshin case, [52] the Court applied the Contracts (Rights of Third Parties) Act 1999, section 8 of which provides that a party can and must enforce its rights by arbitration if that right is conferred on it by virtue of a contract between two distinct entities that contains an arbitration clause. The case expressly stated that an arbitration clause clearly aimed at a two-party situation, which was not conceived for a situation involving a non-signatory, need not stand in the way of an application of the Contracts (Rights of Third Parties) Act 1999. [53]

Thus, although circumscribed, there is scope under English law for binding a non-signatory to an arbitration agreement as a party to the proceedings.

U.S.A.

In the United States, although no uniform federal law for piercing the corporate veil exists, [54] courts generally do so when a non-signatory corporation controls another corporation and uses it as a signatory for an improper purpose, such as to perpetrate a fraud or to bring about substantial injustice or inequity. [55] Under this theory, the totality of circumstances must be examined in the particular factual situation. [56]

The U.S. Court of Appeals for the Fifth Circuit in the Bridas case concluded that Turkmenistan should be bound to the arbitration agreement because the contracting party, a state-owned enterprise, was the “alter ego” of Turkmenistan. [57] The court applied the established two-step test for piercing the corporate veil which requires dominant control of the shareholder over the company and a misuse of its separate legal personality to commit fraud. [58] The court considered numerous criteria established in prior case law (such as common stock ownership, common directors, inadequate capitalization of the subsidiary), to which it added certain additional factors specifically tailored to state-owned enterprises (such as local legislation, the company’s capacity to sue and to hold property, among others). [59]

To apply the alter ego doctrine to justify the disregard of a corporate entity, the court must determine that there is such unity of interest and ownership that separate personalities of the corporations no longer exist, and that failure to disregard the corporate form would result in fraud or injustice. Flynt Distributing Co. v. Harvey, 734 F.2d 1389, 1393 (9th Cir. 1984); accord FMC Distributing Co. v. Murphree, 632 F.2d 413, 422 (5th Cir. 1980); Kirno Hill Corp. v. Holt, 618 F.2d 982, 985 (2d Cir. 1980). However, a stringent showing is required before a court will pierce the corporate veil. Hidrocaburos y Derivados CA v. Lemos, 453 F.Supp. 160, 172 (S.D.N.Y. 1977). The courts do not lightly disregard the separate existence of related corporations, even in deference to a strong policy favoring arbitration of private commercial disputes. Coastal States Trading, Inc. v. Zenith Navigation SA, 446 F.Supp. 330, 387 (S.D.N.Y. 1977). [60]

More cases and Concluding sentence.

Some Important Considerations

Written form of the arbitration agreement

Many legal systems impose no requirement that agreements to arbitrate must take the form of signed documents. Some countries enforce arbitration agreements made orally. [61] National arbitration statutes often recognize consent memorialized in unsigned written provisions, [62] as does thee UNCITRAL Model Law. [63] The New York Convention covers agreements to arbitrate concluded through unsigned exchanges of letters and telegrams, [64] and some courts have read the Convention to permit arbitral jurisdiction derived from unsigned contracts such as purchase orders. [65]

Most significantly, the fact that a “non-signatory” might be bound to arbitrate does not dispense with the need for an arbitration agreement. Rather, it means only that the agreement takes its binding force through some circumstance other than the formality of signature. The legal framework for normal commercial arbitration (whether statute, treaty or institutional rules) continues to require some assent to arbitrate, whether express, implied or incorporated by reference to other documents or transactions.

Daly – arguments for and against consent

Because arbitration is based on a contractual agreement, each party usually consents to use arbitration as a dispute-resolution mechanism before the dispute arises. [66] Thus, extending arbitration agreements to include nonsignatories may weaken or even destroy this important foundation of the arbitral process. [67]

Despite these arguments, consent has never been an absolute notion in the context of international arbitration. For example, under the LCIA Rules, consent of all of the parties to an arbitration is not required to join a third party to the arbitration; rather, consent may be inferred. [68] Consent is also interpreted broadly by many national courts, which look to honor and validate arbitration agreements. [69] Some federal district courts have even held that parties intended to arbitrate a dispute stemming from an agreement that referred to an arbitral forum that did not exist. [70] Thus, consent has always been a flexible concept in arbitration.

Park – determining consent, p 12

When a non-signatory denies having consented to arbitrate, the very existence of that contract remains at the heart of the parties’ dispute. An arbitrator whose decision rests on a single version of contested facts (the assertion that “X” agreed to arbitrate) would be open to the charge of having engaged in a circular exercise, presuming the very fact that remains open for determination and starting from the contested conclusion whose truth must be evaluated.

Brinsmead – suggestions for amendment to UNCITRAL ML

Perhaps acknowledging the controversy surrounding this issue, UNCITRAL has recently suggested the following possible amendments to the Model Law:

The Swiss Rules, for instance, expressly provide, under Article 4, paragraph (2) that: “Where a third party requests to participate in arbitral proceedings already pending under these Rules or where a party to arbitral proceedings under these Rules intends to cause a third party to participate in the arbitration, the arbitral tribunal shall decide on such request, after consulting with all parties, taking into account all circumstances it deems relevant and applicable”.

The Working Group might wish to consider whether an express provision on third party intervention should be included in any revised version of the UNCITRAL Rules. [71]

This formulation would not require the application of the law of any particular jurisdiction to resolve this question; consequently, its enactment (and incorporation into national laws) would not necessarily clarify some of the main concerns in this area.

Chapter 2

Binding Non-Signatories to Arbitration under Indian Law

Malhotra, 2nd ed., p 182

A person who is not a party to the arbitration agreement is not bound by the agreement. He is neither bound by the award resulting from the agreement, nor is he entitled to enforce it. For instance, in Union of India v Dalmiya Engineering (P.) Ltd., [72] in a contract with the government of India, the trade union of the labour employed by the respondent-contractor, was not recognized as a party to the arbitration agreement. Likewise, in Mulk Raj v Union of India, [73] the High Court of Jammu and Kashmir held that after the competent authority had assessed the compensation, the government of India had no locus standi to question the compensation and reference to arbitration could only be made in the event of disagreement between the persons from whom land had been acquired or requisitioned on the one hand and the competent authority on the other, as they were the only two parties contemplated under law. However, if some measure of privity is established between a stranger and a party to the arbitration agreement, he may be treated as a party to the reference. [74] A person who is not a party to the reference, therefore, is entitled to enforce a charge created by the award in his favour, in terms of a compromise between the parties to the reference. [75]

Arbitration is largely dependent on the consent of parties; however, the objective of arbitration is the speedy and inexpensive disposal of “the matters involved, so that they may not become the subject of future litigation between the parties.” [76] To achieve this objective, some principles of arbitration require to be redefined.

In India, courts so far have generally confined arbitration to parties to the arbitration agreement, whether signatories or not. This traditional approach has been accepted and followed in a long line of cases over a hundred years: Arumugaswamy v Kumara Ettappa Swamy [77] ; Alice Vandepitte v P.A. Insurance Co. [78] ; Tarachand v Syed Abdul Razak [79] ; Patanjal v Rawalpindi Theatre [80] ; G.M. Oriental F&G insurance v Mahendra Prasad [81] ; and Union of India v Dalmiya Engineering [82] .

In National Textile Workers’ Union v P.R. Ramakrishnan, [83] redefining the principle in Re Bradford Navigation Company, [84] a constitutional bench of the Supreme Court held that the workmen of a company had locus standi to be impleaded in winding up proceedings and contest the petition rejecting the statement of the law that it is only the company, the creditors and the contributories who are entitled to appear on the winding up the petition.

But three decisions militate against this traditional and anti-modern approach. In Indian Organic Chemicals Ltd v Chemtex Fibres Inc, [85] Mridul J. was confronted with an issue of string contracts entered into between different entities. Agreement “A” was bet