Fraud and Arbitration — Deconstructing the Russell Principles*

Events over the past decade have led to much debate and reflection on an important question at the core of dispute resolution in India — the arbitrability of fraud. In N. Radhakrishnan v. Maestro Engineers¹ (“Maestro”), the Supreme Court had stood in the way of disputes being arbitrated if there was an allegation of fraud. This obviously led to much chagrin as the simple way to avoid a reference was to trump up an allegation of fraud and sidestep the law. Six years later, in 2016, an opportunity presented itself for course correction and the Supreme Court attempted to clear the way in A. Ayyasamy v. A. Paramasivam² (“Ayyasamy”) without explicitly overruling Radhakrishnan¹. In fact, both Sikri and Chandrachud, JJ. separately concluded that it would be overbroad to allow Radhakrishnan¹ to be interpreted to refer to all types of fraud permitting an avoidance of arbitration and that it ought to be limited only to instances of serious fraud which would go to the root of the contract itself.

This clarification in Ayyasamy² would not have been strictly necessary in all cases where there was an allegation of fraud if one traces the beginning of the jurisprudence on the subject which held the ground that the person accused of fraud has the right to redeem himself in a court of law. Conversely, the accuser of fraud could not himself refuse the reference on the grounds of fraud. For the purpose of tracing the roots, turn to the seminal case of Russell v. Russell³ from 1880 (“Russell”, also cited in Ayyasamy²). Since Russell⁴, there have been multiple judgments on the issue in India, most recently — Avitel Post Studioz Ltd. v. HSBC PI Holdings (Mauritius) Ltd.⁵ (“Avitel”), Vidya Drolia v. Durga Trading Corpn.⁶ (“Vidya Drolia”) and N.N. Global Mercantile (P) Ltd. v. Indo Unique Flame Ltd.⁷ (“N.N. Global”). This article attempts to explore the issue of fraud and arbitrability through the lens of Russell⁴ and its application and evolution in the Indian jurisprudence.

THE GENESIS: RUSSELL⁴

In Russell⁴, partner W.A. Russell gave notice to the other partners — Henry Russell and Robert Crawford for the partnership to determine. This was permitted according to the partnership deed. As a counterblast, Henry Russell commenced an action against W.A. Russell alleging fraud, claiming the notice was void and sought an injunction against acting on such notice. W.A. Russell then invoked the arbitration clause of the partnership deed and filed a motion in court for the reference of all disputes to arbitration. Henry Russell objected to arbitration on the ground of fraud by W.A. Russell and requested for the continuance of his court proceedings. Both the issues were heard together by the Court. While refusing to grant an injunction, the Court held⁴:

… It must be an injury, as a rule, to the person charged with fraud to have it published, and I must say that I am by no means satisfied that the mere desire of the person charging the fraud is sufficient reason for the Court refusing to send the case to arbitration.

… that rule ought only to be applied, as a matter of course, without investigating the circumstances, in cases where the person charged with the fraud desires the enquiry to be public.

***

The next question … is … what foundation there is for the charges, … There must be sufficient prima facie evidence of fraud, not conclusive or final evidence, because it is not the trial of the action, but sufficient prima facie evidence.⁸

The two principles

Thus, there were two principles that broadly emerged:

(i) First, on an allegation of fraud, arbitration could be resisted if the person charged with the fraud wanted a public enquiry.

(ii) Second, there should be a prima facie case of fraud irrespective of who was resisting it for the Court to consider refusing a reference to arbitration.

For the sake of convenience, I will call it the Russell⁴ Principle 1 and Principle 2, respectively, though it really is circular for reasons explained below.

Principle 1 does not seem to have gotten its due place in the trajectory of arbitrability of fraud in India as compared to the other principles. The Chancery Division’s reasoning for Principle 1 was that anyone who faced an injury to his reputation had the right to get it cleared in public in open court and by a Jury. Jury trials represented trial by a society of peers where a not guilty verdict would exonerate the shame in a public fashion. Though Section 11 of the Common Law Procedure Act, 1854 of England⁹ provided a wide discretion to the Court to enter or refuse reference to arbitration, the same was understood to be exercised in exceptional cases once the parties had agreed to arbitrate.³ Jury trials were abolished in India in 1973 with the passing of the Code of Criminal Procedure, 1973¹⁰ and now it is the Judge who rules on the evidence. Though still in public domain, the element of peers ruling on a person’s innocence is no longer available. Even so, Principle 1 has been recognised by the Indian Courts even after abolishment of jury trials and change in the laws in England¹¹ until recently¹².

While applying Principle 1, the Indian Courts have not been consistent. Two early High Court judgments which dealt with Principle 1 divergently are Manindra Chandra Nandy v. H.V. Low and Co. Ltd.¹³ and Narsingh Prasad Boobna v. Dhanraj Mills¹⁴. While the Calcutta High Court observed¹³ (and misapplied Principle 1)—

8. … provided a prima facie case of fraud is made out, the action will be allowed to proceed, although it is the party alleging the fraud who desires the public enquiry.¹⁵

The Patna High Court observed¹⁶:

… if the party resisting an application to stay a civil suit is the party making a charge of fraud, different considerations arise as the person charged does not desire trial by a civil court.¹⁶

One of the earlier Supreme Court decisions dealing with this principle of law was the case of Abdul Kadir Shamsuddin Bubere v. Madhav Prabhakar Oak¹⁷ (“Abdul Kadir”). Here the reference was under Section 20 of the Arbitration Act, 1940¹⁸ — a provision which gave expansive discretionary powers to the Court for the purpose of deciding whether a matter should be referred to arbitration. The Supreme Court recognised the application of Principle 1 to the facts of the case correctly but as the allegations of fraud were held not serious and like in Russell⁴ only based on suspicion, the matter was referred to arbitration.¹⁹

Thus, it is important to note that Principle 1 does not have an independent standing and is subservient to Principle 2. The test of Principle 2 (discussed below) also has to be met for Principle 1 to succeed. This gives rise to a peculiar situation. If there is a serious allegation of fraud, the reference to arbitration may fail (irrespective of who is resisting the reference) as the Court is likely to hold that it cannot be arbitrated. If a sufficient prima facie case of fraud²⁰ is not made out by the accuser, then, even if the accused wishes to resist arbitration, the Court may hold that it is not a serious or sufficient charge and refer it to arbitration. It is for this reason, that it may be interesting to observe how Principle 1 has played out in the Indian courts.

PRINCIPLE 1 — THE STEPCHILD

Of the twenty-nine cases²¹ that were looked at where Principle No. 1 and/or the Russell⁴ judgment were discussed, in sixteen cases, it was the accuser of fraud who attempted to resist arbitration. Of those sixteen cases, eight times the Court refused to refer to arbitration holding that there was a prima facie case of fraud and/or serious fraud. Of the thirteen times the accused resisted arbitration, six times the Court refused to refer the matter to arbitration on the ground that the fact-situation prima facie pointed at serious fraud or that a civil court would be able to appreciate complex and voluminous evidence better than an arbitrator. It appears from these cases that the issue of fraud vexes the person charging the fraud quite often. Perhaps the reason is that they would like the accused to be held accountable in public and hence prefer a court trial. It may also be that if the accuser feels heavy evidence needs to be adduced in support of its claim, then a Court is a better forum. Conversely, it may well be that the person accused of fraud prefers the privacy of an arbitration to avoid public cynosure, especially if they have a weak case. Principle 1 in the long run up to Avitel²² seems to have lost steam and only plays a small supporting role in the reasoning of a Court while declining or entering a reference for arbitration. Even so, Russell⁴ is often cited for Principle 1 in decisions relating to fraud and arbitrability. Though Principle 2 does not trace its inception to Russell⁴, it is reiterated there and forms a strong basis of the verdict of the Chancery Division.

PRINCIPLE 2 — THE CHAMELEON

Principle 2 has gained a life of its own from a prima facie case of fraud in the 1880 judgment to a prima facie case of serious fraud in the Indian context²³. This issue takes on a subjective form which has caused the courts to spend much time trying to define its contours. In Russell⁴, the allegation of fraud was made on affidavit and no details were provided. The Court held the allegation was a—

… statement of belief, sufficient on a motion of this kind to call upon the party for a denial, but useless for any other purpose.²⁴

Thus, in Russell⁴ the Court referred the matter to arbitration because a prima facie case of fraud was not met. In the Indian context, the Courts have oscillated between a prima facie case of fraud and a prima facie case of serious fraud as grounds for refusing arbitration. Though really this difference was illusionary given that there was no objective threshold for what amounted to serious fraud. While Principle 2 came into the spotlight with the judgment of Abdul Kadir¹⁷ in 1962, it is really between 2009 and 2011 that a spate of judgments like Maestro,¹²⁵ Afcons Infrastructure Ltd. v. Cherian Varkey Construction Co. (P) Ltd.²⁶ (“Afcons”) and Booz Allen & Hamilton Inc. v. SBI Home Finance Ltd.²⁷ (“Booz Allen”) cemented the importance of the issue of arbitrability of serious fraud²⁸. In Maestro¹, the Supreme Court accepted that malpractices of finance amounted to a serious allegation of fraud that could not be dealt with by the arbitrator. However, no cogent reason was provided as to why the arbitrator would not be competent enough to deal with the same. In Afcons²⁶, the Supreme Court of India spelled out certain types of cases which it held to be not suitable for the process of alternative dispute resolution. This list included in Entry (iv) cases involving serious and specific allegations of fraud, fabrication of documents, etc. This was followed by Booz Allen²⁷ where it clarified that actions in rem were outside the purview of arbitration. Though these judgments held that serious fraud was not arbitrable, in reality they were and are cited regularly to resist arbitration even in simpler cases.

It took only 6 years for the tide to turn for Principle 2.²⁹ In Ayyasamy², the Supreme Court held that mere allegation of fraud simpliciter may not be enough to nullify an arbitration agreement. It held that a court has to consider what the nature of the dispute is and then strict and meticulous enquiry into the allegations is needed. Only when the court is satisfied that the allegations are serious and of a complicated nature that the reference to arbitration should be refused. The Court clarified that serious fraud would include allegations which make a case of criminal offense³⁰, allegations which are complicated and require appreciation of voluminous evidence, allegations of forgery/fabrication of documents, allegations of fraud against the arbitration agreement itself/fraud permeating the contract. The Court also held that allegations of fraud touching upon the internal affairs of the parties (with no implication on the public domain) could be referred to arbitration. The Supreme Court in Ameet Lalchand Shah v. Rishabh Enterprises³¹ (“Ameet Lalchand”) and Rashid Raza v. Sadaf Akhtar³² (“Rashid Raza”) followed Ayyasamy² and also distinguished between serious and simple fraud. In N.N. Global⁷, the Court has gone as far as stating that the judgment in Russell⁴ is obsolete. Though the Court in N.N. Global⁷ does not distinguish between Principle 1 and Principle 2, the observation has to be perhaps read in light of Principle 2 as the Court prefaces the obsoletion with the observation that allegations of fraud no longer relate to the competence of the arbitrator and voluminous evidence. As Principle 2 is not unidimensional in its invocation, it would be useful to see how allegations of serious fraud have manifested in cases:

Suit for accounts: In Abdul Kadir¹⁷, the allegation was that the accounts were not made up to date, were incomplete and incorrect. On a closer look, the Court observed the allegations were based only on a suspicion. The Court held that such allegations on correctness of entries of accounts are often made in account suits but do not amount to serious fraud. In Maestro¹, the Court while relying on Abdul Kadir¹⁷ for the ratio on serious fraud not being arbitrable, took a different turn on the appreciation of facts. The Court held that malpractice in account books and finances of the partnership could not be dealt with by an arbitrator. The position seems to have turned a full circle in Ayyasamy² and Avitel²². In Ayyasamy², the issue of signing a cheque without the consent of other partners was held to be an accounts issue and not one of serious fraud or involving any complex issue³³. In Avitel²², the Supreme Court on a prima facie finding, allowed the relief of a full deposit in a Section 9 of the Arbitration & Conciliation Act, 1996 (“the Act”) proceeding to safeguard the pending enforcement proceedings under Section 48 of the Act. This was despite the objection of the respondent that the foreign award would not be enforceable under Indian laws as there were serious allegations of fraud including siphoning of funds. Thus, it can be said that inter-party accounting irregularities and siphoning of funds can be dealt with in arbitration.

Voluminous evidence better suited for Court: The Court in Maestro³⁴ accepted the argument that when detailed material evidence (both documentary and oral) is required to prove malpractice, then the arbitrator would not be a competent authority. Even the Court in Ayyasamy² observed that there could be complex cases which could only be tried by the Court due to voluminous evidence. This factor has caused much heartburn, as the question is often raised : Why cannot an Arbitral Tribunal conduct a commercial trial even if the matter is complicated? There seems to be little clarity on what matters may require such complex voluminous evidence and why a Tribunal cannot deal with the same, especially after the enactment of Section 26 and Section 27 of the Act. The international arbitration space is thriving with complex private disputes³⁵. The old perception of ADR³⁶ still manifests in contemporary Indian arbitral jurisprudence despite the enactment of a modern statute (the Act) and its consequent amendments. The Supreme Court in N.N. Global⁷ has called out on this issue while observing that

96. … Arbitrability of fraud is no longer an issue relating to the competence of the arbitrator, or dealing with voluminous evidence. …

***

116. … In contemporary arbitration practice, arbitral tribunals are required to traverse through volumes of material in various kinds of disputes such as oil, natural gas, construction, industry, etc.³⁷

Public domain: If the allegation at hand, has a public element, it is unlikely the same can be arbitrated as a private dispute. In World Sport Group (Mauritius) Ltd. v. MSM Satellite (Singapore) Pte. Ltd.³⁸, Part II of the Act was under consideration. The Supreme Court held that consideration of fraud was not relevant for the purposes of Section 45. Even so, the impugned order of the Bombay High Court made³⁹ an interesting observation on the issue of arbitrability when it touches the public domain. It observed that when the issues of fraud have a bearing on the position of BCCI in the game of cricket, the involvement of the general public in the game and the television rights which are conferred for viewing the games by public, and also presence of BCCI being necessary, then, the matter is fit for an open public trial by a court of law. Thus, even an indirect public element may lead to a stay on arbitration. On the other hand, in Rashid Raza⁴⁰, the Supreme Court held that the affairs of partnership and siphoning of funds do not concern the public domain and the dispute was arbitrable. Taking this further, the Court in N.N. Global⁷ held that allegations of fraud with respect to the invocation of bank guarantee are arbitrable as it arises from disputes between parties inter se and is not in the realm of public law. It would be interesting to see the same Court’s observations below in the case of criminal matters which it holds to be in the public realm.

Effect of parallel criminal proceedings: An interesting discussion in Avitel²² is about the effect a criminal proceeding has on a reference for arbitration. As the standard of proof is different in civil and criminal proceedings, the mere filing of a criminal case would not amount to the allegations of fraud being upscaled to serious in the civil proceedings. While the concept of finding of facts in civil and criminal matters not having a bearing on each other is trite law, it has been expounded in a crystal-clear manner in the context of reference to arbitration matters in Avitel²². Thus, both Afcons²⁶ and Booz Allen²⁷ may have to henceforth be read with a rider that that merely because criminal proceedings can be or have been instituted in respect of the same subject-matter, it would not mean that a dispute which is otherwise arbitrable, ceases to be so.⁴¹ However, it is to be kept in mind that the recent case of N.N. Global⁷ (a three-Judge Bench) after noting the observations of Ayyasamy² (a two-Judge Bench), Vidya Drolia⁶ (a three-Judge Bench) and Avitel²² (a two-Judge Bench), and without departing from them, states that the criminal aspect of fraud, forgery or fabrication which have penal consequences and criminal sanctions can be adjudicated only by a court of law since it may result in a conviction, which is in the realm of public law. To this extent it would appear there is perhaps some pull back from Avitel²² if the observations were read in isolation without comprehending the context of criminal law. The matter though seems settled if one turns to Vidya Drolia⁶ (also a three-Judge Bench) where the Court has concurred with the Avitel²² view that Maestro¹ “has no legs to stand on”⁴² and then proceeded to overrule it, stating that allegations of fraud can be the subject-matter of an arbitration if they relate to a civil dispute, leaving only the caveat that “fraud, which would vitiate and invalidate the arbitration clause, is an aspect relating to non arbitrability”⁴³.

THE WAY AHEAD

While the above factors are not exhaustive and the outcome of each case will depend on its facts, they do provide a strong indication that courts are steadfast in permitting commercial fraud to be arbitrated. It goes without saying that any serious allegations of fraud cannot be based on mere suspicion for the court to consider refusing arbitration. In Avitel²², the Supreme Court while relying on Ayyasamy², has tried to distill the various factors into a two-prong test for meeting the threshold of a serious allegation of fraud⁴⁴. The first test is satisfied when the arbitration clause itself cannot be said to exist in a clear case in which the Court finds that the party against whom breach is alleged, cannot have been said to have entered into the agreement at all. The second test is where allegations are made against the State or its instrumentalities of arbitrary and fraudulent conduct which would require adjudication in a writ court.⁴⁴ Though the intention to streamline the threshold for fraud by the two-prong test is well meaning, it is quite narrow. It is possible that litigants will rely on Ayyasamy² to keep the door of serious fraud open a little wider. They will have further assistance in N.N. Global⁷ which observes that fraud not being arbitrable is a wholly archaic view.

Russell⁴ to Avitel²² and N.N. Global⁷, has been a long journey for arbitrability of fraud. While Indian courts have relied on the principles propounded in Russell⁴ to grapple with questions of fraud, the Supreme Court’s tests laid down in Ayyasamy² and Avitel²² have attempted to forge touchstones for these principles. This will hopefully bring in some predictability on how courts will look at the issue of fraud and arbitration and is a progressive step towards strengthening the arbitration regime in India. This is also in keeping with the international trend of private resolution for inter-party disputes. In a country like India, where the case pendency is a staggering 3,78,96,456 cases⁴⁵, and bound to get worse due to the Covid-19 Pandemic, the importance of ADR cannot be overemphasised. All efforts should be made by the courts to give ADR the impetus needed to thrive.

“Why should not the childish quarrels of princes be settled through the arbitration of these learned men.” — Desiderius Erasmus…. sometime in the 1500s.⁴⁶

---