The Supreme Court of India (“Supreme Court”) recently ruled on the arbitrability of fraud in the case of Avitel Post Studioz Ltd. v. HSBC PI Holdings [2020] (“Avitel”). The judgement lays down the tests to determine “serious allegations of fraud” and thereby disputes which cannot be resolved through arbitration.

Various developments in the jurisprudence of the arbitrability of fraud in India have led to diverging legal positions adopted by the judiciary. The Supreme Court in A. Ayyasamy v. A. Paramasivam (“Ayyasamy”) attempted to settle the debate on this issue (as discussed on the blog). Previously, an inconsistency existed between the judgements of N. Radhakrishnan v. Maestro Engineers (“Radhakrishnan”), which held that a case would not be arbitrable if it involved serious allegations of fraud, and Swiss Timing v. Organizing Committee, Commonwealth Games (“Swiss Timing”), which held that cases involving allegations of fraud can be resolved through arbitration. Ayyasamy thus sought to offer clarity by distinguishing between “mere allegations of fraud simplictor” and “serious allegations of fraud”, holding the former to be arbitrable and whereas the latter as non-arbitrable.

Despite the attempt made by the Supreme Court to offer clarity in Ayyasamy, ambiguity persisted in the modus operandi of the substantive law in India qua arbitrability of fraud. Ayyasamy thus gave rise to problems of a hazy and inconsistent interpretation of “serious allegations of fraud” (discussed here on this blog). In particular, it held that “a strict and meticulous enquiry into the allegations of fraud” would be conducted in order to determine whether an allegation of fraud is serious or not. Going against the nature and core principles of arbitration, this was bound to result in unnecessary judicial intervention and delay, if a clear cut formula was not derived to distinguish the two categories of fraud.


Tests for Determining the ‘Fraud Exception’ 

The Arbitration and Conciliation Act 1996 1996 (“Act”) does not specifically exclude any category of cases as non-arbitrable. Section 8 states that the judicial authority has the power to refer the cases to arbitration if a valid arbitration agreement exists. Further, Section 34(2)(b) provides for awards to be set aside if the subject matter is not arbitrable by the law in force. Despite various landmark judgements and recommendations made by the Indian Law Commission (in its 246th Report), no legislative clarity existed. The exceptions to arbitrability of disputes are therefore created by judicial decisions.

Through these decisions (specifically Ayyasamy) it was extrapolated that in Indian law “serious allegations of fraud” are not arbitrable. Thus, the Supreme Court in Avitel focused on determining what would exactly constitute the “serious allegations of fraud” exemption to the arbitrability of disputes. It accordingly laid down that “serious allegations of fraud” would arise only when either of the following two scenarios are satisfied:

  • The first scenario involves an arbitration agreement that cannot be said to exist. This would involve cases in which a party cannot be said to have entered into an agreement to arbitrate at all. Therefore, only in cases where the allegations of fraud are directed towards the ‘agreement to arbitrate’ or are such that if proved, it would vitiate the arbitration clause along with the agreement, would this test apply. Section 16(1) of the Act provides for a wide interpretation of the arbitration clause. It states that the arbitration clause shall be treated as an agreement in itself and the invalidity of the contract shall not entail ipso jure the invalidity of the arbitration clause. Therefore, through Section 16(1) even when the contract is claimed to be fraudulently induced, the substantive validity of the arbitration clause is not compromised. Application of the first test together with this Section 16(1) would considerably reduce the scope of judicial intervention on grounds of fraud. Furthermore, the Supreme Court referred to the work of Gary Born [International Commercial Arbitration by Gary B. Born 2nd Edn., Vol. I, p. 846], and agreed with his rationale that matters in which a party procures an agreement to arbitrate by fraud rarely arise, even in cases where fraud may have actually been committed in connection with the underlying contract. Thus, this test reflects an internationally accepted approach and stance which endorses minimal intervention by the courts. [Para 14 of the judgement]
  • The second scenario is where allegations of arbitrariness, fraud or malafide conduct are made against the “state or its instrumentalities”. Such cases would not be arbitrable as the questions raised would concern matters of public law, thereby attracting implications not only on the parties but concerning the public domain as well. These cases would necessarily be settled in a writ court as issues related to fundamental rights of the people would arise (such as Article 14 of the Indian Constitution). However, with regards to the “public domain” ambit of fraud, the Supreme Court stated that allegations of impersonation, false representations and diversion of funds are all internal affairs of the parties having no “public flavour” and are thus not to be construed as “serious allegations of fraud” so as to attract the fraud exception. Therefore, unless these allegations are made against the state and its authorities, the dispute would be arbitrable and not be adjudged in a court of law. [Para 14 of the judgement]



Avitel in essence deviated from the position taken in Ayyasamy, including inter alia its conformity with Radhakrishnan, and adopted the reasoning given in Swiss Timing instead. It noted that the earlier decisions did not consider the judgement in Hindustan Petroleum Corporation Ltd. v. Pinkcity Midway Petroleums, which stated that under Section 8, it is mandatory for the court to refer the dispute to arbitration if the parties have a valid arbitration agreement. The Supreme Court opined in Avitel that the previous cases failed to take into account a combined reading of Section 5 (limitation on judicial intervention), Section 8 and Section 16 of the Act which reflected a new approach to arbitration. The Supreme Court seemed keen to adopt this approach and thus found Radhakrishnan to be wanting in law.

In Avitel, the Supreme Court sought to clear out multiple anomalies. It addressed the ambiguity arising due to the lack of a specific category of non-arbitrable cases being carved out in the domestic legislation and the decision of the Indian Parliament to not incorporate the recommendation of the Indian Law Commission in this regard. According to the Supreme Court, “the parliament has left it to the courts to work out the fraud exception on a case by case basis”. However, this is worrying as it will lead to a “a case by case” determination of what constitutes fraud and will result in unnecessary judicial intervention and delay. This approach would harm the existing regime of arbitration to a considerable extent.

The Avitel decision is certainly a welcome development after inconsistent interpretations of “serious allegations of fraud”. However, it does not come without its own concerns. The second scenario it presents, where “allegations of such frauds are made against the state or its instrumentalities”, brings to fore certain potential problems as parties can now conveniently raise an allegation or defence of fraud to avoid recourse to arbitration. The test can be misused to avoid arbitral proceedings in cases which involve government sectors, authorities or instrumentalities. India being the second most populous country in the world, there are a large number of contracts which involve the government. Therefore, a blanket application of this test would result in an influx of cases being relegated for adjudication by the courts. This would again lead us back to the initial position of judicial interference by way of closely examining cases involving allegations of fraud to determine arbitrability. Such consequences emerging from this test would thus undermine arbitral principles such as kompetenz-kompetenz and also lead to unwarranted delays which is contrary to the objectives of the Act.



Despite the potential scope for misuse of one of the two tests, the Supreme Court has offered clarity on what makes an allegation a “serious” one. There has been a massive shift from the initial positions adopted by the Supreme Court as it has shed off its interventionist attitude and diverged from its arbitration averse approach. It has essentially adopted a pro-arbitration rhetoric albeit while retaining its protectionist approach, not for reasons of lack of confidence in arbitration, but to secure and reinforce the rights of citizens in the public fora. However, a judicial trend of consistent application of the second test (regarding “states and its instrumentalities”) needs to be developed to negate the use of this test as a pretext to avoid recourse to arbitration. Thus, a pragmatic approach to arbitration needs to be adopted in practice.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration
This page as PDF

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.