SPOLIATION OF EVIDENCE AND SETTING ASIDE

One of the issues that was addressed in the Singapore Court of Appeal’s decision of Oni Global Pte Ltd & Anor v GNC Holdings, LLC [2026] SGCA(I) 3 was whether the issue of spoliation of evidence in the arbitration would render the arbitral award in conflict with the public policy of Singapore such that the award should be set aside.

 

Spoliation of Evidence. It was undisputed that the executive vice chairman of one of the parties had deleted text messages which were potentially relevant to the arbitration (Judgment [44]).

However, the arbitral tribunal declined to dismiss the claims made by that party despite finding that it was reasonable to infer that the spoliation of evidence was done “to conceal harmful information and gain an unjust litigation advantage”, finding that there was insufficient basis to conclude that the destroyed evidence would have altered the outcome (Judgment [44]).

The other party sought to set aside the arbitral award on the basis that the destruction of evidence amounted to fraud on the tribunal, which substantially impacted the making of the award (Judgment [44]).

 

Procedural fraud dealt with by the tribunal. This did not find traction with the Court of Appeal. The Court of Appeal made clear that the “great caution possible” should be exercised by a supervisory court before reopening and re-litigating issues of alleged procedural fraud which were fully dealt by with the arbitral tribunal (Judgment [51]).

At Judgment [51(a)], the Court of Appeal stated that the court accords “substantial defence” to a tribunal in the exercise of its procedural discretion, and it extends to even questions of application of Singapore law as the lex arbitri. As such, even if a tribunal erred in its application of the law in Singapore, that would be an error of law that would be insufficient to justify curial intervention, given the principle of minimal curial intervention.

The Court of Appeal then held at Judgment [51(b)] that “… [w]here the tribunal is fully aware of the circumstances [i.e. the fraudulent conduct], the award does not readily come within the ambit of an award induced or affected by fraud. Evidently, the tribunal is better placed to make the assessment of the significance of the procedural breach.”

The distinction drawn is between the “making” of the award, versus the “correctness” of the award. As set out in Judgment [51(c)], “[i]n the absence of further evidence that was not before the tribunal, or a finding of an impermissible flaw in the tribunal’s process (such as a breach of natural justice), there will typically be no room for the court to substitute the tribunal’s findings as to the materiality of concealed information for its own conclusions.”

 

Subject to the Court. Of course, as the Court of Appeal made clear at Judgment [48], in determining whether an award contravenes the public policy of Singapore, a court can, and must, undertake its own assessment. The issue, however, is “… the extent to which the court should accord deference to a tribunal’s decision on the same issue where it has already been canvassed before the tribunal as an issue going to the disposal of the arbitration.

Hence, while the final determination of whether procedural fraud offends the public policy of Singapore lies with the enforcement court (Judgment [53]), the question is whether there are circumstances which warrant the re-litigation of the tribunal’s decision (Judgment [54]).

 

Application to the case. On the facts of the case, the Court of Appeal declined to set aside the award, finding that there were no circumstances which warranted the re-litigation o the tribunal’s decision on the issue of spoliation of evidence (Judgment [54]).

The Court of Appeal found that the tribunal was alive to the issue of alleged fraud and had provided detailed reasons as to why it considered that the fraud could be addressed via the drawing of limited adverse inference, characterising these as procedural rulings which are accorded deference by the court. The Court of Appeal also held that the tribunal’s decision did not involve any obvious errors of fact or law. We set out Judgment [54] – [56] and [58] below for ease of reference.

“54 ... The fraud in question involved what was in effect a failure to comply with discovery obligations in the arbitration. The Tribunal, being fully alive to the nature of the alleged fraud, provided detailed reasons for why it considered that the fraud could be cured with the drawing of a limited adverse inference. Such procedural rulings are accorded deference by the court (see [51(a)] above).

55 We do not consider that the Tribunal’s decision involved any obvious errors of fact or law so as to displace the deference we would typically accord such a ruling. ...

56 Further, the Tribunal’s approach is not inconsistent with decisions rejecting challenges because of a lack of evidence as to the materiality of the Deleted Messages. Where an applicant asserts procedural fraud in the form of the concealment or non-disclosure of material information or documents as the basis for engaging the public policy grounds in the IAA, the applicant should typically establish that the information or documents concealed is “so material that earlier discovery would have prompted the arbitrator to rule in favour of the applicant” (Bloomberry Resorts and Hotels Inc v Global Gaming Philippines LLC [2021] 3 SLR 725 at [106]).

58 As with the case in CYE v CYF, ONI has accepted that it does not know the contents of the Deleted Messages. Before the SICC and this court, ONI did not submit on the contents of the Deleted Messages. Instead, ONI argues that it cannot be faulted for its failure to adduce sufficient evidence of the Deleted Messages, as this was a direct result of GNC’s destruction of evidence. ONI also asserts that “Mr Wong would not have destroyed irrelevant documents”. But the Tribunal was aware of all this and considered that it could address the adverse consequences to ONI with a more limited sanction than ONI sought. We are unable to see that this was so obviously wrong that it would warrant our intervention. We therefore affirm the SICC’s conclusion on this issue.” (emphasis in original)

 

Conclusion. The distinction between the correctness of an award, versus the making of the award, as a crucial one. An award can be incorrect, but that would not typically be grounds for setting aside the award. Given that arbitral rules and procedures often do not provide for appeals from the arbitral award, this distinction is a factor to bear in mind in deciding whether the appropriate the mode of dispute resolution should be litigation or arbitration.

Another point to note is the application of the above distinction in the context of a failure to comply with discovery obligations and a tribunal’s decision flowing from such a failure. Even if there was spoliation of evidence by another party, if the issue was fully argued before a tribunal, it would be difficult to convince a court that the spoliation of evidence would be a ground for setting aside the award. This is because even if the tribunal could be said to have dealt with the issue wrongly on the merits, that would likely be insufficient per se to ground a setting aside application, unless there is an issue with the process of the decision.

 

This publication is not intended to be, nor should it be taken as, legal advice; it is not a substitute for specific legal advice for specific circumstances. You should not take, nor refrain from taking, actions based on this publication.

Xian Ying Tan