Singapore Court of Appeal Clarifies Assessment of Fraud Damages in a Falling Market

Introduction

In Kupetz, Jonathan and others v Terraform Labs Pte Ltd and others [2026] SGCA(I) 1, the Singapore Court of Appeal clarified how damages for fraud should be assessed where an asset that is acquired in reliance on the fraud collapses in value in a falling market. The decision highlights the importance of identifying when the fraud ceases to be operative, after which any loss suffered by the buyer may be attributed to his own decision to hold on to the asset instead of to the fraud. 

Background Facts

The claimants were representative claimants who had invested in TerraUSD, or UST, an algorithmic stablecoin. The defendants were developers of or entities associated with UST. The claimants claimed that the defendants had made misrepresentations about UST’s stability, including as to how robustly UST would be pegged to the US dollar. When UST later de-pegged from the US dollar, its value collapsed, and the claimants suffered investment losses. The defendants conceded before trial that they had made false and fraudulent representations, but did not admit reliance, causation or loss.

The trial judge found that seven of the ten representative claimants had relied on the misrepresentations and suffered loss. The trial judge then determined that by a certain time which was referred to as the “Cut Off Time”, it would have been apparent that the representations were not true, so the claimants’ reliance on the representations would have ended. He awarded the seven claimants the difference between the price at which they had invested in UST, and the value of the UST they held based on the opening price at the Cut-Off Time, which was referred to as the “Cut Off Price”. The total award of damages was around US$451,000, a fraction of the total amount claimed.

The claimants appealed. They challenged the trial judge’s approach to the Cut-Off Time and the Cut-Off Price as a finding on whether the claimants had mitigated their loss, which they argued the defendants had not properly pleaded. They also argued that the trial judge had erred in applying the Cut-Off Time to all the representative claimants, and in using the Cut-Off Price for the purposes of a notional sale at the Cut-Off Time.  

Causation or Mitigation?

The Court of Appeal held that the trial judge’s analysis was properly one of causation, and not mitigation as the claimants had argued.

The Court re-affirmed the general principle that where a claimant has been fraudulently induced to buy an asset, he may recover the purchase price less the value of the benefits received. But those benefits must be valued as at a certain date. The Court held that the starting point in determining this date is the date the misrepresentation ceases to be operative—that is, when the fraud was discovered. Once the fraud is discovered, if the claimant exercises a free choice to hold on to the acquired asset, the chain of causation is broken and any further losses are no longer regarded as caused by the fraud. However, if the claimant remains locked into the property due to circumstances beyond his control, such that she cannot be said to have been a free and informed decision to hold on, further losses may still be causally attributed to the defendant. The issue of mitigation may then arise, but only after the claimant has established the causal link between the fraud and the loss.

This is an important clarification. In claims of this nature, defendants often argue that claimants held on to the relevant assets too long, so their losses should not fall on the defendants. Claimants may answer that their losses would not have arisen had they not been induced to invest in the first place. They may also contend that they sought reasonably to mitigate their loss by waiting for the market to bounce back. The Court of Appeal’s decision makes it clear that notwithstanding such arguments, the burden remains on the claimant to establish when the fraud operated—and ceased to operate—on the claimant’s mind and, in turn, what loss was caused by the fraud.

When did the Fraud Cease to Operate?

The trial judge had adopted 12 May 2022 at 12.01am UTC as the Cut-Off Time. He found that one of the defendants had on 11 May 2022 published tweets which, when read as a whole, would have indicated to a reasonable investor that UST’s value was not stable and that the mechanism for its peg to the US dollar had failed. This was followed by a 14-hour period for investors to read, digest and act. The trial judge considered that, by then, a reasonable investor would have appreciated that UST was not stable and that the mechanisms said to maintain its peg had failed.

The Court of Appeal upheld the use of this Cut-Off Time for the representative claimants. The Court accepted that, on the evidence before the trial judge, a reasonable investor considering the whole of the 11 May 2022 tweets would have appreciated that UST’s value was not stable. The Court was not persuaded by the argument that a temporary re-pegging which occurred between 7 and 9 May 2022, or optimistic language in the tweets, meant that investors could reasonably continue to treat the earlier stability representations as operative. By 11 May 2022, UST had traded far below the one-dollar peg, including an intra-day low of US$0.2998. The tweets referred to the need for collateralisation. In context, they pointed to a breakdown of the stablecoin.

At the same time, the Court was careful not to overstate the representative effect of that finding. It held that the Cut-Off Time did not necessarily bind the represented claimants whose claims remained to be determined later. Individual circumstances may matter. A claimant who had not read the relevant tweets by the Cut-Off Time might not yet have discovered the fraud. A claimant who read them and took reasonable steps to sell but was locked into a platform, unable to transfer tokens, or otherwise unable to exit may be able to show that later losses remained causally attributable to the fraud.

This part of the decision is significant for group or representative claims involving financial products. The Court recognised that common issues may properly be decided in a representative tranche, especially where the meaning of a public statement or market event is common to all claimants. But this does not preclude a represented claimant from later putting forward, or having to establish, her own basis for why she had suffered loss. The parties should not assume that a ruling on causation one way or the other for one set of representative claimants will apply equally to all represented claimants. 

What was the Value of the Benefit Received?

The Court of Appeal’s key intervention concerned price. The trial judge had valued the representative claimants’ UST holdings as at the Cut-Off Time based on the opening price on 12 May 2022, US$0.8011. But the Court of Appeal held that this was too mechanical. This is because it treated the claimants as all having sold by the Cut-Off Time. But it was unrealistic to assume that all remaining UST could have been sold at precisely that moment and at precisely that price.

The Court observed that where many claimants were assumed to be selling into a highly volatile and distressed market, the act of selling might itself depress the price or require time to execute. This was not the ordinary case of one claimant entering a deep and functioning market without affecting price. UST’s daily price range was extreme. On 11 May 2022, it moved between US$0.847 and US$0.2998. On 12 May 2022, it moved between US$0.829 and US$0.3626. Selecting a single opening price was therefore potentially arbitrary.

The Court adopted US$0.60485, the average of the opening and closing price on 12 May 2022, which had been proposed by the defendants. In the Court’s view, that price better reflected the reality that in the circumstances, investors would probably have obtained lower average sale prices than US$0.8011.

Significantly, the Court declined the claimants’ request to admit fresh evidence that they were locked into their holdings and could not sell at the proposed price. The Court held that they had had the opportunity to adduce this evidence at trial, but they did not do so and must be held to that forensic choice.

A key message is that the courts will not insist on artificial precision in valuing loss where the relevant market is highly volatile. At the same time, the claimant must put forward all the market and valuation evidence they need at trial. That evidence will have a direct impact on proving causation and the quantification of damages. 

Key Takeaways

The Court of Appeal’s decision may incentivise sellers of financial products who have made misstatements to correct them early and unequivocally. Although this may invite claims, a clear and timely correction starts the clock running for potential claimants: once the misrepresentation ceases to be operative, responsibility for losses arising from subsequent market movements shifts to the buyer. The sooner the correction is made, the smaller the window of recoverable loss.

The corollary is that claimants must plead and prove not only reliance at acquisition, but also the basis for recovering any loss sustained after the misrepresentation was made known. As noted above, the Court explained that one situation is where the claimant remains locked into the property due to circumstances outside his control, such that he cannot be said to have made a free and informed choice to hold on to the property. A claimant may find it difficult to show she was locked in if she did not even attempt to sell the property, unless it is obvious that any attempt would have failed.

An important unresolved question is how the Court will assess knowledge acquired through social media and other public channels, as opposed to official statements issued by the seller. In this case, the Court rejected the argument that claimants should be treated as having read the 11 May tweets if they had not in fact done so. But future cases may test the boundaries of this approach, especially where the claimant is a sophisticated or institutional investor expected to have wide access to information. The approach that would apply in a case involving innocent or negligent—as opposed to fraudulent—misrepresentation also was not decided in this case.

The overall lesson is that while the law may be generous to victims of deceit, it is not indifferent to causation. In a collapsing market, the decisive question may be not simply whether the defendant lied, but when the lie stopped causing the loss.

The decision reflects the complexities that commonly arise in financial disputes, and the importance of proper advice on the nuances and technicalities of commercial litigation. As experienced practitioners in this field, Rajah & Tann’s Banking & Financial Disputes Practice is well placed to advise on how best to navigate potentially complex legal proceedings, particularly in the realm of banking and financial disputes. For further queries, please feel free to contact our team on this page. 

This article is authored by Banking & Financial Disputes Partner Torsten Cheong.