Follow us

The Court of Appeal has refused to set aside a settlement agreement on the basis of new evidence indicating that the claimant's case had been fraudulently exaggerated:  Hayward v Zurich Insurance Co plc [2015] EWCA Civ 327.

Settlement agreements may be set aside on the same grounds as any other contract, including on the basis that one of the parties was induced to enter into it by a counterparty's misrepresentation concerning a material fact. However, where the representations relied on comprise the very allegations advanced by the claimant as part of the claim being settled, the courts have generally taken the view that the defendant has, in deciding to settle, taken into account the risk that the statement was false and has foregone the opportunity to challenge it. The issue in the present case was the extent to which the position is different where, as here, the representation was not merely false but could later be shown to have been fraudulently made.

In holding that the innocent party was nevertheless bound by the agreement, the Court of Appeal's decision shows the limits, in this context, of the principle that "fraud unravels all".  It confirms that a fraudulently advanced case will not necessarily entitle a defendant to rescind a settlement agreement – particularly in cases where the evidence suggests that, at the time of settlement, the defendant had at least some indication of the possibility of fraud and therefore settled "with its eyes wide open".

The decision illustrates the courts' "robust disinclination" to interfere with settlement agreements unless there are exceptional circumstances, given the important public interest in the finality of settlements.


An employee brought proceedings against his employers in respect of a workplace injury. The employers' insurer conducting the defence admitted liability but disputed quantum, primarily on the basis that the employee had exaggerated the extent of his ongoing condition. Prior to trial a settlement was reached and the terms recorded in a settlement agreement.

Two years later, the employee's neighbours provided the insurer with further evidence indicating that the claim had been dishonestly exaggerated and that the claimant had in fact fully recovered from his injuries over a year prior to the settlement. The insurer commenced proceedings seeking rescission of the settlement agreement (or damages for deceit in the alternative). It alleged that it had been induced to enter into the agreement by the employee's fraudulent misrepresentations, in the form of statements as to his condition in his pleadings and witness statements.

The County Court set aside the settlement agreement on that basis. It held that, although the evidence indicated that the insurer did not believe the injury claims and strongly suspected exaggeration, the law only required that a party had been "influenced by" a representation in entering a contract. The County Court considered that this requirement was satisfied; the insurer undoubtedly took into account in its settlement deliberations the fact that the claimant's assertions would be put before the court and the risk that they would be believed. In that sense, they influenced the insurer's decision to settle on the terms it did.


The Court of Appeal (Underhill, Briggs and King LJJ) overturned the County Court's decision and ruled that the settlement agreement was binding, although it expressed some regret in doing so. Key to the court's reasoning was the conclusion that the insurer's entry into the agreement had not been induced by the claimant's statements in the relevant sense.

Disagreeing with the lower court, the Court of Appeal held that, for the purpose of establishing reliance, it was not sufficient that a litigant took into account the risk that its opponent's evidence would be believed by the judge, although it did not itself believe the evidence. Briggs LJ noted that it is an everyday feature of litigation for parties to be "influenced" by their opponent's factual pleadings in that sense when assessing whether to settle and at what level. That does not constitute reliance in the relevant sense.

To the extent that earlier authorities have suggested that "influence" was sufficient to constitute reliance, they were simply confirming that the representation did not need to be the sole cause of entry into the contract. They were not suggesting that a party could be induced by a statement if it was not in fact misled. The authorities were clear that, while a party need not have had "blind faith" in the truth of a representation in order to establish reliance, it "must have given some credit to its truth, and been induced into making the contract by a perception that it was true rather than false".

Where the statements being relied on as misrepresentations were among the very matters alleged by the claimant in the proceedings settled by the agreement (rather than representations as to collateral matters), there was clearly no reason in principle to rescind an agreement purely on the basis that a defendant could subsequently show that the allegations were unfounded.

The court acknowledged that that the situation might be different in cases where a defendant discovered that the factual statements made by a claimant were not merely false or unfounded but were fraudulently advanced, because in principle "fraud unravels all". However, importantly, the court drew a distinction between cases where, as here, the defendant had some indication of the possibility of fraud at the time of settlement (whether or not they had pleaded it expressly) and those where it did not. In the former case, the decision suggests that the court is likely to uphold the settlement, on the basis that the possibility of fraud was taken into account in deciding to settle. As Briggs LJ observed, entering into a settlement agreement is a form of risk management and there is no reason a party should be able to walk away from it merely because  that risk later diminishes or disappears (either because of new evidence or otherwise). To hold otherwise would seriously undermine the finality of settlement agreements and run counter to the public policy in favour of encouraging settlement.


In any case where a party seeks to set aside a settlement agreement on grounds of fraud, much is likely to turn on the evidence as to that party's state of knowledge and belief at the time of settlement. In that regard, the decision is consistent with the courts' approach in construing settlement agreements, including determining the scope of a settlement and whether it encompassed claims subsequently raised. In such cases, where at the time of settlement a party had some knowledge that it may have further claims, although ignorant of the detail, the authorities suggest that a court may be more likely to conclude that the settlement was intended to cover such claims than if the possibility of further claims was completely unknown (see our post here on a recent decision illustrating this).

Although the Court of Appeal in the present case acknowledged that rescission of a settlement agreement is "in principle at least" available where, at the time of settling, a defendant had no suspicion regarding a subsequently uncovered fraud, the court's emphasis on the public policy factors against interfering with settlements suggests that defendants seeking such relief may face a high evidential bar.

Note: This decision was overturned on appeal. See our post on the Supreme Court decision here.

Related categories

Key contacts

Alan Watts photo

Alan Watts

Partner, Global Co-Head of Class Actions and Co-Head of Partnerships, London

Alan Watts
Maura McIntosh photo

Maura McIntosh

Professional Support Consultant, London

Maura McIntosh
Jan O'Neill photo

Jan O'Neill

Professional Support Lawyer, London

Jan O'Neill