In Hayward -v- Zurich Insurance Company PLC [2015] EWCA Civ 327 today the Court of Appeal overturned a decision that a claimant should repay a large part of a personal injury award from an earlier settled action.  This decision makes it very difficult for defendants to reopen settled cases.


The claimant had been injured in an accident at work.  Liability was admitted and shortly before trial the action settled for £134,973.11. In the defence in the first action the insurer had argued that the claim was inflated and exaggerated and it settled for around a third of the stated value.  After the settlement the insurers were informed that the claimant’s claim for damages was inflated and largely inflated.  The insurers issued proceedings against the claimant for fraudulent misrepresentation and claimed rescission of the agreement.  An order striking out the action was overturned on appeal and the matter progressed to trial. The insurer was successful at trial and the action was held to haave a value of £14,720 the claimant was ordered to repay the balance. The claimant appealed.


  • The insurer had settled the action after alleging that the original claim was overstated and fraudulent.
  • The insurer could not be allowed to reopen a case because there was better evidence to establish one of the factors it had taken into account when deciding to settle the matter .
  • To allow the insurer to reopen the case would make settlements difficult, if not impossible.


  1. I believe, with all respect to him, that the Judge’s approach was wrong; but I have had some difficulty reconciling my preferred reasoning with that of the Court on the earlier occasion. I will start by setting out what I believe is the correct analysis without reference to that reasoning.
  2. I start from the fact that the contract of which rescission is sought is a contract to compromise a disputed claim and that the misrepresentations on which the claim for rescission is based consist of (some of) the very averments of fact which the claimant made in advancing that claim – that is, the averments in his pleadings and witness statements – and which had been put in issue.[1] It is important not to lose sight of that context. In particular, we are not concerned with what might be called collateral representations designed to induce the settlement (as in cases like Gilbert v Endean(1878) 9 Ch D 259 or Dietz v Lennig Chemicals [1969] 1 AC 170).
  3. It seems to me clear that in such a case a defendant will not be entitled to seek to have the agreement set aside at some later date only on the basis that he can (now) show that the claimant’s factual statements of the case being advanced were wrong (or, to use the technical language, “false”)[2]. In deciding to settle the defendant takes the risk that those statements are in fact untrue (or, to put it more accurately, would not be proved at trial) and pays a sum commensurate with his assessment of that risk. He could have taken the case to trial in order to disprove the statements in question; but by settling he agrees to forego that opportunity and he cannot reserve the right to come back later for another attempt. If it were otherwise no settlement would be final.
  4. As to how that position is best analysed in legal terms, it would be possible to say that the defendant does not in this context as a matter of fact rely on statements of the character in question but only on his own assessment of all the material facts. This is essentially Mr Sims’s second way of putting his case: see para. 12 above. But arguably that is something of a fiction: at least in some cases, and at least in some sense, the defendant will to a greater or lesser extent rely on the claimant’s statements about factual matters relevant to liability or quantum in deciding whether, or at what level, to settle. It seems to me that the more satisfactory analysis is that the defendant, by entering into the settlement, necessarily implicitly agrees not thereafter to seek to have it set aside on the basis that the statements made in support of the claim were false; another way of putting that would be that he agrees not to rely on them for the purpose of deciding whether to settle.
  5. I have thus far been referring to misrepresentation generally. The position is no doubt different where the factual statements advanced by the claimant and relied on by the defendant are not merely false but fraudulent. That appears from Callisher v Bischoffsheim (1870) LR 5 QB 449. That case is of course the locus classicus for the proposition that the settlement of an ill-founded claim is nonetheless binding, but the Court went out of its way to emphasise that that would not be the case where the claim was fraudulent. As Cockburn CJ put it at p. 452:

“It would be another matter if a person made a claim which he knew to be unfounded, and, by a compromise, derived an advantage under it: in that case his conduct would be fraudulent. If the plea had alleged that the plaintiff knew he had no real claim against the [defendant], that would have been an answer to the action.”

(See also per Blackburn J at p. 452 and Mellor J at p. 453.) In fact in that case the analysis was in terms of absence of consideration – that is, that the forgoing of a bona fide but unfounded claim is good consideration for a payment made in settlement of it but not the forgoing of a fraudulent claim. But I do not think it makes any difference whether the compromise is challenged on the basis of absence of consideration or misrepresentation: the principle must be the same. Thus, in terms of the analysis above, while it may be fair to treat the defendant as having taken the risk of the claimant’s statements in support of his claim being wrong, it will not – absent any indication to the contrary – be fair to treat him as having taken the risk of them being dishonest.

  1. But the qualification “absent any indication to the contrary” is important. The reason for treating fraud differently is that the Court will not normally accept that a defendant has taken the risk that the claimant’s case is not just ill-founded but dishonest. But what risk the defendant is to be treated as having accepted must depend on the circumstances of the particular case. If it is in any case sufficiently apparent that the defendant intended to settle notwithstanding the possibility that the claim was fraudulently advanced, either generally or in some particular respect – the paradigm being where he has previously so asserted – there can be no reason in principle why he should not be held to his agreement even if the fraud subsequently becomes demonstrable. In such a case the position is no different from that discussed at paras. 16-17 above. It cannot be right that a defendant who has made an allegation of fraud against the claimant but decided in the end not to have it tested in the court should be allowed, whenever he chooses, to revive that allegation as a basis for setting aside the settlement. It may stick in the throat that the claimant can retain the reward of his dishonesty, but the defendant will have made the deal with his eyes open to the possibility of fraud, and there is an important public interest in the finality of settlements.[3]
  2. If those principles are applied to the facts of the present case it is in my view clear that Zurich ought not to be entitled to rely on the fraudulent misrepresentations found by the Judge. The statements in question are, as I have said, the statements made by the Appellant in his pleadings and witness statements. The employers had in their Defence not simply put them in issue but positively asserted that they were dishonestly advanced: see para. 2 above. Ms Adams argued that the relevant paragraphs did not amount to a plea of fraud, but I cannot see how an averment that the Appellant was exaggerating his disability “for financial gain” can be anything else.[4] Against that background it is in my view necessarily implicit in the settlement agreement that the employers, and Zurich, gave up the right to have it set aside if they were subsequently in a position to prove the identical dishonesty already alleged.
  3. I indicated earlier that I had found some difficulty in reconciling that reasoning with the decision of this Court on the earlier appeal. Although much of the argument on that occasion was concerned with issues of estoppel per rem judicatam and abuse of process which are not material for our purposes, it seems also to have gone rather further. In the course of his judgment in the County Court Judge Yelton said this:

“In my judgment, the issue of fraudulent exaggeration is clearly raised in the pleadings and was clearly compromised in the Tomlin order. As already stated, it was specifically alleged that the then claimant “has exaggerated … his current physical condition”: I see no material difference between that assertion and the allegation which [counsel] puts before the court to the effect that the evidence of Mr and Mrs Cox is such as to suggest that the claimant had fully recovered by the time of the settlement. The defendant in the first action put fraud into issue, and the claimant settled at a figure far less than his original claim. What Zurich now has is better evidence than they had at the time of the consent order, but on the principles set out above it seems to me that does not give them the right to have an issue reopened. They did not need to compromise in 2003 but having elected to do so it seems to me they are bound by it. The question of discretion in the court simply does not arise.”

Although the Judge appears to have treated that conclusion as part of his reasoning on res judicata, it seems that in this Court Mr Sims relied on it as founding a more broadly-based estoppel. Both Smith LJ and Moore-Bick LJ rejected that argument. Smith LJ said, at para. 29:

“The judge was of the view that the allegation of fraud raised by Zurich in the present action was essentially the same as the defence of exaggeration which had been pleaded in the first action. He was also satisfied that that allegation had been compromised by the settlement. I can see that an allegation that a disability is being exaggerated for gain amounts to fraud and that that allegation of fraud is similar to the allegation now made in the second action. However, in my judgment it is not the same allegation. Nor do I consider that it is clear exactly what was compromised in the first action. For those reasons, I must respectfully disagree with the judge that the first allegation has created an estoppel in respect of the second.”

She went on, at para. 30, to say that the parties had done no more than compromise the value of the claim and had reached no agreement “on the extent of Mr Hayward’s exaggeration or whether it was in fact fraudulent at all”. Likewise Moore-Bick LJ said, at para. 60:

“A large part of Mr. Sims’s argument depended on the broad submission that the issue of fraud had been “compromised” by Zurich in a way that gives rise to an estoppel. In my view, however, even accepting that Zurich can be treated as a party to the settlement agreement, that proposition calls for careful analysis. The settlement involved an agreement on the part of Zurich and its insured to pay a sum of money to Mr. Hayward in consideration of his agreeing to abandon his claim and consent to a stay of the proceedings. Zurich did not consent to judgment and no decision was made on any of the issues raised in the action (apart from the admission that Mr. Hayward had suffered an injury of some kind). Neither side accepted the other’s case in relation to the nature and degree of his injuries. I am not sure, therefore, what is meant by the expression “compromised the issue of fraud”, other than that Zurich agreed to pay a sum of money to Mr. Hayward, despite the fact that it disputed his honesty. Nothing in the settlement agreement involved an admission on the part of Mr. Hayward that his claim was exaggerated or an admission on the part of David S. Smith or Zurich that it was genuine; nor did David S. Smith or Zurich agree to withdraw the allegation of exaggeration. It is difficult to see, therefore, how Zurich could be precluded by the agreement from raising the issue of Mr. Hayward’s honesty in another context, if it were to become relevant.”

Arguably Smith LJ and Moore-Bick LJ do not put the point in quite the same way[5], but it is nevertheless clear that both decided, as a matter ofratio, that the fact that the employers had alleged deliberate exaggeration prior to the settlement did not preclude them from relying on it subsequently as a ground for rescission.

  1. It is a debatable point whether that conclusion, so reasoned, precludes me from deciding the appeal on the basis developed in the previous paragraphs. The Court had not of course heard any evidence and was concerned only with whether Zurich’s claim should be struck out; but I am not sure that the Judge’s findings of fact are relevant on the present issue. I could point to the fact that the discussion was couched in terms of estoppel rather than contract, but I have an uneasy feeling that that is in this context a distinction without a difference. In the end I think the fair thing is to park that question and consider whether my reasoning can be re-cast in a form which, albeit perhaps less satisfactory, avoids the potential conflict.
  2. I believe that such a course is indeed possible. Although at para. 17 above I have expressed a preference for a contractual analysis, I regard the alternative analysis based on reliance as perfectly viable. I repeat that it is important to recall the very particular context in which the reliance is said to occur – that is, that the contract in issue is a settlement agreement and the misrepresentations relied on comprise the very allegations advanced as part of the claim being settled. No doubt in one sense those allegations do operate on the mind of the defendant in his consideration of whether to settle, and at what level: he may actually believe them to be true, but even if he does not he will inevitably be influenced by the possibility that they will be believed by the Court – which is of course what the Judge found to be the case here. But to my mind that does not constitute reliance in the relevant sense. The defendant is not concerned with the truth or otherwise of the statements as the factor motivating his action. Rather, he is treating them simply as part of the claimant’s case. It is inherent in the antagonistic relationship of claimant and defendant that in deciding whether to settle he has to form an independent judgment about whether the disputed statements made as part of the claim are (to the extent that they are material to the outcome) likely to be accepted by the Court. I do not believe that a relationship of reliance arises in that context.
  3. I draw some support for that approach from the decision in Kyle Bay Ltd v Underwriters Subscribing under Policy no. 019057/08/01 [2007] EWCA Civ 57, in which this Court upheld a finding that an underwriter who settled a claim had not relied on a misrepresentation made by an assured as to the effect of the policy because he had treated it as a contention rather than a representation. Neuberger LJ said, at paras. 41-42:

“41. … I can see the force of the point that the statements as to the effect of the Policy … did induce the Settlement, in the sense that they were a causative factor in the claimant and [the underwriter] agreeing on a settlement figure of around £200,000, rather than around £300,000.

42. However, I think the Judge effectively found that [the underwriter] merely treated the statements as contentions which he decide to assume were right for the purposes of the Settlement, and which he would thereafter investigate (as he did) with a view to obtaining compensation for the claimant. In effect, he chose to treat them as correct for the time being, without necessarily believing that they were accurate, … .”

That is not on all fours with the present case, but it does illustrate a rather similar distinction being applied.

  1. I would for those reasons allow the appeal in this action. The result is that the settlement in the original action remains binding and the judgment of Judge Moloney must be set aside. The result is unattractive because it means that the Appellant retains the benefit of a settlement far in excess of the value of his actual loss, though I dare say somewhat reduced by the incidence of costs in these protracted proceedings. But there is a wider principle at stake, that parties who settle claims with their eyes wide open should not be entitled to revive them only because better evidence comes along later.

Lord Justice Briggs

  1. I agree that this appeal should be allowed. I would gladly have embraced any sound basis for upholding the trial judge’s decision to strip the Appellant of the grossly inflated amount which he received upon the settlement of his fraudulently exaggerated claim. But in my opinion the judge’s finding that Zurich was induced into making the settlement agreement by reliance on the Appellant’s dishonest misrepresentations about his continuing injury was based upon a view of the law for which there is no authority, which is wrong in principle and the recognition of which would have most unfortunate consequences.
  2. The principle upon which the judge relied appears to be this: that if A makes a contract with B, to the terms of which A has been influenced into agreeing because he fears that a statement by B which he believes is untrue may nonetheless be believed and acted on by C, then A may later rescind his contract with B if he can prove the untruth of the statement. Put in negative terms it comes to this: that the requirement common to the law of misrepresentation and deceit that the misrepresentation must induce the contract sought to be rescinded is only a bare causation requirement, and does not require that the representee be in any way misled. Nor does it appear to require that A learn something about the untruth of the statement after making the contract. It is enough if the mere making of the misstatement, rather than A’s perception that it was true, had a causative effect upon A’s entry into the contract. It appears that A would be free to seek rescission merely using the evidence he already had when making the contract, because he had in the meantime simply repented of it, or because his apprehension about the effect of the statement upon C had gone away.
  3. In my judgment the authorities on rescission for misrepresentation speak with one voice. For a misstatement to be the basis for a claim to rescind a contract, the claimant 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 judges and text-book writers have used the word ‘influenced’ as the touchstone for reliance they have done so in order to allow for belief in the truth of the misrepresentation to be a contributory rather than sole cause of the representee’s entry into the contract: see for example Clerk and Lindsell on Torts (21st ed.) para 18-35. They have not thereby intended to allow in any case where the representee can show that he was influenced into making the contract by the mere making of a representation which he did not believe was true.
  4. It is not necessary to cite copious authority for this principle. It is sufficient to note that in Moore-Bick LJ’s judgment in the earlier appeal he said that the question whether the Appellant’s fraud was an inducement to Zurich to enter into the settlement agreement depended upon:

“the extent to which, if at all, it was in fact misled when it approved that agreement”

  1. I readily acknowledge that the representee may not need to have had blind faith in the truth of the representation. He may only have been prepared to give the representor the benefit of considerable doubt. If by doing so he was influenced into making the contract, then he may succeed. But in this case Zurich did not merely disbelieve the Appellant’s assertions about the continuing effect of his injury. It went so far as to plead (under a statement of truth) that they were fraudulent. Again I acknowledge that the subsequent discovery by the representee that a statement which he had thought genuine but mistaken was in fact fraudulent might be sufficient for rescission because, in principle at least, fraud unravels all. But in this case there can be no such unravelling. Zurich alleged that the statement was fraudulent from the outset. All that happened thereafter was that better evidence of the fraud came to light than was available when the settlement contract was made.
  2. In my opinion the true principle is that the equitable remedy of rescission answers the affront to conscience occasioned by holding to a contract a party who has been influenced into making it by being misled or, worse still, defrauded by his counterparty. Thus, once he discovers the truth, he must elect whether to rescind or to proceed with the contract. It must follow that, if he already knows or perceives the truth by the time of the contract, he elects to proceed by entering into it, and cannot later seek rescission merely because he later obtains better evidence of that which he already believed, still less if he merely repents of it. This seems to me to be a fortiori the case where, as here, the misrepresentation consists of a disputed claim in litigation, and the contract settles that claim.
  3. Nor is there anything contrary to conscience in holding a person to a contract made in order to deal with the risk that a statement which he believes to be untrue and even fraudulent may nonetheless persuade someone else, even a judge. The contract is made with his eyes open about the probable untruth of the statement. His contract is a form of risk management, and there is no reason why he should be enabled to walk away from it merely because that risk later diminishes or disappears.
  4. To extend the law of rescission in the manner here under consideration would have the most unfortunate consequences. The first would be that it would become almost impossible to compromise a whole swathe of litigation if settlements were vulnerable to being set aside in this manner. Apprehension by one party that his opponent may persuade the trial judge of matters which he denies, and disbelieves, is an everyday characteristic of litigation, and a healthy driver towards settlement, as every mediator knows. If the principle contended for were correct, almost any litigant could say that he was influenced to settle a case for more than it was worth because of a fear that the judge might believe his opponent, even though he did not. To be able to treat as an actionable misrepresentation the opponent’s statement of his case merely because of such an everyday apprehension would expose almost any settlement to subsequent attack if fresh evidence became available. Indeed, there is nothing in the reliance test propounded by the judge that would even make the obtaining of fresh evidence a necessary condition. The public policy which encourages settlement of litigation would be gravely undermined if, in effect, dissatisfaction on either side led, with or without later forensic research, to the settlement being impugned on the ground that the opponent’s case contained a misrepresentation which, without being believed, influenced the terms of settlement.
  5. In that context I accept that fraud stands in a different category but not if, as here, the settlement compromised an allegation of fraud already on the pleadings. For my part this does not depend on estoppel, but rather on a robust disinclination to set aside contracts in settlement of litigation, save where the traditional requirement of the party seeking rescission to show that he has been misled or defrauded is well and truly satisfied. For that reason I have encountered less difficulty than my Lord with the outcome of the strike out application in the earlier appeal to this court. My understanding of that decision is merely that there was no relevant estoppel sufficient to make the further pursuit of Zurich’s case an abuse of process.
  6. The second unfortunate consequence would be that there would be no easy way of confining this more generous approach to inducement to the rescission of contracts in settlement of litigation. If the mere making of a misstatement rather than belief in its truth is to be a sufficient influencing factor, then there is no telling in what contractual contexts it may be applied, with debilitating effects upon contractual force and certainty.
  7. For those reasons I would also allow this appeal.

Lady Justice King

  1. I agree with both judgments.

Note 1    I should say that it was not argued before us that the Judge’s reference at para. 6.6 of his judgment to the Appellant having continued his misrepresentations even after the disclosure of the 1999 video meant that any distinction could be drawn for the purpose of reliance between his initial exaggeration of the extent of his injuries and anything said subsequently. Nor in any event would such a distinction be sustainable.     [Back]

Note 2    For convenience I refer to the party who might wish to have the agreement set aside as the defendant but of course in principle it could be the claimant.    [Back]

Note 3    There is perhaps some parallel with the Court’s willingness to recognise the validity of the settlement of a claim based on an illegal transaction, as in Binder v Alachouzos [1972] 2 QB 151.     [Back]

Note 4    I note that that was also the view of Smith LJ in the earlier appeal: see para. 29 of her judgment.     [Back]

Note 5    Maurice Kay LJ identified a difference between the reasoning in the two judgments, which he thought it unnecessary to resolve; but what matters for present purposes is that the argument based on the effect of the agreement itself was rejected.    [Back]