//
© 2026 Zoe. All rights reserved.
Zoe is a legal information platform. Always consult the Federal Register of Legislation for authoritative text.
[1998] HCA 38
High Court of Australia
1998-05-20
Brennan CJ, Hayne JJ, Hayne J
Original judgment source is linked above.
High Court of Australia Brennan CJ McHugh, Gummow, Kirby and Hayne JJ Unity Insurance Brokers Pty Ltd v Rocco Pezzano Pty Ltd (P64/1996) [1998] HCA 38
The question in this appeal is whether a plaintiff claiming damages for breach of contract is entitled to damages for loss arising from the plaintiff compromising legal proceedings with a third party where the proceedings arose out of the breach of contract.
The appeal is brought by Unity Insurance Brokers Pty Ltd (the broker) against an order of the Full Court of the Supreme Court of Western Australia. That order dismissed the broker's appeal against a judgment holding it liable in damages for breach of a contract made with the respondent. The broker contends that, although it breached its contract with the insured by failing to exercise reasonable care and skill in obtaining an insurance policy, the insured failed to prove that it had suffered a loss from that breach. That contention is based on the ground that the insured sought to prove its loss simply by proving the difference between the sum that it would have received from the insurer, if due care had been exercised, and the sum that it did receive in settlement of the claim with the insurer. The broker contends that to calculate its loss the insured was required to prove the sum that it would have received if it had litigated the matter against the insurer and that the insured cannot rely on the settlement sum in calculating the difference.
The broker agreed to obtain insurance for the respondent, Rocco Pezzano Pty Ltd, (the insured) in respect of the insured's business premises in East Cannington, near Perth. Pursuant to the agreement, the broker obtained a special risks insurance policy from NZI Insurance Australia Ltd (the insurer). Among the risks covered by the policy was damage by fire to the insured's plant, machinery, stock, contents and premises. During the currency of the policy, a fire at the premises extensively damaged many of those items. The insurer refused to indemnify the insured for the damage because "of material non-disclosure of prior claims". The non-disclosure was caused by the broker who disclosed to the insurer only one of twelve claims that the insured had made against insurers during the previous thirteen years.
As a result of the insurer's refusal to indemnify it, the insured sued the insurer for breach of its promise of indemnity and the broker for breach of its duty to exercise reasonable care and skill in obtaining the policy. The insurer, relying on s 28 of the Insurance Contracts Act 1984 Cth (the Act), asserted that it was entitled to reduce its liability under the policy to nil by reason of the non-disclosure. In these circumstances, the legal advisers of the insured, including senior counsel advised it to settle the action against the insurer for $900,000. The shortfall between the amount of the settlement and a full indemnity was more than $800,000. The insured contends that, because the broker placed it in the position where it had to settle for this sum, the broker is liable to it for the difference between the amount of the full indemnity and the sum of $900,000 received in the compromise settlement.
In the action between the insured and the broker [6] , Mr Pezzano, a director of the insured, gave evidence that he had been advised that the insured should accept the sum of $900,000 rather than "losing everything because there was no disclosure" of the full claims history. He also testified that the insured's lawyers had told him that the insurer "wouldn't pay any more than that". Senior counsel and the insured's solicitors who gave this advice were not called as witnesses. Nor was there any other evidence concerning the reasons for the settlement. There was, however, evidence that the initial offer from the insurer was $740,000 and that this had been eventually increased to the sum of $900,000 which was then accepted.
(1) This section applies where the person who became the insured under a contract of general insurance upon the contract being entered into:
(a) failed to comply with the duty of disclosure; or
(b) made a misrepresentation to the insurer before the contract was entered into;
but does not apply where the insurer would have entered into the contract, for the same premium and on the same terms and conditions, even if the insured had not failed to comply with the duty of disclosure or had not made the misrepresentation before the contract was entered into.
(2) If the failure was fraudulent or the misrepresentation was made fraudulently, the insurer may avoid the contract.
(3) If the insurer is not entitled to avoid the contract or, being entitled to avoid the contract (whether under sub-section (2) or otherwise) has not done so, the liability of the insurer in respect of a claim is reduced to the amount that would place him in a position in which he would have been if the failure had not occurred or the misrepresentation had not been made.
In Biggin, Devlin J and the English Court of Appeal considered a claim for damages where the defendant had sold defective material to the plaintiff who had resold it and had later been forced to settle claims made against it by third parties for supplying the defective material. The plaintiff claimed that the amounts paid in settlement of the claims against the defendant were losses recoverable from the defendant for its breach of contract at first instance. Devlin J rejected the claim. He said [17] :
Was the compromise a foreseeable consequence of the making of the claim? In my judgment, it was not a consequence in the legal sense at all. It flowed from the voluntary acts of the plaintiffs. His Lordship said that a plaintiff who settles does so at his own risk and that if "he compromises erroneously where there is no liability, he cannot make the defendant pay for his error" [18] . It is difficult to know whether his Lordship decided the case on causation or remoteness or both. The reference to foreseeability in the first sentence of the above extract suggests remoteness. The reference to consequence in the second sentence suggests causation. 1. Biggin [1951] 1 KB 422 at 428. 2. Biggin [1951] 1 KB 422 at 427.
On appeal, the Court of Appeal came to a different conclusion. Somervell LJ said [19] :
I think that the judge here was wrong in regarding the settlement as wholly irrelevant. I think, though it is not conclusive, that the fact that it is admittedly an upper limit would lead to the conclusion that, if reasonable, it should be taken as the measure. The result of the judge's conclusion is that the plaintiff must prove their damages strictly to an extent to show that they equal or exceed £43,000; and that if that involves, as it would here, a very complicated and expensive inquiry, still that has to be done. The law in my opinion, encourages reasonable settlements, particularly where, as here, strict proof would be a very expensive matter. The question, in my opinion, is: what evidence is necessary to establish reasonableness? Singleton LJ said [20] : The plaintiffs must establish a prima facie case that the settlement was a reasonable one. If the defendants failed to shake that case, the amount of the settlement can properly be awarded as damages. The position is much the same, though perhaps not quite so strong, as in the case in which damages have been assessed in a suit between other parties involving the same facts. The judgment is not binding, but the court will not lightly disregard it in the absence of fresh evidence or new factors. Birkett LJ agreed with both judgments [21] . 1. Biggin & Co Ltd v Permanite Ltd [1951] 2 KB 314 at 321. 2. Biggin [1951] 2 KB 314 at 325. 3. Biggin [1951] 2 KB 314 at 327.
With great respect to the Court of Appeal, the reasoning is far from clear. First, it is not clear whether their Lordships even considered issues of causation and remoteness. Perhaps their Lordships assumed that the settlement was causally connected with the breach and that it was not too remote a consequence of that breach if the settlement was reasonable. But they did not say so. Moreover, the reference to public policy in the judgment of Somervell LJ suggests that they did not consider the case in terms of causation or remoteness.
It is impossible, however, to think that their Lordships intended that money paid in reasonably settling a third party action following a breach of contract could be recovered from the defendant even though there was no causal connection between the settlement and the breach. But, even on this restricted reading of the judgments of the Court of Appeal, their Lordships' reasons are at least open to the interpretation that every reasonable settlement of litigation resulting from a breach by the defendant must be regarded as within the contemplation of the defendant. That proposition cannot be accepted.
Whether a settlement was within the contemplation of a defendant or a reasonable person in its position must depend upon the nature of the contract between the plaintiff and the defendant, their actual or imputed knowledge of the consequences of a breach, and the nature of the third party's claim against the plaintiff. As a general rule, a contract breaker must be taken to have reasonably contemplated that its breach may force the innocent party into litigation with third parties and that the innocent party may conclude that it is in its best interest to compromise the third party's claim. But it does not follow that the fact that it was reasonable for the plaintiff to compromise the claim against the third party necessarily means that the settlement was within the reasonable contemplation of the defendant. That is so even in those cases where the defendant's breach was proved to be causally connected with the settlement. Each case must depend upon its own facts.
In my opinion, Biggin should not be used as an authority in this country. Whether the claim is in tort or contract, the question whether the plaintiff can recover from the defendant moneys paid in settlement of a third party action depends on ordinary principles of causation and remoteness. That proposition also applies in a case such as the present where the settlement is an element in the calculation of damages.
One further point should be made about the Court of Appeal's judgment in Biggin. Somervell LJ expressed the opinion that, while the client can prove that the settlement was made as the result of legal advice, the evidence of the advisers is not ordinarily relevant or admissible. With great respect, I am unable to accept that the evidence of the legal advisers is not normally relevant or admissible in such a case. On the contrary, in most cases where the settlement is made on legal advice, the evidence of the relevant legal advisers is vital. This is because the risk involved in the litigation and the reasoning which led to the settlement are the factors that will determine whether or not the settlement was reasonable. If an unreasonable settlement is made on bad legal advice, the innocent party's remedy is against the legal adviser, not the contract breaker.
The final question in the appeal is whether the trial judge's finding that the settlement was reasonable can stand. The insured called no evidence from its legal advisers. The learned judge said [22] :
It seems to me, in all of the circumstances, that, viewed from the plaintiff's perspective in January 1994 (when the settlement with NZI was achieved), there was a reasonable prospect that NZI would succeed in its defence on the non-disclosure issue (that, as I have said, having been the only issue which, on the evidence, caused the plaintiff to consider it necessary to settle for less than the amount of its claim after taking into account the proper application on the average provisions of the policy).
Once that is accepted, and having regard for the fact that the plaintiff had been advised by senior counsel to settle with NZI for payment of the sum of $900,000 (and Mr Pezzano said in evidence not only that he was told that it was better to accept this amount than to lose everything "because there was no disclosure" of the full claims history but also that his lawyer told him that NZI "wouldn't pay any more than that"), it seems to me that the settlement was reasonable.
Of course the plaintiff was required, in circumstances in which it proposed to recover the balance of its claim from the defendant, to act with the defendant's interests in mind as well as its own. I am, in this respect, troubled by the fact that no evidence was placed before me to suggest that the defendant was invited to comment on the reasonableness of the proposed settlement [23] However I am, in the end, satisfied that this did not render the settlement unreasonable in circumstances in which the defendant was maintaining a complete denial of liability to the plaintiff, in which it had denied knowledge of the plaintiff's prior claims history and in which it was ascribing to the plaintiff the very act, which had given rise to the non-disclosure, of which it had itself been guilty. Moreover, it may, in this respect, be apposite to quote what was said by Lord Macmillan (albeit in a somewhat different context) in Banco de Portugal v Waterlow [24] cited in McGregor on Damages [25] as follows: "Where the sufferer from a breach of contract finds himself in consequence of that breach placed in a position of embarrassment the measures which he may be driven to adopt in order to extricate himself ought not to be weighed in nice scales at the instance of the party whose breach of contract has occasioned the difficulty. It is often easy after an emergency has passed to criticise the steps which have been taken to meet it, but such criticism does not come well from those who have themselves created the emergency. The law is satisfied if the party placed in a difficult situation by reason of the breach of a duty owed to him has acted reasonably in the adoption of remedial measures and he will not be held disentitled to recover the cost of such measures merely because the party in breach can suggest that other measures less burdensome to him might have been taken."
I am satisfied, in the light of the evidence before me, that the defendant's breach of duty (whether contractual or tortious) was such as to leave the plaintiff exposed to a real risk of complete failure in relation to its claim against NZI, that the circumstances were such that it was reasonable for it to have sought to compromise that claim and that the compromise which it did in fact achieve was reasonable in all of the circumstances.
The facts are detailed in the judgments of Kirby J and Hayne J and I do not repeat them in detail.
The respondent, Rocco Pezzano Pty Ltd, engaged the appellant, Unity Insurance Brokers Pty Ltd, as its insurance broker. The appellant negotiated a contract of insurance between the respondent and NZI Insurance Australia Ltd (the insurer) for a policy period of twelve months ending 29 June 1993. The contract was "a contract of general insurance" within the meaning of s 11(6) of the Insurance Contracts Act 1984 Cth (the Act). Section 28 of the Act states:
(1) This section applies where the person who became the insured under a contract of general insurance upon the contract being entered into:
(a) failed to comply with the duty of disclosure; or
(b) made a misrepresentation to the insurer before the contract was entered into;
but does not apply where the insurer would have entered into the contract, for the same premium and on the same terms and conditions, even if the insured had not failed to comply with the duty of disclosure or had not made the misrepresentation before the contract was entered into.
(2) If the failure was fraudulent or the misrepresentation was made fraudulently, the insurer may avoid the contract.
(3) If the insurer is not entitled to avoid the contract or, being entitled to avoid the contract (whether under subsection (2) or otherwise) has not done so, the liability of the insurer in respect of a claim is reduced to the amount that would place him in a position in which he would have been if the failure had not occurred or the misrepresentation had not been made.
The respondent had a substantial history of prior claims which it disclosed to the appellant as its broker. The appellant failed to comply with the duty of disclosure in respect of the respondent's claims history. The consequence was that, whilst the insurer was not entitled to avoid the policy, its liability in respect of a claim was reduced by s 28(3) of the Act to the amount which would place it in a position in which it would have been if the failure had not occurred. Rather than acquiring a policy which was fully enforceable according to its terms and not subject to the operation of s 28(3) of the Act, the respondent acquired something of lesser value. This was a cover which was open to controversy between it and the insurer and from which there was a reduction in what otherwise would have been the liability of the insurer.
The consequence of the operation of s 28(3) was that this was not a situation where the respondent was necessarily to be provided with a full indemnity either by the insurer (under the policy) or by the broker (by reason of the broker's breach of contractual retainer or its negligence).
For the ascertainment of the mischief to the remedy of which s 28(3) was directed, it is permissible [29] to have regard to the Report of The Law Reform Commission on Insurance Contracts (the Report) [30] , which preceded the enactment of the legislation. The Report stated [31] :
The nature and extent of the insurer's redress should depend on the nature and extent of the loss which it has suffered as a result of the insured's conduct. It should no longer be entitled to avoid a contract, and a heavy claim under that contract, merely because it has suffered a small, even insubstantial, loss as a result of a non-disclosure or misrepresentation. Of a proposed provision in terms identical to s 28(3), the Report stated [32] : An insurer is not entitled to avoid a contract for innocent misrepresentation or non-disclosure, but may reduce a claim by the amount of the loss it has suffered as a result of the misrepresentation or non-disclosure. The amount by which the claim is reduced is the amount which would place the insurer in the position it would have been in if the failure to comply with the duty of disclosure had not occurred or the misrepresentation had not been made. For example, if the insured would have charged a higher premium had the misrepresentation not been made, it would be entitled to reduce the claim by the amount of the additional premium. If the insurer would not have entered into the contract at all, it would be entitled to pay nothing in respect of the claim except the premium paid by the insured. If the insurer would have inserted a different term in the contract, then the insurer would only be liable for the amount for which it would have been liable if that term had been a term of the contract. The principle is the one generally applied in assessing damages for misrepresentation. It is accepted that one application of s 28(3) is to cases where, if disclosure had been made, the insurer would have accepted the risk but on different terms. There has been a division of opinion, which it is unnecessary now to resolve, as to the amount, if any, recoverable by an insured where the insurer would have declined the risk altogether had proper disclosure been made [33] . The litigation between the appellant and the respondent has been conducted on the footing that in a case where, were it not for the failure of the insured to comply with its duty of disclosure, the insurer would not have accepted the proposal at all, s 28(3) may operate to reduce the liability of the insurer to nil. 1. Newcastle City Council v GIO General Ltd (1997) 191 CLR 85. 2. Report No 20 (1982). 3. at p 117. 4. at p 284. 5. See Advance (NSW) Insurance Agencies Pty Ltd v Matthews (1989) 166 CLR 606 at 621-622; Twenty-First Maylux Pty Ltd v Mercantile Mutual Insurance (Australia) Ltd [1990] VR 919 at 926-928.
In a fire on 20 September 1992, during the currency of the policy, the respondent suffered damage to the insured property. It is accepted that, had it admitted liability under the policy, the insurer would have met the respondent's claim in the sum of $1,720,287.04. However, there has been no finding as to the amount to which, pursuant to s 28(3), the liability of the insurer was reduced so as to place it in the position it would have occupied if the failure to comply with the duty of disclosure had not occurred. It is that circumstance which is at the heart of the appeal brought by the broker to this Court.
The insurer denied liability to indemnify the respondent, in respect of the claim made after the fire, by reason of the non-disclosure. The respondent commenced proceedings in the Supreme Court of Western Australia against the appellant and the insurer. The action against the insurer was settled for $900,000. The action against the appellant proceeded to trial and the respondent recovered judgment in the sum of $1,041,166 [34] . The respondent supports that outcome on the footing that the amount for which it recovered judgment was assessed on the basis that the difference between what the insurer would have paid had it admitted the claim and the amount of the settlement represents the difference in value between the insurance cover which the appellant should have obtained for the respondent and the more limited cover which results from the application of s 28(3) of the Act. Implicit in this position taken by the respondent is the proposition that the amount of the settlement reached in the action between the respondent and the insurer is to be taken in the determination of the liability of the appellant to the respondent as the reduced amount in respect of the liability of the insurer under s 28(3).
The judgments in the Supreme Court of Western Australia
The question is not whether [the insurer] considered that it was or was not entitled to reduce its liability to nil but, rather, that of whether [the respondent] acted reasonably in compromising its claim against [the insurer] for $900,000. 1. Rocco Pezzano (1995) 8 ANZ Insurance Cases ¶61-288 at 76,213.
The contract of brokerage was not to provide the respondent with a chance or opportunity, uncertain of realisation, to obtain a commercial advantage or benefit. The respondent enlisted the appellant's services to obtain secure insurance cover. The broker failed to exercise reasonable care, skill and diligence, with the result that the cover provided by the policy was limited by the reduction in the liability of the insurer by s 28(3) of the Act. The difference between the value of what should have been provided and what was provided was the respondent's loss. The court should not have been "concerned with valuing a chance or prospect that the respondent might have lost" [39] . The question was one of assessing the measure of the respondent's loss in the events that happened consequent upon the fire of 20 September 1992 and the denial by the insurer of all liability under the policy.
Next, it was said for the respondent, with particular reference to an observation by Somervell LJ in Biggin & Co Ltd v Permanite Ltd [40] , that "the law encourages reasonable settlements". No doubt, with respect to the action between the respondent and the insurer, this was so. But it is not the policy of the law, at least in the absence of their agreement [41] , to compel or cajole parties to settle or to foreclose their right of access to the courts for the determination of their respective legal rights and obligations. That right of access to curial determination is deeply rooted in constitutional principle. In Bremer Vulkan Schiffbau und Maschinenfabrik v South India Shipping Corporation Ltd [42] , Lord Diplock said:
Every civilised system of government requires that the state should make available to all its citizens a means for the just and peaceful settlement of disputes between them as to their respective legal rights. The means provided are courts of justice to which every citizen has a constitutional right of access in the role of plaintiff to obtain the remedy to which he claims to be entitled in consequence of an alleged breach of his legal or equitable rights by some other citizen, the defendant. Nor does the policy of the law in encouraging settlements necessarily carry with it the consequence that a settlement between A and B of their respective legal rights and obligations upon a particular subject matter provides a determinative measure of the obligations between B and C with respect to that subject matter, so as to relieve B of the necessity to prove the loss B seeks to recover from C. 1. [1951] 2 KB 314 at 321. 2. cf Brooks v Burns Philp Trustee Co Ltd (1969) 121 CLR 432 at 451-453. 3. [1981] AC 909 at 977. See also R v Lord Chancellor; Ex parte Witham [1997] 2 All ER 779 at 784; R v Secretary of State for the Home Department; Ex parte Pierson [1998] AC 539 at 574-575.
Here, the respondent sued the appellant for a breach of a term implied in their contract that the appellant would exercise reasonable care, skill and diligence in providing the respondent with insurance advice and services. The respondent alleged that it had sustained damage which was represented by the difference between what would otherwise have been payable under the policy and the amount of $900,000 recovered from the insurer under the settlement.
The respondent also alleged that the appellant owed it a duty of care to take all reasonable steps in conducting its business so as not to expose or cause the respondent to suffer unnecessary loss and damage. The trial judge held that, whether contractual or tortious, the appellant had breached its duty to the respondent. In The Commonwealth v Amann Aviation Pty Ltd [43] , Deane J, as an example of the diminishing significance of the differences between the rules governing recovery of damages in tort and contract, referred to "the gradual assimilation of the tests of "within the contemplation of the parties" (in contract) and "reasonably foreseeable" (in tort)".
What is significant is that the obligation undertaken by the appellant was not to indemnify the respondent in respect of liability of the respondent to any third party. It is true that obligations of one party to indemnify another with respect to that party's liability to a third party stand in a special position as regards the effect given to settlements of disputes with that third party. The present was not such a case. The obligation of the appellant was to procure the provision to the respondent of a contract of insurance with secure and certain cover against certain risks. Nevertheless, considerable attention in argument was given to that category of cases involving an obligation to indemnify, and it is appropriate to give it further consideration so as to emphasise why the decisions in litigation within it are not indicative of the outcome in the present case.
A number of the decisions to which we were referred, particularly the decision of Devlin J [44] and the English Court of Appeal [45] in Biggin, concerned situations where A was obliged (by express stipulation or by implication of law) to indemnify B against the liability of B to C. For example, A may be the insurer of B in respect of claims made by C upon B, or A may be (as in Biggin) the supplier of defective goods to B for resale by B to C. In Biggin, it was not disputed that A ought to have foreseen the claim by C upon B. At the time of the contract between A and B, A had been informed by B of B's negotiations with C and of the purpose for which the goods were required by C [46] . Devlin J posed the issue as one of "the extent of the indemnity which the law grants [to B] by way of damages for breach of contract" [47] . Again, as illustrated by the decision of the Victorian Full Court in Wright v Langlands' Foundry Co [48] , A may have contracted with B to supply within a particular period goods or services to B for use in B's performance of the principal contract between B and C, and A's delay has placed B in breach of the contract with C.
[T]he plaintiff, having been put in the position of having to take all steps in connection with the litigation of the claims against him at his own risk, is entitled to recover, as damages, such sums as he paid to settle those actions, provided that he shows that he acted reasonably in making the settlement. 1. (1933) 34 SR (NSW) 88. 2. Edwards (1933) 34 SR (NSW) 88 at 98.
to determine whether the defendants should have regarded it as a serious possibility that one of the consequences of their breach would be that the plaintiffs would compromise the claim. Was the compromise a foreseeable consequence of the making of the claim? In my judgment, it was not a consequence in the legal sense at all. It flowed from the voluntary act of the plaintiffs. The Court of Appeal reached the opposite conclusion. 1. Biggin [1951] 1 KB 422 at 428.
A claim against an insurer is compromised for non-disclosure
The insured claimed, against the broker, the difference between the settlement sum and the sum which would have been recovered had the policy issued by the insurer been fully enforceable (the shortfall). That claim required a number of hypotheses to be made in order to succeed. First, that there was no substance in the insurer's alternative defences of arson by the insured and of under-insurance. (For the first of these, at least, the broker was not responsible and possibly for neither.) Secondly, that had there been disclosure of the insured's unpromising claims record, the insurer would not have issued a policy at all to the insured. Thirdly, that the settlement achieved in the negotiations between the insured and the insurer was, objectively, an accurate measure of the insurer's exposure to the insured in terms of the policy which it would have issued had there been no failure to disclose.
No evidence was called at the trial, either from the insurer or from anyone knowledgeable about its practice, as to what it would have done if the non-disclosure had not occurred, that is, if the broker had revealed the full claims history of the insured. There was, therefore, no affirmative evidence as to whether the insurer, accurately informed about the insured's claims history, would have issued a policy at all; issued a policy in the same or different terms; or issued a policy at a higher premium.
An expert in insurance broking, Mr Frank Fuller, gave evidence. That evidence was accepted by the primary judge. Mr Fuller was asked whether he could have "placed the [insured] with an ISR [Industrial Special Risks] insurer in 1992". He said "I would think so. I am not sure [at] what cost or on what terms but, yes, I would place the cover". He thought that the premium "would be a bit higher". In cross-examination, he acknowledged that "[m]aybe not all insurers", with disclosure of the insured's past claims history, would have provided cover. He was pressed with the question whether "a reasonable insurer" would issue such a policy. He answered affirmatively. He also affirmed that the insurer which issued the instant policy was "a reasonable insurer". None of this evidence was contradicted.
It was in this state of the evidence that the primary judge upheld the insured's claim to recover from the broker the entire shortfall constituted by the difference between what he found would have been recoverable under the policy and what was agreed had been recovered from the insurer following settlement. The broker complains that this result was founded on an incorrect approach, unexplored assumptions and the absence of relevant evidence.
The primary judge found that the only issue which ultimately affected the settlement, as between the insured and the insurer, was non-disclosure. Mr Pezzano had given evidence that, at the mediation conference preceding the settlement, non-disclosure was the only question raised. The insurer did not raise the issue of arson [65] . Mr Pezzano specifically denied that he had been advised that there was a risk that the insured might lose its case against the insurer entirely on the strength of the latter's allegation of arson. The trial judge noted that no admissible evidence had been tendered at the trial to suggest that a defence of arson would have succeeded or that the fact of arson could have been proved. In relation to the defence of under-insurance, the primary judge observed that the parties were agreed in relation to the figures to which the average provisions would have been applied by the insurer had there been no other dispute about liability under the policy [66] . He accepted those figures and adjusted the insured's entitlement accordingly. In this way his Honour, in the claim against the broker, confined his assessment of the "reasonableness" of the settlement by reference solely to the issue of non-disclosure.
The question is not whether the plaintiffs acted reasonably in settling the claim, but whether the settlement was a reasonable one. Therefore, even on the Biggin approach, the repeated references to the reasonableness of the insured's conduct in settling the claim involved a misdirection. However, in other parts of his reasons, the primary judge held that the settlement itself was "reasonable" or "not unreasonable" [72] . He concluded [73] : I am, in the end, satisfied that this [failure to consult the broker] did not render the settlement unreasonable in circumstances in which the [broker] was maintaining a complete denial of liability to the [insured], in which it had denied knowledge of the [insured's] prior claims history and in which it was ascribing to the [insured] the very act, which had given rise to the non-disclosure, of which it had itself been guilty [T]he [broker's] breach of duty (whether contractual or tortious) was such as to leave the [insured] exposed to a real risk of complete failure in relation to its claim against [the insurer] [T]he circumstances were such that it was reasonable for it to have sought to compromise that claim and that the compromise which it did in fact achieve was reasonable in all of the circumstances. 1. [1951] 2 KB 314 at 326. 2. Rocco Pezzano Pty Ltd v Unity Insurance Brokers Pty Ltd (1995) 8 ANZ Insurance Cases ¶61-288 at 76,213-76,214, per Steytler J. 3. Rocco Pezzano Pty Ltd v Unity Insurance Brokers Pty Ltd (1995) 8 ANZ Insurance Cases ¶61-288 at 76,214, per Steytler J.
The issue in the present case is somewhat different from that raised in Biggin. Here, no one contested that the settlement was relevant to fix the maximum recovery for the indemnity claimed by the insured. Here also, no solicitor or counsel were "submitted to cross-examination" to justify the settlement. Some of the considerations that seemed to have affected the primary judge in this case were the reasonableness of the conduct of the insured rather than the objective reasonableness of the settlement. In Biggin, only the latter was said to be relevant. Significantly, Biggin concerned a situation where A was obliged to indemnify B against the liability of B to C. In the present case, B (the insured) has rights against both A (the broker) and C (the insurer). Nevertheless, the discussion in Biggin of the question for decision is sufficiently pertinent to repay attention to the reasons of Devlin J and of the English Court of Appeal written nearly fifty years ago.
Devlin J, at first instance, stated his approach thus [77] :
[T]he general rule is that a defendant who is required to indemnify a plaintiff against his liability to a third party is entitled to have the existence and precise extent of that liability proved against him in proceedings to which he is a party. That does not mean that a plaintiff may not compromise with his adversary save at the risk of losing his indemnity or at the cost of offering the defendant a choice between the compromise or the proof. A plaintiff acts in such a matter at his own risk and for his own benefit. If he compromises erroneously where there is no liability, he cannot make the defendant pay for his error. If he compromises successfully, he will be that much in pocket. It is true that if the defendant pays him more than the compromise, the plaintiff will be paid more than he has lost. But that is a thing which is bound to happen whenever the nature of the indemnity if precisely defined, and the area of its operation consequently circumscribed, and an event takes place outside that area. Devlin J rejected the argument that the duty on the plaintiff to mitigate its loss extended to acceptance of the settlement [78] . He then dealt directly with the suggestion that the liability of a third party could be fixed by the opinions and estimates of counsel in settlement negotiations in which the third party was not involved [79] : It is conceded that counsel's views on the facts and law on which the existence of liability is based are irrelevant, and that that is a matter for the court. Why as a matter of logic are his views on the facts and the law relating to the quantum of liability any different? What happens if the court thinks that nominal damages only have been proved? Is it permissible to investigate the chain of counsel's reasoning to see if it contains any error? Would an incorrect appreciation of the facts be a fatal error, or must it be an error of law? If in the case of a large number of items counsel has erroneously included one, does that invalidate altogether a settlement for a global sum? And so on. The point is not, I think, improved by the alternative submission that the settlement made must be reasonable. If this means that the sum paid must be the reasonable value of the claim when all the material facts are taken into consideration, it can only be that which the court would itself have awarded, and the inquiry is neither shortened nor affected by the advice of counsel. Therefore, it is contended that reasonableness must be judged as at the time of the settlement and on the information then available. This means that instead of weighing the true facts the court would have to weigh only those, however incomplete and erroneous, which were known to counsel. What standard of reasonableness can be applied to that? 1. Biggin [1951] 1 KB 422 at 426-427. 2. Biggin [1951] 1 KB 422 at 428. 3. Biggin [1951] 1 KB 422 at 429.
In the Court of Appeal, the judges were at pains to demonstrate the error of Devlin J's conclusion that the settlement was wholly irrelevant. Their Lordships showed, clearly enough, that the quantum agreed fixed, in effect, a cap upon what the claimant could recover [80] . However, perhaps because of the terms of the headnote to the report of the Court of Appeal's decision, the latter has been accepted, since 1951, as standing for a much larger proposition. This is that the settlement, if reasonable, is a conclusive determination of facts, even where that issue is in contest by persons who were not parties to the settlement. As a matter of logic, and of common justice, that cannot be so.
The central plank of the broker's argument was that the reforms of insurance law effected in Australia by the Act made it impossible, or at least highly unlikely, that the insurer could, by reason of the accepted non-disclosure, have escaped liability in proceedings between the insured and the insurer.
At common law, material non-disclosure on the part of an insured was ordinarily fatal to the insured's claim under the policy if the undisclosed fact would have affected the decision of a prudent insurer [81] . The rule was straightforward but often unfair in its operation. The Australian Law Reform Commission, in its review of the law of insurance contracts (upon which the Act is based), pointed out that decisions on whether to accept or reject insurance frequently depended as much upon statistical data as upon the previous claims experience of the proponent [82] . On the recommendation of the Commission, the law in this regard was changed by the Parliament. The change narrowed substantially the circumstances in which an insurer could refuse indemnity under a policy for material but non-fraudulent failure to comply with the duty of disclosure. The relevant provision is s 28 of the Act. It reads:
(1) This section applies where the person who became the insured under a contract of general insurance upon the contract being entered into:
(a) failed to comply with the duty of disclosure; or
(b) made a misrepresentation to the insurer before the contract was entered into;
but does not apply where the insurer would have entered into the contract, for the same premium and on the same terms and conditions, even if the insured had not failed to comply with the duty of disclosure or had not made the misrepresentation before the contract was entered into.
(2) If the failure was fraudulent or the misrepresentation was made fraudulently, the insurer may avoid the contract.
(3) If the insurer is not entitled to avoid the contract or, being entitled to avoid the contract has not done so, the liability of the insurer in respect of a claim is reduced to the amount that would place him in a position in which he would have been if the failure had not occurred or the misrepresentation had not been made.
Arguments for accepting the compromise as conclusive
The law encourages the settlement of legal disputes. The courts could not cope if it were not the fact that the overwhelming majority of cases are compromised. This fact, evident in civil litigation, now finds reflection even in criminal proceedings, once resistant to plea bargaining [87] . Even more than in 1951, when Biggin was decided, the courts encourage consensual settlement of civil claims. Procedures (such as reference to referees, arbitrators and other non-court dispute resolution) and cost rules (such as penalties for rejection of settlement offers shown by later judgment to have been reasonable when made) [88] promote settlement as an indispensable attribute of a modern justice system. In part, this is out of recognition of the desirability of promoting agreements between parties rather than the imposition of a decision by a third party. In part, it arises from the increasing appreciation of the public, as well as the private, costs inherent in the litigation of claims to finality [89] . The substantive law is not untouched by these contemporary developments. Rules of the common law should promote, and not discourage, the lawful settlement of claims. Any rule which effectively discourages settlement or encourages parties to litigate issues to finality which they are content to settle would need to be closely examined. In the present case, to have obliged the insured to litigate its case against the insurer in order to avoid the contest now presented would (where reasonable settlement was otherwise achievable) have imposed burdens on the parties and the courts which the common law, as the product of reason, would not readily require.
The difficulties with the insurer were, as was found, the result of the broker's breach of duty. It was the broker's failure to disclose the claims history revealed to it by the insured which occasioned the difficulties in which the insured found itself. Had the broker disclosed that claims record, it would either have obtained cover from the insurer on the terms negotiated, or on other terms (as to premium and risk reduction measures), or it would have gone elsewhere to endeavour to secure the placement of the risk, possibly at a higher premium and on conditions. In these circumstances, the insured invoked the principle expressed by Lord Macmillan, in a different context, in Banco de Portugal v Waterlow & Sons Ltd [90] . His Lordship said that "if the party placed in a difficult situation by reason of the breach of a duty owed to him has acted reasonably in the adoption of remedial measures he will not be held disentitled to recover the cost of such measures merely because the party in breach can suggest that other measures less burdensome to him might have been taken". By analogy with this and other legal doctrines, the insured said that it scarcely lay in the mouth of the broker to criticise the best settlement which the insured could procure with the insurer when the cause and origin of the problem was the broker's own default.
The insured also relied on the fact that it was competently advised by senior counsel and that it had accepted this advice in settling with the insurer as it did. What more could the insured do, it was asked rhetorically, than to obtain appropriate legal advice and act on that advice when the risks advised, at that stage of the litigation, were that it might lose against the insurer altogether? This could occur if the insurer could show that, having regard to the claims experience, it would never have offered this policy to the insured and was therefore entitled to reduce its liability to nil. The offer of settlement was, by any account, a very significant sum. It was prudent for the insured to accept it. Objectively, it represented a "reasonable settlement", not merely as between the insured and the insurer but as against the world.
In the particular facts of the case, as found by the primary judge, there were some elements which lent credibility to the possibility that the insurer would have been entitled to deny indemnity altogether, notwithstanding s 28 of the Act. In an answer given by the broker to an interrogatory administered by the insured, the broker stated that no written reply had been received by it to a facsimile of 13 February 1991 sent to the insurer. This had sought a quotation for the placement of the insured's insurance business. The answer to the interrogatory went on to state that "it is believed by the [broker] that [the insurer] orally informed the [broker] that no quote was to be provided". In evidence, an officer of the broker was asked whether he had ever received a response to the facsimile. He replied "I don't recall receiving a response to that, no". The primary judge found this evidence to be compelling. He concluded that it was indicative of the fact that the insured's claims history was such as "to deter [the insurer] from quoting in respect of the insurance requested and of the fact that it did not take the history into account in later affording the [insured] the insurance provided under the cover note and the policy" [91] . On this basis, the Full Court concluded that the trial judge was in the best position to evaluate the facts as to the likely response of the insurer had there been full disclosure. He did so in the absence of specific evidence from the insurer, which no side had chosen to call.
As to who bore the forensic burden of tendering evidence from the insurer, the insured submitted that, once an apparently reasonable settlement was arrived at on appropriate advice, between the insured and the insurer, it was for the broker, if it wished to test the objective reasonableness of the settlement, to call evidence from the insurer (or other evidence of an equivalent kind) to displace the inference that the settlement was objectively the best that could possibly have been procured. In Biggin [92] , Somervell LJ remarked that the defendant might, by cross-examination, "seek to show - and perhaps successfully show - that [the settlement] was not reasonable. He may do so or call evidence which leads to the same conclusion. He might in some cases show that some vital matter had been overlooked". By parity of reasoning, it was urged that it was for the broker, and not the insured, to call any evidence from the insurer's camp. Its failure to do so gave rise to an inference that evidence supporting its attack on the "reasonableness" of the settlement could not have been procured, whether from the insurer or otherwise.
It is true that the law favours settlement. But it cannot do so at the price of a departure either from basic principles of law or the requirements of procedural fairness. It is fundamental to our legal system that, ordinarily, the party who asserts must prove. Where a claim is framed in the tort of negligence, the plaintiff must prove each ingredient of that tort: duty of care, breach and consequential damage. A failure to prove any ingredient will, in principle, be fatal to recovery. Whether the requisite causal connection has been shown between a particular breach of duty and particular loss or damage is essentially a question of fact. It is to be resolved on the probabilities as a matter of commonsense and experience [93] . If the claim is based on breach of a contractual duty to exercise reasonable care, and the plaintiff fails to establish that there was a cause of action which has been lost by breach, all that may be recovered is nominal damages [94] . Thus, claims in tort and contract (as here) equally postulate that the plaintiff must prove its loss. The burden of doing so, both evidentiary and forensic, is upon the plaintiff. Talk of the desirability of settlement is directed at the position as between A and B. Neither as a matter of logic nor of fairness can settlement between A and B govern the rights of A against C, unless C was party to that settlement or otherwise accepts what A and B have done. Biggin, in the Court of Appeal, correctly held that the settlement sum was not irrelevant because it fixes the upper limit of what may be claimed as an indemnity [95] . Often, for the avoidance of costs (now including possibly penalty costs), a third party in the position of the broker will accept such a settlement as defining the ambit of its liability. It may be more likely to do so if it is consulted in the negotiations and especially if its concurrence is procured [96] . But the desirability of settlement, and the modern laws and practices which facilitate it, cannot relieve a court of the duty of resolving lawfully and justly contested issues which are tendered for decision. That duty arises from the nature of a court as a court. Like Gummow J I regard this as rooted in a basic constitutional principle. Important as considerations such as efficiency, cost-savings and the promotion of settlement may be, they cannot detract from a court's duty in this regard [97] . Settlements will continue to be secured because of the pressure of court business, the saving of costs and the parties' evaluation of their own best interests. But such settlements must be achieved within a legal system which determines contested issues on the evidence and, relevantly, determines whether a party asserting damages has proved the damage.
As to the complaint that the broker was the occasion of the insured's predicament, so much may be conceded once it is found that the broker was in breach of its duty of care to, or contract with, the insured. But that merely sets the scene for the elucidation of the damage (if any) which flows as a result. Damage being disputed on the pleadings and resistant to compromise, it had to be proved as any other claim must be: by the tender of relevant evidence and the application to the facts found of the applicable principles of law. When the common law governed the liability of insurers, in cases of material non-disclosure, the position was much simpler. Then, indeed, proof of a negligent omission by a broker to notify the insured's adverse claims history (of which it had been informed) would, with very few exceptions, have given rise to an inference that the rejection of indemnity by the insurer would be the consequence of the broker's default. Now, however, the law has moved along. By the Act, material non-disclosure (except in case of fraud) is no longer fatal to an insured's claim. The insurer is specifically disentitled from avoiding the contract on that basis (s 28(3)). A hypothetical exercise must be carried out. That exercise is addressed not to the consideration of what, by the application of the pressures and evaluation of the risks of litigation, the insurer is willing to pay to avoid the risk of being found liable to indemnify the insured in the entirety promised under the policy [98] . Instead, the inquiry is concerned with something more specific and normative. This is to discover "the amount that would place [the insurer] in a position in which [it] would have been if the failure had not occurred or the misrepresentation had not been made" (s 28(3)). This is quite a different inquiry from that posed by the question whether a settlement negotiated at the door of a court was, or was not, "reasonable". So far as I can see, the last-mentioned question is irrelevant to the issue presented by the terms of s 28(3) of the Act. It is concerned with an inquiry which takes the decision-maker into the practice and likely response of the insurer which has issued the contract of insurance in question. The cases [99] and the textbooks [100] agree that s 28(3) is concerned with the practice of the particular insurer. In effect, the sub-section requires the elucidation of what that insurer would have done had the non-disclosure not occurred. The mere assertion of the insurer does not have to be accepted. In many cases, those seeking to unravel this point secure discovery. They try to find, from the treatment of similar cases, the way in which (if at all) the insurer in question would probably have been prepared to deal with the case in hand. Evidence of the insurer's practice lies at the heart of the issue presented by s 28(3) of the Act [101] . No doubt, if the particular insurer has no practice or is newly licensed or has lost its files, the matter is left to inference. But evidentiary silence on the point is perilous. The silence is not, in my view, filled by the sounds of a settlement achieved long after the proposal for insurance was tendered, in circumstances where the non-disclosure (and possibly other problems) are well identified and where the often unequal bargaining power of the parties looms larger than the hypothetical legal question which s 28(3) presents.
As to the suggestion that a settlement between the insured and the insurer must bind the broker (although it did not know of, or participate in, the negotiations) because the insured was so advised by senior counsel, I would answer with the response to a like proposition given by Devlin J in Biggin. The decision on settlement falls to be made at one point of time. The case against the broker falls to be determined on the facts available at a later point of time, when more and different information may be available to the court. Why should the broker be bound in law by the judgment of counsel not appearing in its interests? Why should a court of law deciding the claim against the broker be bound by counsel's views? I make no reflection on counsel advising in this case. But the proposition must be tested as one of universal application. What if counsel advised the settlement because of problems with other commitments? What if counsel were a timorous soul who found the burdens of trials irksome? What if counsel's mind had not fully grasped the new regime introduced by s 28 of the Act but continued to address the problem of non-disclosure on assumptions resting on the common law rather than the novel provisions of the Act [102] ? None of these defaults was suggested in this case. But the possibility that they could exist along with other extraneous explanations of a settlement, serves to illustrate the unwisdom, and potential injustice, of embracing a principle which forfeits the lawful proof and determination of claims against a third person to a settlement, advised and effected between two other persons.
It is true that there were, for default of evidence from the insurer, snippets of facts to which the parties severally referred in order to address the issue presented by s 28(3) of the Act. Thus, the insured laid emphasis upon its own unattractive claims history and the failure of the insurer, in 1991, to respond to an invitation to bid for its insurance business. The broker emphasised the testimony of the expert witness that a reasonable insurer would have offered insurance, although at a higher premium, and that the particular insurer here was a reasonable one. These generalities fall far short of the evidence necessary to establish the insurer's practice such as would ordinarily be required in the case of contest to address the issue posed by s 28(3). They leave too much to guesswork and speculation.
The question, then, is ultimately which party bore the burden of calling the evidence upon the basis of which a proper conclusion could be reached to answer the hypothetical question presented by s 28(3). Clearly, the insured considered that it had done enough by proving that it had settled against the insurer on legal advice. For the reasons which I have given, I do not believe that such evidence answered the question presented by s 28(3). True, it proved what the insurer, in respect of a claim, would actually pay following the failure of the insured to comply with the duty of disclosure. But it did not prove the reduction that should be made, in lieu of avoidance of the contract of insurance, to place the insurer in the position it would have been in if the failure had not occurred. That being an element in the calculation of the insured's loss in its claims against the broker, it was simply not established by the evidence tendered at the trial. Upon a matter in respect of which it bore the burden of proof, the insured therefore failed to establish an ingredient of its claim. Subject to what follows, the claim based upon the tort of negligence should have been dismissed. The claim for breach of contract should have attracted no more than nominal damages.
In its notice of appeal the broker initially sought only an order allowing the appeal and dismissing the insured's action against it. Subsequently, as a result of argument before this Court, the broker sought orders, in the alternative, that the judgment below be set aside and in lieu thereof judgment for the [insured] for damages of $1 or alternatively $9,106 be paid in the contract claim or that the action be remitted to the Supreme Court for further hearing according to law.
The first of these alternatives took into account the fact that, even on the broker's own case, it was accepted that breach of the contractual (and tortious) duty of care was established. The insured would therefore be entitled, at the least, to nominal damages for breach of contract. Given the established fact of the payment to the insured of the settlement sum, it is not plain why, or on what footing, the broker would be liable to pay the insured the increased premium applicable in respect of a like policy of insurance issued by the insurer or another insurer.
The second alternative arose from questions during argument in which it was suggested that, in the approach taken at the trial, there had been no real trial of the issue of damages. At least, there had been no trial addressed to the legally relevant question. The broker added this form of relief to its notice of appeal reluctantly. It primarily urged that the insured had had its chance, had elected to present its case against the broker in a particular way and should not now be permitted to reformulate its case and start again. There is force in that submission. Ordinarily, a party will be bound by the way in which it presents its case. The interest of finality in litigation argues against the reagitation of questions simply because it has been demonstrated that a party failed to prove a claim in the way the law requires.
As against this consideration, it may be accepted that this Court enjoys the powers of the Full Court of the Supreme Court of Western Australia. It should exercise those powers as that Court ought to have done, had it correctly identified the error in the approach of the primary judge which I would find. By O 63, r 10(2) of the Rules of the Supreme Court 1971 WA, the Full Court has the power "to give any judgment, and make any order which ought to have been made, and to make such further or other order as the case may require".
This large power must be exercised in a way consistent with the trial process and with the principle of finality. In fairness to the insured, it may be said that the decision of the English Court of Appeal in Biggin had stood for a long time, had been applied in several Australian cases and approved by textwriters. It was, therefore, perhaps understandable that the insured should have approached the proof of its claim against the broker as it did. As against that, there is no doubt that the broker, at the trial, signalled its submission which I have now upheld. It said, in terms, that the broker would not be actually negligent if it had arranged a policy which the insured was entitled to enforce against the insurer. The burden of proving all elements of the claims, including damage, rested on the insured. Moreover, after judgment was given, the broker sought, in a context where the determination of the quantum of damages was outstanding, to reopen the issues at trial relevant to the reasonableness or otherwise of the settlement between the insured and the insurer. In particular, the broker applied, whilst the matter was still before the trial judge for final orders, to lead evidence about the issue of arson and to have produced to the Court letters or documents disclosing advice given by the insured's solicitors or counsel before the insured settled the case with the insurer [104] . As these questions were not explored in the Full Court's reasons, and as the findings on arson (and the average provisions) were not attacked on appeal to the Full Court [105] , it must be assumed that the broker's application to reopen was not pressed. There was no application by the insured to reopen.
In my opinion, the issue of the insured's entitlement to damages against the broker was never properly tried. Although on one view, this was by reason of the election of the insured and despite clear signals from the broker, upon another view it was because of a lack of clarity in the law which should not now prevent a trial being had which addresses the correct legal questions. Had the Full Court reached a similar view, it is provided with ample power by its Rules to permit amendments to be made so as "to ensure the determination on the merits of the real question in controversy between the parties" (O 63, r 10(3)).
As the matter remains alive, because of the previous uncertainty of the applicable law and because the insured has established every ingredient of its claim except damages, the question of the damages (if any) to which the insured is entitled against the broker should be remitted for trial in conformity with the opinions which I have expressed. At such trial, the insured may indeed prove that its apprehensions were correct and that the insurer, by its established practice, would not under any circumstances have issued the policy to it had non-disclosure not occurred. The insured may be able to convince the Court that a proper application of s 28(3) of the Act required reduction of the insurer's exposure to nil. If that were done, then subject to any other amendments which were permitted, the judgment entered in the first trial would once again follow. But then, at least, it would be upheld on the basis of evidence tendered by a plaintiff to prove an essential ingredient to its claim, disputed by the defendant. The latter would, in the normal way, have the opportunity of testing the propositions. It could call its own evidence in response. In the way this trial occurred, those facilities were denied to the broker incorrectly and, in my view, unjustly.
The appeal should be allowed with costs. The judgment and orders of the Full Court of the Supreme Court of Western Australia should be set aside. In lieu thereof, it should be ordered that the appeal to that Court be allowed with costs; and the judgment and orders of the primary judge set aside. The proceedings should be remitted to the Supreme Court of Western Australia for retrial limited to the amount of damages (if any) to which Rocco Pezzano Pty Ltd (the plaintiff) is entitled to recover from Unity Insurance Brokers Pty Ltd (the defendant). The costs of the first trial should abide the outcome of the second.
Rocco Pezzano Pty Ltd, the respondent, leased premises at East Cannington in Western Australia for use in its business of selling fruit and vegetables. In December 1990, that company (the insured) appointed Unity Insurance Brokers Pty Ltd, the appellant, as its insurance broker. The appellant (the broker) arranged an industrial special risks policy for the insured with NZI Insurance Australia Ltd (the "insurer" or "NZI") for the period from 29 June 1992 to 29 June 1993. One of the risks covered by the policy was damage to the insured's premises at East Cannington, and its plant, machinery, stock and contents, by fire.
On 20 September 1992, those premises were extensively damaged by fire and plant, equipment, stock and contents were damaged or destroyed. The insured claimed on the policy but in February 1993 the insurer denied liability. The letter to the insured from the insurer's solicitor said that the insurer denied liability "by reason of material non-disclosure of prior claims" but reserved the insurer's right "to rely on other grounds for denying indemnity".
The insured sued the insurer and the broker, alleging that the insurer was not entitled to deny liability but that, if it was, the broker was responsible for any non-disclosure. In June 1993, the insurer filed an amended defence in the proceeding adding two further grounds of defence: first, that the insured's loss was caused by arson arising through its own deliberate conduct and second, that the insured had not insured for the full value of the insured property and the amount payable under the policy should be reduced according to the formula prescribed by the policy. (Application of that formula would, so the insurer alleged, have reduced the amount payable under the policy to $928,293.62, less an excess of $200. By contrast, the trial judge found that, had the insurer otherwise admitted liability under the policy, it would have paid the insured $1,720,287.04.)
On 24 January 1994, the insured compromised its claim against the insurer, accepting $900,000 in full settlement of the claim. The insured continued its proceeding against the broker.
An order was made that the issue of liability in the proceeding between the insured and the broker should be tried separately. On that trial, it was not disputed that the insured had not made full disclosure to the insurer of its prior claims history on its proposal for this insurance policy. Nor was it disputed that the broker had completed some parts of the proposal, including the part relating to prior claims history. There was a dispute whether a full account of the insured's claims history, which had been supplied to the insurer in 1991, but which was no longer complete by 1992, was sufficient satisfaction of the insured's duty of disclosure but that question was resolved against the broker. The present appeal was conducted on the basis that there was not full disclosure of the insured's claims history. Similarly, although it was contended at trial that the broker had not breached its duties to the insured, it was not disputed on the hearing of this appeal that the broker had failed to perform its retainer with reasonable care and skill by completing the proposal for the policy without disclosing to the insurer the claims history of the insured: a claims history which it knew. (It matters not, for present purposes, whether that breach was a breach of contract or a tortious breach of a duty of care.)
The trial judge held that the broker's breach of duty "was such as to leave [the insured] exposed to a real risk of complete failure in relation to its claim against [the insurer], that the circumstances were such that it was reasonable for it to have sought to compromise that claim and that the compromise which it did in fact achieve was reasonable in all of the circumstances". He therefore gave judgment for the insured against the broker for damages to be assessed. Subsequently, the parties having agreed on the facts relevant to the assessment of damages, the trial judge gave judgment for the insured for $1,041,166. The broker appealed to the Full Court of the Supreme Court of Western Australia but that appeal was dismissed. It now appeals to this Court.
At trial, the insured called no evidence from any officer or employee of the insurer. Thus there was no direct evidence adduced by the insured of what effect, if any, full disclosure to the insurer of the insured's claims history may have had on that insurer. The Insurance Contracts Act 1984 Cth provides, in effect, that an insurer may avoid the contract if there has been a fraudulent failure to disclose or a fraudulent misrepresentation by the insured (s 28(2)), but that if the insurer is not entitled to avoid the contract (or being entitled to avoid it, has not done so) "the liability of the insurer in respect of a claim is reduced to the amount that would place him in a position in which he would have been if the failure had not occurred or the misrepresentation had not been made" (s 28(3)). The insurer had pleaded that the insured had made a material non-disclosure by failing to disclose its claims history and also pleaded that the insured had misrepresented its claims history. The pleading was treated at the trial of the proceeding between insured and broker as not alleging fraud. Accordingly, the proceeding between insured and broker was conducted on the basis that the insurer had not been entitled to avoid the contract of insurance because of the failure to disclose the prior history but its liability was reduced under s 28(3) of the Insurance Contracts Act.
There being no direct evidence from any officer or employee of the insurer, the broker submitted that the insured had failed to show what the position of the insurer would have been if the true claims history of the insured had been disclosed. Accordingly, so the argument ran, the insured had failed to show that it had suffered any loss as a result of the broker's breach of duty.
Much emphasis was placed in the submissions on behalf of the broker upon what was asserted to be the necessity to call evidence from the insurer. But to focus upon the manner of proving a particular proposition (in the sense of focusing upon who should be called to give evidence about it) may distract attention from identifying what it is that the insured had to prove to succeed in its claim against the broker. It was submitted on behalf of the broker that the insured could succeed in a claim for more than nominal damages for breach of the contract of retainer (and could succeed in its claim for negligence) only if it established that it could not have recovered from the insurer any more than it had obtained by way of settlement. Thus, it was submitted, it was necessary for the insured to prove against the broker the case which the insurer had pleaded against its insured. In this case, it was said that the insured had to prove that the insurer would have been entitled, under s 28(3), to reduce its liability to the insured to nil - on the basis that the insurer would not have undertaken the risk if the insured had made full disclosure [106] . Nothing less than that would suffice, it was submitted, because there was evidence at the trial that, if there had been full disclosure, reasonable insurers would nevertheless have given the cover sought by the insured for a premium only marginally higher than the premium that was in fact charged by this insurer. Thus, so the argument went on, if the insured had prosecuted its claim against the insurer to judgment, the only consequence of the insured's non-disclosure would have been a comparatively small reduction in the amount payable under the policy - a reduction much less than the discounting of the claim by about 40 per cent or 50 per cent that the settlement of $900,000 represented.
Whether the compromise of a claim was reasonable must be judged objectively, not subjectively. Thus whether a party to litigation has received advice to settle may be important in deciding whether that person's conduct in settling the case was reasonable but, standing alone, the fact that a litigant was advised to settle at a particular figure reveals little or nothing about whether the settlement reached was reasonable. This is not to say that evidence may not be led that such advice was given and adopted; it may. But evidence of that kind does not conclude the issue. What will usually be much more important is the reasoning that supported the advice that was given for that will ordinarily reveal why it was thought reasonable to compromise the claim as it was.
Next, the question whether the settlement was reasonable must be judged by reference to the material the parties had available to them at the time the compromise was reached. It is not to be judged according to whether material which was obtained later shows that the opposite party could or could not have prosecuted or defended the claim successfully but according to the assessment which could properly be made at the time of settlement of the chances of success or failure.
Often that will require consideration of whether the party that later seeks to say that the settlement was reasonable had made sufficient inquiries and had sufficient information available to it to warrant reaching a compromise. In turn that may invite attention to whether the cost of seeking further information would outweigh the benefit that it was reasonable to expect may be obtained from doing so, but it does not assume knowledge of the opposite party's brief to counsel.
All of these, and no doubt other, considerations may bear upon the question whether the settlement arrived at was reasonable. And it is inevitable that there will be no single answer to the question "for what amount was it reasonable to compromise this claim" - there will be a range of answers. What is a reasonable compromise of the claim will almost always require consideration of the chances of the parties succeeding in their respective claims or defences and that prediction of likely outcomes must always be imperfect and imprecise. To state the obvious, that is why the compromise of a claim, which is a monetary claim that will succeed entirely or fail entirely, will usually fasten upon a figure that is less than would be recovered if the claim were to succeed and why it is that there will be a range of figures within which the reasonable observer may conclude that settlement of the claim would be reasonable.
No doubt this may be contrasted with the case which is fought to judgment. Then the liability of the party in the position of the insurer in this case would be fixed by the judgment. The broker's argument in this case (and the argument which found favour with Devlin J at first instance in Biggin & Co Ltd) is that its liability should be fixed by reference to that sum and no other and that if the liability of the insurer has not been fixed by judgment in an action brought against it, it should be fixed in the action between insured and broker.
I do not agree. First, as I have indicated, acceptance of this argument would discourage settlement of the claim between insured and insurer. Secondly, to subject the broker to liability based upon a settlement that is found to be reasonable is not unjust. And it is not unjust even though there may well have been a range of figures within which settlement could reasonably occur and even though the decisions whether to settle and at what figure to settle are decisions over which the broker has no control. It is always necessary to recall that the broker was in breach of duty. There is no injustice in leaving the wrongdoer to bear the consequences of the decisions made in response to that wrongdoing by the party harmed - so long as those decisions are reasonable. Reasonableness informs much of the law of contract and, in particular, the assessment of damages for breach [115] . This means, for example, that if the party wronged has acted reasonably, the wrongdoer may be liable for all the loss that the plaintiff has suffered, even if the plaintiff's conduct has increased the loss [116] . Conversely, the party wronged is not bound to take all possible steps to mitigate its loss, only those steps which are reasonable [117] .
Considerations of legal professional privilege are not the only considerations that may affect the evidence that may be led. If the settlement has been reached at a mediation, the rules under which the mediation was conducted, whether rules of court or privately agreed rules, may restrict what may be revealed about what occurred at the mediation. But that may mean only that the reasonableness of the settlement reached may have to be demonstrated without resort to that material. If that is a problem, it affects the party contending that the settlement is reasonable; it is not a problem that causes any injustice to the party against whom the evidence of the settlement is to be led.
I have already referred to the finding of the trial judge in this matter that the settlement which the insured reached with the insurer was reasonable. There were several strands in the reasoning in support of that conclusion (reasoning which was substantially approved by the Full Court) to which reference should be made.
As I have already noted, the insurer relied, in its amended defence, not only on defences of non-disclosure and misrepresentation but also on a defence of arson and a defence based on the average provisions of the policy. Only the first of these defences - of non-disclosure and misrepresentation - concerned conduct of the broker. The trial judge found that the insured was motivated, in arriving at the settlement which it did, solely by its concern that the insurer might succeed in the defence of non-disclosure. In particular, he found that the defence of arson played no part in persuading the insured to settle with the insurer and that it did not influence the amount for which the insured was prepared to settle. As for the question of average, he found that, although the insurer and insured disputed the values to which the relevant provisions should be applied, the insured settled the claim having regard to the figures which it propounded and not those put forward by the insurer. Accordingly, the trial judge found that the dispute between insured and insurer about these figures was not material to whether the settlement was reasonable.
It follows that the case which the insured sought to make at trial was that, having regard only to the defence of non-disclosure, the settlement which it reached with the insurer was reasonable. That is, even though the insurer's pleading of other grounds of defence (arson and average) might have been said to have injected further doubt into the fate of the litigation, and thus, have warranted a further discounting of the insured's claim against the insurer, the insured contended, in its claim against the broker, that those defences of arson and average should be entirely disregarded in considering whether the settlement was reasonable.
The trial judge's finding may be read as focusing on the subjective reasoning of the insured - the insured did not take certain matters into account in deciding whether to settle for the amount ultimately agreed. As I have said, the test to be applied is an objective test; the subjective reasoning of the insured does not determine whether the settlement was reasonable. What is important for present purposes is the question whether, having regard only to the defence of non-disclosure, the settlement was reasonable.
The trial judge concluded that the insured was entitled to assume that, at any trial between insured and insurer, the insurer would "lead evidence in support of its contention that, had full disclosure of the claims history have been made, it would not have entered into the contract of insurance" and that the insured was not bound, in its action against the broker, to lead evidence from the insurer to the effect that its assumption was correct.
The first of these conclusions may be read as emphasising what the insured thought. As I have said, the question is whether the settlement was reasonable, and that is not concluded by identifying whether the parties settling thought it was or why they thought it was. As to the second of these propositions, I accept that the insured was not bound to call the insurer to show that the insurer would have declined the risk but, in order to demonstrate that settling its claim against the insurer for $900,000 was reasonable, the insured did have to show that there was a real risk that the insurer could have satisfied a court that it would not have entered the contract of insurance if the true claims position had been disclosed. It was this, and only this risk, which was said to warrant the insured accepting such a large discount on its claim against the insurer. (Showing that this, or another, insurer might have charged a higher premium would not have been enough in the particular circumstances of this case. The only evidence about this aspect of the matter was evidence from an experienced insurance broker that, if full disclosure had been made, some insurers would have accepted the risk but at a premium which he described as "a bit higher".)
The insurer had pleaded that if full disclosure had been made it would not have taken the risk. The fact that the defence had been taken was the first step in demonstrating that there was a risk that the insured may have failed in its claim on the policy, but the fact that the defence was taken says nothing about what chance it had of succeeding. At most it showed that the legal practitioner who prepared the pleading had instructions sufficient to warrant drawing the pleading in the form it took. No larger assumption could be based on the form of the pleading.
As the Full Court pointed out, there was evidence led at the trial which was said to indicate the attitude of the insurer to the risk. In February 1991, the broker attempted to procure industrial special risks insurance on behalf of the insured over its premises, plant, equipment and the like for the 1991-1992 year. The broker disclosed the insured's then full claims history. It asked NZI and other insurers for a quotation but there was no record of NZI providing any such quotation to the broker and an employee of the broker gave evidence that he did not recall receiving a response to the request for quotation. In addition, the broker gave an answer to an interrogatory to the effect that the broker believed that "NZI orally informed the [broker] that no such quote was to be provided". This answer was tendered in evidence. Finally, the request for quotation sent by facsimile to NZI was produced by NZI in answer to a subpoena and was also tendered. That request bore markings that appeared to read "no quote". This evidence, taken together, warranted the finding that NZI had refused to offer a quotation for the insured's business in February 1991.
There being evidence that the insurer had refused the risk in 1991, I consider that it was open to the trial judge to find that there was a real risk that the insurer would have proved that it would have refused the risk in 1992 if full disclosure had been made of the claims history. (During 1991 there had been two further claims and thus the insured's history was no better than it had been in 1991 when the insurer refused the risk.)
Was there, then, enough material to show that the settlement was reasonable? The trial judge and the Full Court held that the evidence that was led, exiguous as it was, was sufficient to show that it was reasonable and I am not persuaded that that conclusion was wrong. The settlement discounted the insured's claim by between 40 and 50 per cent. Presumably, then, the insured is to be taken to have estimated the chances of the insurer succeeding in its defence as being about that level. It is as well to remember, however, that there are uncertainties inherent in litigation and that predictions of the chances of success in litigation can never be precise. It follows that the comparison between the amount of the settlement and the amount of the claim can never be anything more than a general indication of what the parties see as the risks of continuing the litigation. There was evidence available to the insured, at the time it compromised with the insurer, that the insurer had refused to deal with it in the previous year and the insurer alleged that if it had known all the facts it would not have dealt with the insured in the year in question. No doubt it was on this basis that senior counsel retained to advise the insured about the possible compromise told the insured that it should settle "because the way things were [the insured] would lose" and that it was better to accept the sum of $900,000 rather than "losing everything because there was no disclosure" of the claims history.
Unity Insurance Brokers Pty Ltd
Rocco Pezzano Pty Ltd
(1972) 127 CLR 454
(1986) 161 CLR 653
(1995) 182 CLR 1
(1968) 120 CLR 516
(1987) 9 NSWLR 310
(1968) 120 CLR 635
(1991) 171 CLR 506
(1991) 28 FCR 68
(1982) 150 CLR 258
(1976) 51 ALJR 170
(1997) 191 CLR 85
(1989) 166 CLR 606
(1994) 179 CLR 332
(1988) 166 CLR 351
(1969) 121 CLR 432
(1991) 174 CLR 64
(1991) 24 NSWLR 564
(1980) 144 CLR 253
(1995) 184 CLR 281
(1987) 8 NSWLR 514
(1991) 22 NSWLR 389
(1993) 176 CLR 332
(1989) 97 FLR 284
(1993) 10 WAR 322
(1979) 46 FLR 386
(1992) 28 NSWLR 593
(1995) 38 NSWLR 404
(1997) 189 CLR 146
(1992) 58 SASR 102
(1993) 176 CLR 344
(1992) 176 CLR 408
(1954) 90 CLR 613
(1990) 29 NSWLR 569