Norman Hay PLC (in Members’ Voluntary Liquidation) v Marsh Limited

“Grounds (a) and (b) cover statements of case which are unreasonably vague, incoherent, vexatious, scurrilous or obviously ill-founded and other cases which do not amount to a legally recognisable claim or defence … .”

“Paragraph 1.4 of the PD (Striking Out a Statement of Case), para.3APD.1, gives examples of cases where the court may conclude that particulars of claim disclose no reasonable grounds for bringing the claim: those claims which set out no facts indicating what the claim is about; those claims which are incoherent and make no sense; and those claims which contain a coherent set of facts but those facts even if true, do not disclose any legally recognisable claim against the defendant.”

“If the Insured Person becomes legally liable to pay damages in respect of:

accidental bodily injury (which shall include death, illness and disease) to any person; and/or

accidental loss of or damage to material property, occurring during the Period of Insurance and arising out of a Journey”.

“Again, generally speaking, in order to claim under a liability policy where the insured has settled the claim of the third party, the insured still has to demonstrate that it was or would have been liable to the third party. It cannot simply rely on the fact of the settlement to demonstrate either liability or that the amount of the settlement was reasonable. In order to show the settlement was reasonable, the insured must show that the amount of damage for which it would have been liable is at least as much as the amount paid under the settlement.”

“an insured can only be obligated to pay sums by law if there is an actual liability to do so. One cannot be obligated by law to pay sums if there is only an alleged liability”.

“I will not consider any of their evidence on what a jury would be likely to find or could have found on particular topics. I have to decide what the outcome of the case would have been, applying Texas law and procedure to the facts as I find them and acting as both judge and jury”.

Under Texas law the assured was not legally liable in tort so the assured was not entitled to an indemnity from its liability insurers.

The Sage Settlement Sum was paid by the Claimant … from funds held in an escrow account which had been constituted as part of the sale of the Claimant's former subsidiaries to the Buyer.

…

But for the Defendant’s breaches of duty as set out above, the Claimant and/or IMP would have had in place valid and effective insurance cover that would have provided an indemnity in respect of liabilities to Ms Sage arising from the Accident such that the Claimant would not itself have had to fund the settlement of those liabilities via the Sage Settlement.”

Mr Shapiro KC submitted that this is wrong since it was Quaker which paid the settlement sum since, pursuant to Clause 9 of the relevant sale contract, the sums paid into escrow were paid by Quaker, not Norman Hay.

“What damage they have suffered does not depend upon whether Eagle Star would have been entitled as a matter of law to repudiate liability under their standard policy, but whether as a matter of business they would have been likely to do so. What the employers have lost is the chance of recovering indemnity from the insurers. If Eagle Star would not have been entitled to repudiate liability in law, cadit quaestio: the damages recoverable would amount to a full indemnity. Even if they would have been entitled in law, however, to repudiate liability, it does not in my view follow that the employers would be entitled to no damages. The court must never consider in that event, what were the chances that an insurance company of the highest standing and reputation, such as Eagle Star, notwithstanding their strict legal rights, would, as a matter of business, have paid up under the policy.”

“The fundamental principle governing the measure of damages is that the claimant should be put, so far as money can do so, in the position he would have been in had the defendant discharged his duty. In claims against insurance brokers, the claimant typically alleges that he was uninsured when, but for his broker’s negligence, he would have been insured. Therefore the main (and often the only) item of damages claimed is the amount which would have been payable by the insurers (or reinsurers) but for the broker’s breach of duty. If there is no doubt that the insurers (or reinsurers) would have satisfied the client’s claim, then this loss is plainly recoverable.

In assessing the claimant’s loss, the court is not strictly concerned with what the insured was entitled to recover under the relevant policy of insurance (where some policy was arranged). Instead, the court has to assess, on the balance of probabilities, what would have occurred had there been no breach of duty by the broker. Consequently, if the court finds that an insurer would or might have made a payment to the claimant but for the broker’s negligence, then the claimant will recover damages even if (as a matter of law) the claimant would not have been entitled to any payment from the insurer. The court will assess the likelihood that the claimant would have received a payment from the insurer. If, as a result of the broker’s negligence, there is uncertainty as to the claimant’s likely recovery from the insurer, then such uncertainty will be resolved in favour of the claimant.”

“Loss of a chance has appeared in actions against insurance brokers by their clients for inadequately insuring them. These claims have from time to time been met with the argument that no action lay because the clients would not have been able to claim on the policy which should have eventuated on the ground that some fault on the client’s part would in law have allowed the insurance company to repudiate the policy. The courts, however, have seen the issue as one of the degree of chance that the insurance company would in practice have repudiated on account of the insured’s fault. In the result they have in some cases held that in practice there would have been no repudiation and have allowed full recovery, as in Fraser v B.N. Furman (Productions) and Dunbar v A. & B. Painters, while in others the full claim has been discounted, by one third in Everett v Hogg Robinson, and by 70 per cent in O. & R Jewellers v Terry and Jardine Insurance Brokers. It was necessary in this last case to assess the chance not only of whether the insurance company would have taken the defence on the policy available to it but also of whether, had the broker’s duty been performed, the claimant would have had any policy at all. It was the assessment of the loss of this double chance which led to the very substantial discount there. … .”

“It is not enough for the claimant to show that the broker was negligent: he must in addition show that the broker’s negligence caused the loss he is complaining of. So a client cannot complain if the insurance that should have been obtained would not in fact have covered him in the events that happened, and a fortiori if he would have been uninsurable even if the broker had acted properly. Again, if it is plain that, even assuming the broker had placed the risk properly, the underwriters would have rightly refused to pay anything for some unconnected reason, then there will be no substantial damages. However, this is a difficult plea to sustain: in the absence of convincing evidence that the claim would not in fact have been paid, the claimant is likely to succeed.

In many cases of the above type, it will be found that there was some likelihood that the insurer would in fact have paid even if not liable to do so, whether ex gratia or for other good commercial reasons. Since the claimant can recover only his actual loss, it follows that in such a case damages may have to be reckoned on the basis of the loss of a chance of receiving payment, on the principle of the decision in Allied Maples Group Ltd v Simmons & Simmons. So in Everett v Hogg Robinson Ltd, brokers negligently made a misrepresentation to the underwriters which the latter correctly invoked to avoid liability. The brokers argued that the underwriters would, even absent the misrepresentation, have repudiated on the basis of the assured’s undisclosed claims record, but Kerr J found that the underwriters would probably have compromised the claim for commercial reasons, and merely reduced the award by one-third to take account of the possibility that they would not have. Where the claimant’s putative right against the underwriters is uncertain – for instance where it is arguable whether non-disclosure would have allowed repudiation at all – and the assured reasonably accepts a settlement of his claim, he is it seems entitled to recover the difference between the amount received under the settlement and the full amount.”

“indemnify the Buyer against and covenants to pay the Buyer an amount equal to the Recoverable Liabilities when the Buyer or any Group Member incurs or suffers, in each case arising out of, or in connection with, the Indemnified Incident.”

“the motor accident which occurred on 22 November 2018 in Ohio, US and in which Mr Nigel Kelsall, a sales director for certain Members was fatally injured and in which Ms Heather Sage was injured.”

“all Losses, liabilities, claims, Costs, damages and expenses that the Buyer or any Group Member does or will incur or suffer, all claims or proceedings made, brought or threatened against the Buyer or any Group Member by any person and all Losses, liabilities, claims, Costs, damages and expenses the Buyer or any Group Member does or will incur or suffer as a result of defending or settling any such actual or threatened claim or proceeding.”

“Once any Relevant Claim which is the subject of any Held-over Amount has been agreed or determined, then the Seller and the Buyer shall jointly instruct the Escrow Agent to pay the relevant Withdrawal Amount to the Buyer out of such Held-over Amount, and then pay the remainder (if any) of the Held-over Amount to the Seller, in each case from the relevant Escrow Account.”

“… Typically a claimant in a broker’s negligence action will have suffered a loss for which it is not properly insured (because no insurance was in place at all, or because the policy in place did not respond to the loss for some reason, or because the sums insured were inadequate). Usually, the claimant’s allegation against its broker is that, but for the broker’s negligence, it would have been properly insured and would have received a payment from an insurer for some or all of its loss.”

“the fact that the terms of a settlement were entered into upon legal advice establishes, at least, that those terms were prima facie reasonable. It is then for the defendant to displace that inference by evidence to the contrary, by establishing, for example, that some vital matter was overlooked … .”

Norman Hay, Mr Chapman KC submitted, will properly be able to establish that the settlement with Ms Sage was entered into upon legal advice (by way of witness statement in the usual way), and that at that point the onus will shift to Marsh to displace the inference that the terms were prima facie reasonable.

Under English law a liability policy is, generally speaking and in the absence of wording to the contrary, a policy which indemnifies the insured in respect of actual liability. That means that, in order to recover from his insurer the insured must show that he was liable to the person who claimed against him. …

In the event of dispute the existence of liability has to be established to the satisfaction of the insurer, or, failing that, by the judge or arbitrator who has jurisdiction to decide such a dispute. It is not, therefore, necessarily sufficient for the insured to show that he has been held liable to a claimant by some court or tribunal or that he has agreed to settle with him. In practice the fact that this has occurred may cause or persuade the insurer to pay, but, if it does not, the insured must prove that he was actually liable. Under English law the ultimate arbiter of whether someone is liable, if insured and insurer cannot agree, is the tribunal which has to resolve their disputes (or any relevant appeal body). It may hold that there was in fact no actual liability and that an insured who thought, or another tribunal which decided, that there was, liability was in error either on the facts or the law or both.

This principle is potentially very inconvenient for insureds. It may mean that they face weak or dubious claims, which it would be commercially expedient to settle, but in respect of which, if they settle, they may not recover against the insurer because the claims cannot be shown to be well founded. In such a situation they may have to soldier on with the defence and hope to persuade the insurer that it is in his best interests to allow them to settle before trial and to indemnify them when they do, on the basis that, if they lose, the insurer is more likely to have to pay, and to pay more than he would if there was no settlement. Even if they are held liable, this may not in practice, and does not in law, mean that they are automatically covered. The insurer may still say that they were not liable.

…

That liability policies require the establishment of actual liability is apparent from considerations of language and English authority. As to the former, ‘liability’ prima facie means the state of being liable and not alleged liability. …

As to the latter, the principle is summarised in MacGillivray on Insurance Law (12th edition) at 29–006 as being that ‘liability insurance provides an indemnity against actual established liability as opposed to mere allegations’. The position which I have set out in paragraphs 17 and 18 above is vouched or supported in several cases which the judge considered including: West Wake Price & Co v Ching [1957] 1 WLR 45 , 48–51; Commercial Union Assurance v NRG Victory Reinsurance [1998] 2 Lloyd's Rep 600; MDIS v Swinbank [1999] 1 Lloyd's Rep IR 516 , 524; Structural Polymer Systems Ltd v Brown [2000] Lloyd's Rep IR 64 , 67; Thornton Springer v NEM Insurance Co Ltd [2000] Lloyd's Rep IR 590 [34]; Enterprise Oil, which contained a pro-insured policy interpretation clause (‘In the event of any conflict of interpretation between the various clauses and conditions the broadest and least restrictive wording to the benefit of the insured shall always prevail’); and Omega Proteins v Aspen Insurance UK Ltd [2010] EWHC 2280 (Comm).”

“The basic rule under English law is that where a policyholder settles its liability to a third party claimant, and wishes to claim under its liability policy, it is not sufficient for the policyholder simply to establish the reasonableness of the settled amount. In order to succeed, the policyholder must prove (i) that it was in fact legally liable …, and (ii) that the amount for which it would have been liable had the matter been litigated is at least as much as the amount paid under the settlement …”.

what Dalamd must show in relation to Aviva’s and XL’s avoidance, and (2) what proof is required in respect of what would have happened had Butterworth Spengler not acted in breach of duty.”

“The first issue relates to whether Dalamd must show that the claim on the Aviva or XL policy would have failed as a result of Butterworth Spengler’s negligence. Dalamd submitted that it was ‘emphatically not necessary’ for it to prove that. Instead, it contended that it was enough to show simply that Butterworth Spengler’s negligence had ‘impaired’ the insured's claims under those policies. What was meant by this was explained as follows: ‘In other words, it is enough to show that [Butterworth Spengler’s] negligence has provided the insurer with a reasonably arguable ground to defend liability.’ It was said that this was so, because it was part of the broker’s duty to protect his client from unnecessary risks, including the unnecessary risk of litigation. … .”

“The second issue relates to how the court should approach a contention that, but for the defendant’s breach, the policy would in any event not have responded by reason of some other point for which the brokers were not responsible, such as, in the case of the Aviva policy here, a breach of the External Storage Condition. Dalamd submits that in this regard, issues as to what the claimant would have done are to be determined on a balance of probabilities basis, but other issues are to be determined on a loss of a chance basis. That includes not only whether insurers would have taken the other point, and whether they would have compromised the case if that point had stood alone, but also what the court would have decided if the point had been maintained to trial by insurers. … .”

“To these contentions, Butterworth Spengler answered that, in relation to both aspects, they were not in accordance with authority, and were unprincipled. They submitted that in a case, which they submitted this one was, where what the insured was claiming was that breach of the brokers’ duty had caused it the loss of an indemnity under the policy, the issues were to be approached as follows. First, questions as to whether the policy was valid or was voidable by reason of the brokers’ negligence were to be determined either as a matter of law (if applicable) or, insofar as factual issues were concerned, on the basis of the balance of probabilities on the material before the court in the action against the brokers. Secondly, questions of whether there was some other ground, for which the brokers were not responsible, on which the insurance would not have provided an indemnity, for example breach of warranty or condition or the application of an exclusion, were also to be determined as a matter of law (if applicable) or on the basis of the balance of probabilities. However, issues of what the insurer might have done as a matter of business – for example that it might not have taken the other point, or might have compromised it – were to be determined on the basis of loss of a chance. What the court would not do, however, was to determine on a loss of a chance basis what another court would have decided had the other point been pursued to trial against insurers. Butterworth Spengler submitted that their contention that these were the correct approaches was supported by Fraser v Furman [1967] 1 WLR 898 and by Dunbar v A&B Painters [1986] 2 Lloyd's Rep 38, and also by other brokers' negligence cases, including Everett v Hogg, Robinson [1973] 2 Lloyd's Rep 217, Gunns v Par Insurance Brokers [1997] 1 Lloyd's Rep 173 and RR Securities v Towergate Underwriting Group Ltd [2016] EWHC 2653 (QB).”

“In my judgment, in relation to the first aspect, the position contended for by Dalamd is not established by authority, and would produce potentially anomalous results. The effect of Dalamd’s case is that, if the insurer puts forward an arguable defence based on the brokers’ breach of duty, then the insured would be able not to pursue the insurer, to claim against the broker and, if there were no other point by reason of which it was said that the policy would not have paid, to recover in full. While, according to Dalamd, it would be open to the broker to say that the insured had been unreasonable in failing to sue the insurer, this, Dalamd said, would be an argument that the insured had failed to mitigate the loss caused by the broker’s conduct in having ‘impaired’ its claim, and it would be difficult for the broker to succeed on such an argument because the duty to mitigate is a low one, and furthermore the burden of establishing a breach of that duty would fall on the brokers.”

“What this would mean is that it would be open for an insured, faced, for example, with an assertion by an insurer in correspondence of what is merely an arguable defence based on the broker’s negligence, not to pursue the insurer but to proceed against the broker and recover the entirety of the indemnity, notwithstanding that a court actually determining the matter would have found that the insurers' defence was clearly a bad one. On Dalamd’s case the broker could only defend itself by showing that the insured had unreasonably failed to mitigate its loss.”

“In my judgment the approach contended for by Dalamd makes unduly favourable to the insured an action against the broker for a breach consisting simply of creating an uncertainty as to cover by comparison with the insured's action against the insurer who, in the type of case I have mentioned in the previous paragraph, is in breach of its obligation to indemnify.”

“It is the case that, if a breach of duty by the broker has caused the insured’s position to be uncertain, and as a result of that uncertainty, the insured has made a reasonable settlement with the insurer, then the insured can sue the broker for the difference between the amount of the settlement and an indemnity under the policy, without having to establish in that action that the defence for which the broker was responsible was a good one. That is what was decided in FNCB v Barnet Devanney and in Ground Gilbey Ltd v Jardine Lloyd Thompson UK Ltd [2011] PNLR 15.”

“On the other hand, I was shown no authority in which, where there had been no settlement with insurers, and where the insured had sued the brokers on the basis that they were in breach of duty in failing to make proper disclosure or to ensure that proper disclosure was made to insurers, the court awarded damages without a finding or concession that the policy was actually voidable. Instead, in Everett v Hogg Robinson (at 222) and in Dunbar v A&B Printers (at 40: ‘… the fact that they were entitled to repudiate …’) the court proceeded on the basis of whether there had actually been an entitlement on the part of insurers to avoid, not whether there was an arguable case that there had been an entitlement to avoid.”

“Given that the issue of whether or not the policy was voidable depends on the facts in existence at the time of the placement or renewal of the insurance this approach is consistent with the ordinary approach of the courts to determining matters of past fact. Furthermore, there appears to me to be a good reason why, at least in the ordinary case, that should be the basis on which the issue is assessed. That reason is that, in cases of alleged brokers’ negligence, it is a commonplace that both the insurer and the broker are sued, very often in the same action. Manifestly, as against the insurer, the issue of whether the policy is or is not voidable has to be determined on a yes/no basis (i.e., as a matter of law, if and insofar as applicable, or insofar as issues of fact arise, on a balance of probabilities). In my judgment the issue should be determined against the broker on the same basis, and there should not be the possibility of a different basis depending on whether the insurer is and continues to be a party to the proceedings (and does not, for example, settle them).”

“As to the second aspect, I accept Mr Shapiro’s submission that the approach he advocated was that adopted in Fraser v Furman. There the Court of Appeal first decided whether there would have been a good defence open to the (putative) insurers based on breach of a reasonable precautions clause, and held that there would not have been as a matter of construction and given the facts of the case (905B–907H), and then, as an alternative ground, considered whether the putative insurers, Eagle Star, would have taken the point. Similarly, in Everett v Hogg, Robinson the way in which the court dealt with the question of whether there was a different point than that for which the brokers were responsible which would have precluded an indemnity was first to consider whether that different point was a good one (223 RHC), and then to consider whether it would have been taken by (re)insurers and, if it would have been, whether it would have been settled. It also appears to be the approach adopted in Dunbar v A&B, in that at 40 LHC May LJ accepted that the facts of the accident meant that insurers could have relied on Memo 2 to escape liability, but then found that the judge had correctly come to the conclusion that insurers would not have taken the point. Equally, in Gunns v Par Insurance Brokers the approach of Sir Michael Ogden QC (sitting as a Deputy High Court Judge) was to decide whether or not the facts were such that the policy would not, in any event, have responded to the loss by reason of breach by the insured of a reasonable precautions clause (at 177-178). … .”

“By reason of considerations similar to those I have referred to in relation to the first aspect, there are good reasons why this should be the approach of the courts. The issue of whether there was a defence by reason of some other non-disclosure for which the brokers were not responsible, or of breach of a condition or the application of an exclusion for which they were not responsible, depends on facts which existed at the time, depending on the nature of the point, either of placement or of the occurrence of the putatively insured loss. Furthermore, in an action against the insurers, the application of such defences would necessarily be decided on the basis of a determination as to whether the defence was or was not a good one (namely, as a matter of law if and insofar as applicable and on the balance of probabilities as regards any issues of fact). The basis on which they are decided in a claim against the brokers should not depend on whether the insurers are (or are still at the point of trial) parties to that action.”

“Mr Hext QC relied heavily in this context on the decision in O & R Jewellers v Terry. In that case the judge, Sir Godfray Le Quesne QC, sitting as a Deputy High Court Judge, included a number of uncertainties in his overall assessment of the insured's chance of recovering the full claim of £850,000. These included not just the issues of whether insurers would have taken other points, and whether they would have compromised them and if so for how much, which, as I have said, are matters which have been assessed on such a basis in other cases, but also ‘If fought to a finish, with what result?’ Mr Hext QC relied on this to contend that the court should not decide whether the other point which insurers might have taken was correct, but simply to decide the chances of how it might have been decided had insurers been pursued to trial and the point then argued. I do not consider that that case is persuasive authority that that is the right approach, and the judge in that case does not appear to have been addressed with arguments on the point similar to those which I have heard. In any event, I consider that the approach which I have attempted to summarise above, is that which is consistent with the preponderance of authority.”

“I did not find persuasive Mr Hext QC’s submission that that approach was unsatisfactory because it would mean the decision of issues on a yes/no basis in the absence of underwriters. The evidence of underwriters would be admissible and could be adduced by either party, if relevant. In any event it is a commonplace for issues to be resolved on the balance of probabilities even though parties involved in the underlying events are not party to the proceedings.”

“As I think Mr Shapiro accepted, there may perhaps be cases in which, by reason of particular facts, the insured may be able to contend against the insurance broker that it has, in effect, deprived the insured of the opportunity of having its claim under the insurance determined by a court, and that in such circumstances different considerations may apply in relation to one or the other of the two issues I have discussed above. I agree with him, however, that the present case is not pleaded on that basis, and Dalamd adduced no evidence to support such a case.”

“All these cases were analysed on the basis of lost chance principles. In Dalamd Ltd v. Butterworth Spengler Commercial Ltd however, Butcher J held that where the broker alleges that the claim would have failed anyway (because of an alternative defence by the insurer), the causation analysis has two stages. The first is to decide whether the alternative defence was a good one or not, which is done in the ordinary way by legal analysis and balance of probabilities (the burden being on the broker to prove that the alternative defence was a good one). The second is to decide (if the alternative defence is held to be good) whether the insurer would have taken the point, and if so with what result: this is a lost chance analysis. The lost chance analysis does not extend to deciding the percentage chances of the insured succeeding at trial if the insurer would have taken it that far this was to be answered ‘yes/no’ at the first stage.”

“It is suggested that this approach should be treated with some care. There may be cases in which it is appropriate for the judge at the broking trial to reach a final determination of the merits of the insurer’s alternative defence: if it is a simple point of policy construction on agreed facts, for example, or an obvious point about the extent of cover. If the alternative defence is more complex, however, then conducting a ‘trial within a trial’ of its merits carries clear risks. The material available to the broking trial judge may be much more limited than that which would have been before the notional judge at the insurance trial, in which the insurer would have had to give disclosure. Important witnesses may be absent and time may have affected the recollection of those who do attend, if negotiations with the insurer were protracted.”

“For these reasons in the analogous context of claims for negligence against solicitors in conducting litigation, where the solicitor says that the claim would have failed anyway the weight of the authorities is generally against the trial judge trying the merits of the original action. So Dalamd should not, it is suggested, be treated as laying down an absolute rule that the merits of an alternative defence available to the insurer always have to be determined on a balance of probabilities, as opposed to assessing the likely value of the claim against the insurer (given the possibility of that defence) on a lost chance basis.”

“This is consistent, it is suggested, with the decision of the Supreme Court in Perry v. Raleys Solicitors, which draws a ‘bright line’ distinction in negligence causation between (a) facts which the claimant has to prove about his own actions and (b) facts which the claimant has to prove about the actions of others or about future events. Only the claimant’s own actions have to be proved on the balance of probabilities; those in category (b) are assessed on lost chance principles. In the context of claims against solicitors for the negligent conduct of underlying litigation, the only action of his own which the claimant has to prove is that he would have brought the claim: if that is proved, generally speaking causation is established and what would have happened thereafter (which measures the strength of the claim) goes into the lost chance analysis. Translating that approach in to claims against broker, the only action of his own which the insured needs to prove is that he would have mounted a claim against the insurer even in the light of the alternative defence. What would have happened thereafter is dependent on the actions of others. If the alternative defence can be shown to be in the ‘clearly right/clearly wrong’ class then the case would lend itself to Butcher J’s approach, but in other cases it would be better for causation to be assessed on a broad view of the merits of the insurer’s putative position.”

“Alternatively, as a result of the Defendant’s aforesaid breaches of duty the Claimant has lost the opportunity of obtaining insurance cover that would have responded to the liabilities arising out of the Accident and/or of avoiding or reducing the liabilities and other losses it has in fact suffered as a result of the Accident, including, in particular, avoiding the need to provide the SPA Indemnity and to fund the Sage Settlement Sum.”

“In the premises, but for the Defendant’s aforesaid breaches of duty, one way or another insurance cover would have been in place that would have provided an indemnity (in full or, in the alternative, at least in substantial part) in respect of the liabilities arising from the Accident and the Claimant itself would not have been subject to an uninsured claim and liability in respect of the same.”

This paragraph underlines the fact that Norman Hay’s case is that “one way or another” it would not have had to fund the settlement with Ms Sage; that, indeed, is what paragraph 39 goes on to state in terms.

Effective cover was available and in place under the IMP Policy. Had the existence of this effective cover for IMP been identified by Marsh, then, Norman Hay would have given instructions for that cover to be retained (even if only for the benefit of IMP) rather than have it cancelled and would have instructed Marsh to seek such cover for Norman Hay and other companies in its group under the global insurance programme.

Alternatively, if liability cover had not been available as part of the global programme of insurance arranged by Marsh, had Norman Hay been made aware by Marsh of the absence of such cover then it would have issued a standing instruction (similar to the instruction it in fact issued after the accident) that its employees should ensure that when hiring cars on business trips the maximum available insurance cover was obtained and that private hire vehicles were to be used wherever possible. In those circumstances, Mr Kelsall would have been obliged to take out adequate insurance when hiring the car in which he was killed and/or used a private hire car instead.