RIDGECREST NZ LIMITED v IAG NEW ZEALAND LIMITED [2014] NZSC 129

RIDGECREST NZ LIMITED v IAG NEW ZEALAND LIMITED [2014] NZSC 129 [27 August 2014]JUDGMENT REISSUED TO CORRECT AN ERRORONEOUS FIGURE IN[12] AND A TYPOGRAPHICAL ERROR IN [33]IN THE SUPREME COURT OF NEW ZEALANDSC 76/2013[2014] NZSC 129BETWEEN RIDGECREST NZ LIMITEDAppellantAND IAG NEW ZEALAND LIMITEDRespondentHearing: 10 March 2014Court: McGrath, William Young, Glazebrook, Blanchard andTipping JJCounsel: C R Carruthers QC and P A Cowey for AppellantB D Gray QC and P M Smith for RespondentJudgment: 27 August 2014Reissued: 19 September 2014Effective dateof Judgment: 27 August 2014JUDGMENT OF THE COURTA The appeal is allowed.B The preliminary question is answered "yes" but subject tothe caveats identified in [62].C The appellant is awarded costs of $25,000 together with reasonable disbursements to be fixed by the Registrar in relation to the appeal.D The orders for costs in the High Court and Court of Appeal are set aside and the respondent is to pay the appellant costs in those courts to be fixed by those courts.____________________________________________________________________REASONS(Given by William Young J)Table of ContentsPara NoThe appeal .................................................................................................. [1]The policy ................................................................................................... [6]The facts in more detail........................................................................... [11]Overview of the problem ........................................................................ [13]Overview of the issues ............................................................................. [16]Does the policy, when construed in the context of the events which happened, require IAG to make payments in relation to the earlier earthquakes? ......................................................................... [20]The approach of the High Court ........................................................ [20]The approach of the Court of Appeal ................................................. [22]Discussion .......................................................................................... [24]Are the losses resulting from the earlier earthquakes to be treated as merged or subsumed in the loss caused by the final earthquake?..................................................................................... [30]Overview............................................................................................. [30]The approach of the High Court ........................................................ [31]The approach of the Court of Appeal ................................................. [32]The marine insurance cases ............................................................... [33]Authorities as to merger in non-marine insurance contexts .............. [41]Discussion .......................................................................................... [48]Is Ridgecrest's claim precluded by the indemnity principle? ............. [53]Conclusion ................................................................................................ [62]The appeal[1] Ridgecrest NZ Ltd (Ridgecrest) was the owner of a building at 215 Gloucester Street, Christchurch which was insured with IAG New Zealand Ltd (IAG). The maximum liability of IAG under the insurance policy in respect of anysingle "happening" was $1,984,000. This was less than the replacement value of thebuilding. During the currency of the policy, the building was affected by four earthquakes which occurred on 4 September and 26 December 2010 and on 22 February and 13 June 2011. IAG commissioned builders to repair the damage to the building caused by the 4 September and 26 December 2010 events but all work stopped after the 22 February 2011 earthquake.[2] It is common ground that by 13 June 2011 at the latest, the building was damaged beyond repair. IAG maintains that the building was damaged beyondrepair after the 22 February 2011 earthquake, whereas Ridgecrest contends that this did not occur until 13 June 2011. It is not possible for us to resolve this aspect of the dispute on this appeal. For ease of discussion, we will refer to the earthquake which damaged the building beyond repair (being either the 22 February or the 13 June 2011 events) as "the final earthquake" and the preceding earthquakes (beingthe 4 September and 26 December 2010 events and possibly the 22 February 2011event) as "the earlier earthquakes".[3] The competing positions of the parties are as follows:(a) Ridgecrest claims that it is entitled to $1,984,000 in respect of the final earthquake, along with the losses caused by the earlier earthquakes.(b) IAG says that its liability under the policy is limited to the cost of repairs actually undertaken to remediate the damage caused by the earlier earthquakes and, in respect of the final earthquake, the maximum amount payable under the policy for any one happening, being $1,984,000.[4] Ridgecrest commenced proceedings against IAG in the High Court. In the course of these proceedings, the parties sought a ruling from the High Court on a preliminary question which was expressed in this way:Where –1. There have been four happenings within a period of insurance;2. Each happening caused damage to the Plaintiff's building;3. Subsequent to the first two happenings repairs were commenced but not completed by the time of the next happening;4. Following the third or fourth happening, the building was damaged so that the cost of repair exceeded the sum insured;5. The building has been damaged beyond repair as a result of either the third or fourth happening;Then:6. Is the Plaintiff entitled to be paid for the damage resulting from each happening up to the limit of the sum insured in each case?An agreed statement of facts was submitted to the Court along with a bundle of documents.[5] In a judgment delivered on 8 November 2012, Dobson J answered the preliminary question in favour of IAG.1 On appeal, the Court of Appeal reached the same conclusion as Dobson J but on different grounds.2The policy[6] The operative clause of the policy provides:A. YOU ARE INSURED FORSudden and accidental loss of or damage to your Business Assets.The policy encompasses loss or damage caused by earthquakes.[7] The policy is also expressed to apply on a happening by happening basis, providing:This Policy covers only those Parts for which a Limit is shown in the Schedule and the maximum amount you can claim under any Part in respect of any one happening (inclusive of fees and costs) is the current Limit for that Part.As noted, the limit specified is $1,984,000.[8] Clauses C1 and C2 of the policy provide as follows:C. THE AMOUNTS YOU CAN CLAIM1. This insurance will pay the amount of loss or damage or the estimated cost of restoring your Business Assets as nearly as possible to the same condition they were in immediately before the loss or damage happened using current materials and methods.1 Ridgecrest New Zealand Ltd v IAG New Zealand Ltd [2012] NZHC 2954, [2013] Lloyd's Rep IR67 [Ridgecrest (HC)].2 Ridgecrest New Zealand Ltd v IAG New Zealand Ltd [2013] NZCA 291, [2013] 3 NZLR 618(O'Regan P, Arnold and Harrison JJ) [Ridgecrest (CA)].2. Where Replacement cover has been agreed by us and specified in the Schedule and following loss or damage you restore or replace the lost or damaged Business Assets this insurance will pay(a) for Buildings(i) where repairable, the cost of restoration of damage to the same condition when new,or(ii) if unable to be repaired because of such damage, the cost of replacement by an equivalent building which meets your requirements at any site provided we shall not pay more than the cost of replacement at the Site stated in the Schedule.Such restoration will use current materials and methods and include the cost of changes to meet the lawful requirements of any local authority or statute but not for work you have already been instructed to do prior to the loss or damage.Under the heading "Replacement Cover" in the schedule the word "yes" appears,with the result that cover under cl C2 was provided.[9] Clause C1 thus imposes on IAG a default obligation to pay either the diminution in value associated with the damage to the building or the estimated cost of restoring it to its pre-earthquake condition using current materials and methods. In contradistinction, the obligation imposed by cl C2 is entirely conditional upon Ridgecrest having restored or replaced the damaged building, but, if that condition came to be satisfied, IAG was required to pay the restoration or replacement cost, up to the liability limit.[10] Although the claim under cl C1 has been pleaded on the basis of estimated costs of repair, we propose to discuss it by reference to diminution in value as this makes it a little easier to understand the practical effect of the happening by happening operation of the limit and, in particular, that what was insured between the earlier and final earthquakes was the building as damaged (and thus as diminished in value) by the earlier earthquakes.The facts in more detail[11] The statement of agreed facts submitted to the High Court was in these terms:The following facts are agreed for the sole purpose of determining the preliminary question to be argued on 26 September 2012:1. At material times, there was a policy of insurance between the parties whereby the Defendant provided replacement cover for thePlaintiff's commercial building situated at 215 Gloucester Street forevents including earthquake damage, with a limit of $1.984m for each happening.2. As a result of an earthquake on 4 September 2010, the Plaintiff'sbuilding suffered damage.3. The cost to repair the building following the 4 September 2010 earthquake was assessed and estimated to be between $110,155.09 (inclusive of GST) and $196,352 (plus GST). The repairs had been commenced, but had not been completed by the time the earthquake described in paragraph 4 below occurred.4. There was a second earthquake on 26 December 2010 that caused further, distinct damage to the building.5. The cost to repair the damage to the building following the 26 December 2010 earthquake was assessed and estimated to be between $200,000 (inclusive of GST) and $337,056 (plus GST). The repairs had been commenced, but had not been completed by the time the earthquake described in paragraph 7 below occurred.6. The Defendant made payments for repairs to the building as a result of the first two earthquakes. The Defendant had paid the sum of not less than $125,609.83 for repairs, consultants and temporary works, by 22 February 2011. However the repair work had not been completed for the damage caused by either of the first two earthquakes prior to the earthquake described in paragraph 7 below.7. There was a third earthquake on 22 February 2011 that caused further, distinct damage to the building.8. The Plaintiff's position is that the cost to repair the damage to thebuilding caused by the 22 February 2011 earthquake was assessedand estimated to be $1.924m plus GST. The Defendant's position isthat as a result of the 22 February 2011 earthquake, the building was damaged beyond repair, or the cost of repairs exceeded the sum insured.9. There was a fourth earthquake on 13 June 2011 that caused damageto the building. The plaintiff says there was 'further and distinct damage' but the defendant says the June earthquake only exacerbated the existing damage.10. The Plaintiff's position is that the cost to repair the damage to thebuilding caused by the 13 June 2011 earthquake has been assessed and will exceed $1.984m plus GST; and further that as a result of that earthquake, the building's structure and foundation were so damaged the building was rendered irreparable. The Defendant'sposition is that it had already paid the agreed sum insured for thebuilding and so had performed its obligations under the policy. In addition, it says the building was already damaged beyond repair prior to the 13 June 2011 earthquake.11. The Plaintiff's position is that the residual value of the building aftereach earthquake was greater than the cumulative cost of the incomplete repairs. The building therefore had residual value that was insured, until the damage suffered on 13 June 2011.12. The Defendant has paid for repairs and other remedial work in the sum set out in the paragraph 6 above and in addition, has on 12 July 2012 paid the Plaintiff under the policy $1,984,000 plus GST (less the excess and less certain costs).[12] The key documents referable to the dispute were also made available to the High Court. These show that in the aftermath of the 4 September 2010 earthquake IAG commissioned Hawkins Construction Ltd to project manage the necessaryrepairs. On 6 December 2010, Ridgecrest wrote to IAG's loss adjustors in theseterms:Further to our telephone conversation on Friday we wish to confirm that as we have become aware that there is an option whereby we are not required to accept Hawkins as project managers for the repairs to our building at 215 Gloucester Street, we have decided not to engage them in this capacity.Therefore we would like to finalize this claim with a cash settlement of $197500.00 +Gst which our excess would have to be deducted from and we understand this would be full and final settlement of this particular claim.If there was a response in writing to this letter, it was not included in the bundle of documents. It is, however, apparent that IAG was not interested in a cash settlement. IAG also insisted on retaining control of the repairs. The result was that Hawkins Construction continued to manage the remediation exercise until it was brought to a halt by the 22 February 2011 earthquake.Overview of the problem[13] The successive losses problem presented by the case is a function of (a) the policy providing for both indemnity and replacement cover, (b) a liability cap which did not purport to be, and was known not to be, the replacement value of the building, and (c) the resetting of that liability cap after each happening.[14] At the heart of the problem are two overlapping points:(a) On a literal reading of the policy, Ridgecrest might be able to recover twice for the same damage. For instance, say a wall which was only superficially damaged by the earlier earthquakes was destroyed by the final earthquake. Allowing claims under cl C1 for the earlier earthquakes and under cl C2 for the final earthquake would result in double counting in respect of the damage caused by the earlier earthquakes. Such a result would rationally be seen by insurers as unintended and it is one which the courts should endeavour to avoid.(b) The effect of the liability cap resetting after each happening is that Ridgecrest, on its argument, is entitled to recover, for successive losses, more than the liability cap. While such a result would be galling to an insurer (that had received a premium based on a liability cap which was less than replacement value), the possibility of over- cap recovery for successive events is contemplated by the way the cap operates on a happening by happening basis.In the rest of these reasons, we will use "double counting" and "over-cap recovery"to refer to the points just addressed.[15] If the policy just provided for indemnity cover, separate claims for successive losses would not cause a double counting problem albeit that it could result in over- cap recovery.3 Assuming both indemnity and replacement cover, but no liability cap, there would necessarily be no scope for over-cap recovery and there would be nopotential for double counting. This is because IAG's discharge of its replacementvalue liability would mop up its already accrued indemnity liability in relation to unrepaired damage caused by the earlier earthquakes. There would thus be no possibility of Ridgecrest being paid twice in respect of the same damage. But, given the liability cap which is lower than the replacement value, the discharge by IAG of3 Assume the building's indemnity value was $3.5m with damage from the earlier earthquakesresulting in a diminution in value of $1.75m and the building being destroyed by the final earthquake. It is difficult to see why Ridgecrest, in that situation, would not be able to recover $1.75m for the earlier earthquakes and the same amount in respect of the final earthquake (being the residual value of the building). In this situation there would be an over-cap recovery but no double counting. This example is further discussed below at [51].its replacement value liability (constrained as it is by the cap) does not, in money terms, necessarily mop up the losses resulting from the earlier earthquakes.Overview of the issues[16] Ridgecrest contends that, properly construed, the effect of the policy is that:(a) after each of the earlier earthquakes, IAG became liable under cl C1to pay "the amount of loss or damage or the estimated cost ofrestoring [the building] as nearly as possible to the same condition [it was] in immediately before the [earthquakes]"; and(b) after the final earthquake, IAG became liable under cl C2 to pay "the cost of replacement by an equivalent building", subject to the limit of$1,984,000.[17] IAG denies that the policy should be construed in this way. It also argues that the claims associated with the earlier earthquakes became merged in the claim associated with the final earthquake. In the High Court and Court of Appeal, it also asserted that in the events that happened, the policy was frustrated.[18] In the High Court, Dobson J resolved the construction and merger arguments in favour of Ridgecrest but found in favour of IAG on its frustration argument. He did so on the basis of a conclusion that a term could be implied into the policy that:4 the scope of liability for subsequent happenings would not extend torequire payment of sums greater than was necessary to effect repairs that were able to be undertaken before the building became irreparable.In effect, he held that the insurance policy covering earthquake-induced losses was frustrated by an earthquake, a proposition which, expressed that way – and of course that is not the way the Judge did express his conclusion – does not seem very plausible. The Court of Appeal was of the view that IAG could not succeed on this argument5 and it was not pursued in argument before us.4 Ridgecrest (HC), above n 1, at [75].5 Ridgecrest (CA), above n 2, at [53].[19] We propose to deal with what remains in issue in the case by reference to three questions:(a) whether the policy, when construed in the context of the events that happened, requires IAG to make payments in relation to the earlier earthquakes;(b) whether the losses resulting from the earlier earthquakes are to be treated as merged or subsumed in the losses caused by the final earthquake; and(c) whether Ridgecrest's claim is precluded by the indemnity principle.6These questions are not truly independent of each other. For instance, if the answerto the first question is a firm "yes", that might be thought to be decisive of thesecond and third questions. The three questions do, however, provide useful headings for discussion. The first two are broadly consistent with approaches taken in the High Court and Court of Appeal and the arguments of counsel, and the third question has been triggered by some commentary in relation to the High Court and Court of Appeal judgments in the present case.Does the policy, when construed in the context of the events that happened, require IAG to make payments in relation to the earlier earthquakes?The approach of the High Court[20] Dobson J observed that there was no provision in the policy entitling IAG toassume control of the repair process and that IAG's obligations under cls C1 and C2were simply to pay money.7 He rejected the argument advanced by IAG that cls C1 and C2 were mutually exclusive alternatives, depending on whether the building was insured on a replacement basis.8 He recorded that the words "you restore or replace"6 The principle under which insurance policies are construed to avoid an insured recovering more than the loss caused by the insured event.7 Ridgecrest (HC), above n 1, at [10]–[11].8 At [27].meant that liability under cl C2 was conditional on such repairs being effected, whereas the scope of cl C1 was unqualified.9[21] Dobson J's conclusion was that:10 the limit of liability is available for the insured to claim in the event of aloss, on as many occasions during the period of insurance as a relevant loss is suffered.The approach of the Court of Appeal[22] The Court of Appeal considered that in the aftermath of the earlierearthquakes, Ridgecrest's claims were advanced under cl C2.11 The Court analysedRidgecrest's statement of claim, which, as it pointed out, either invoked cl C2 or was expressed in terms predicated on the assumption that cl C2 was the dominant provision.12 The Court stated that it was not until the hearing in the High Court that counsel for Ridgecrest invoked cl C1.13 The Court then reviewed the facts, noting that, after the first two earthquakes, IAG commissioned repair work which was underway by 22 February 2011.14 It saw this as consistent with the parties acting on the basis that cl C2 was the operative provision.15 In the course of doing so, the Court referred to the letter of 6 December 2010 but commented that it was sent well before the statement of claim was settled.16 The Court was unimpressed by the argument that payment of a claim under cl C1 could be provisional and followed by a claim under cl C2 if repairs as envisaged by that clause were later effected.17[23] The Court then construed the policy on the basis that the claims advanced were made under cl C2.18 On this basis, it held:[49] In relation to claims relating to a damaged but repairable building, IAG is liable under cl C2(a)(i) in respect of each happening for the cost of restoration of damage to the building to the same condition as when it was9 At [27].10 At [34].11 Ridgecrest (CA), above n 2, at [45].12 At [32]–[36].13 At [36].14 At [37]–[38].15 At [38].16 At [38]–[39].17 At [42]–[43].18 At [47].new. The quantification of the liability depends on the repairs actually being made and the liability equates to the actual cost of those repairs. In the present case the cost of the repairs actually completed after the first earthquake is payable in respect of the claim resulting from the first earthquake. In addition, the cost of the repairs actually completed after the second earthquake is payable in respect of the claim resulting from the second earthquake. This is so even though in neither case was the building fully repaired before the next earthquake struck.[50] In relation to the third earthquake, the position taken by IAG is that the building then became unrepairable, in which case cl C2(a)(ii) requires the cost of replacement of the building to be paid, subject to the maximum liability under the policy. We understand that the replacement cost exceeds that maximum, so the maximum amount becomes payable, less anydeductions. IAG's position in respect of the claim resulting from the fourthearthquake is that there is no liability because the building was unrepairable before the fourth earthquake.[51] On the view taken by Ridgecrest that the building remained repairable after the third earthquake, the liability in respect of the claim resulting from the third earthquake would be zero because no repairs have been undertaken and therefore no costs of repairs have been incurred by Ridgecrest for which a claim can be made. However, on Ridgecrest's viewof the facts, the building became unrepairable after the fourth earthquake in which case the amount payable by IAG in relation to the claim resulting from the fourth earthquake is that required by cl C2(a)(ii) which, as indicated above is, in this case, the amount equal to the maximum liability under the policy.Discussion[24] As noted, the liability cap is expressed to apply on a happening by happening basis. To construe the policy against Ridgecrest would require a liability cap in these terms:The maximum amount you can claim in respect of any one happening(inclusive of fees and costs) is the current Limit and, in the case of a claim under cl C2, is the current limit less the amount of any unrepaired damage resulting from earlier happenings in the same insurance period.The words we have italicised would have to be read into the clause. To do thiswould detract in a significant way from the "in respect of any one happening"constraint on the operation of the cap.[25] Contrary to the view expressed by the Court of Appeal, we see nothing implausible in an approach to the policy under which the insured could receive payment under cl C1 subject to a top-up if, following repair, cl C2 came to beengaged. That said, however, we accept that the conduct of Ridgecrest and IAG between 4 September 2010 and 22 February 2011 appears to have been predicated on the assumption that the primarily operative provision of the policy was cl C2. IAG commissioned repairs on what seems to have been a "new for old" basis and, givencl C2, Ridgecrest was entitled to insist on nothing less. We also accept thatRidgecrest's statement of claim was prepared on the same basis. However, and incontrast to the approach taken by the Court of Appeal, we do not see this as being of controlling significance.[26] Prior to 22 February 2011, the shared assumption was that the building could be repaired within the liability cap of $1,984,000. To be more particular, the expectation was that the building would be repaired and that the costs of doing so would be met by IAG by reference to cl C2(a)(i). But, as a result of the final earthquake, the building was not able to be repaired. The result is that, from that point onwards, cl C2(a)(i) was no longer applicable.[27] After the final earthquake (being the event which caused the "loss or damage" which rendered the building irreparable), Ridgecrest was entitled undercl C2(a)(ii) to the cost of any equivalent building it acquired as replacement, subject to the limit and other conditions specified in cl C2(a)(ii). The entitlement under cl C2(a)(ii) was expressly conditional on Ridgecrest replacing the building, although it may be that IAG has waived that condition.[28] All of this means that Ridgecrest is entitled to cover in respect of the final earthquake under cl C2(a)(ii). Clause C2(a)(ii), however, is not applicable to the earlier earthquakes as the damage associated with those earthquakes was not the damage which rendered the building unable to be repaired. Clause C2(a)(i) is likewise not applicable to the earlier earthquakes because the damage caused by the earlier earthquakes was not, and never will be, repaired.[29] With cl C2 out of the way, cl C1 applies. It follows that the policy entitles Ridgecrest to payment of the "amount of loss or damage" or the estimated costs ofrepair associated with the damage caused by the earlier earthquakes. Accordingly,but subject to our answers to the second and third questions, Ridgecrest is entitled to be paid:(a) under cl C1 for the damage resulting from each of the earlier earthquakes, up to the limit of $1,984,000 in respect of each happening; and(b) under cl C2(a)(ii) for the loss caused by the final earthquake, up to the limit of $1,984,000.Are the losses resulting from the earlier earthquakes to be treated as merged or subsumed in the loss caused by the final earthquake?Overview[30] The merger principles relied on by IAG have their origin in cases involving marine insurance and are reflected in s 77 of the Marine Insurance Act 1908. This provides:77 Successive losses(1) Unless the policy otherwise provides, and subject to the provisions of this Act, the insurer is liable for successive losses, even though the total amount of such losses may exceed the sum insured.(2) Where under the same policy a partial loss which has not been repaired or otherwise made good is followed by a total loss, the assured can only recover in respect of the total loss.IAG maintains that the marine insurance principles should apply with the result thatRidgecrest's partial loss claims in respect of the earlier earthquakes are merged in thetotal loss claim resulting from the final earthquake.The approach of the High Court[31] Dobson J referred to s 77 and to the leading marine insurance cases on merger, namely British and Foreign Insurance Co, Ltd v Wilson Shipping Co, Ltd1919 British and Foreign Insurance Co, Ltd v Wilson Shipping Co, Ltd [1921] 1 AC 188 (HL).and Livie v Janson,20 along with a New Zealand earthquake case, Wright, Stephenson, and Co, Ltd v Holmes.21 We will refer to these cases in more detailshortly. He recorded Ridgecrest's argument that the cases relied on by IAG turn onthe principle that under a marine insurance policy, no cause of action accrues to the insured until the end of the policy period.22 He then went on to say:23[49] I am not satisfied that the doctrine of merger has been applied to marine insurance policies solely because the underwriters would not be liable for unrepaired damage to a vessel until the expiry of the policy. That seems more likely to be a consequence of the settled approach, rather than the rationale for it. A broader basis for merger is that the policy of mercantile law in regulating interests between underwriters and insured interests operates to moderate the extent of liabilities assumed by underwriters. The essence of such contracts is that the insured will be indemnified for losses actually incurred. This is exemplified in the concernnot to produce a profit for the insured .[50] Marine insurance does not contemplate "replacement" of a vessel, atleast within an existing hull policy. Once either a constructive or actual total loss has been established, that exhausts the insurance over a vessel. A vessel built or acquired to replace one that is lost by a marine peril is a new insurance prospect. In contrast, in theory at least, a replacement for an irreparable building could be insured under the same policy. In that sense,the occurrence of a "total loss" under a marine hull policy brings an end tothe contract, whereas a pay-out for what is a constructive total loss of a building may not. Merger applies in marine insurance policies to subsumelesser liabilities that have become a matter of indifference within an insurer'sultimate liability. In those terms the proposition is supported by common sense in a way that arguably justifies a wider application. It certainly invitesanalogy with arguments for the doctrine of frustration to apply .[51] The terms in which the Court of Appeal commented in Wright, Stephenson on Lord Birkenhead's observations about the policy having"wide application" are not a sufficient foundation for treating the doctrine ofmerger as applying to wider categories of material damage insurance. As Mr Carruthers pointed out, under a marine insurance contract, the insurer is only liable for damage to the vessel when the repairs are effected, or at the end of the term of the policy. In contrast, the present policy creates liabilityas soon as damage caused by a "happening" occurs. It is materially easier tojustify the merger of one or more pending claims than it is to re-cast the character of one or more existing claims.[52] In material damage insurance, parties are left to negotiate the scope of cover by contract, and it is a matter of interpreting the terms contracted for. One influence on that is likely to be the fundamental nature of material damage insurance as an indemnity against loss actually incurred, but that cannot justify imposing a rule that merger should apply automatically, or20 Livie v Janson (1810) 12 East 648, 104 ER 253 (KB).21 Wright, Stephenson, and Co, Ltd v Holmes [1932] NZLR 815 (CA).22 Ridgecrest (HC), above n 1, at [38]–[46].23 (citations omitted).uniformly. Indeed, in the present case, there are circumstances where the insurer will be liable for more than indemnification and what amounts to betterment, if the insured restores a damaged building not to its pre-happening condition, but to "as new" condition and claims that cost under clause C2.[53] Reflecting on these considerations, I consider that the greater variety in the terms of general insurance contracts, and the absence of any equivalent to s 77(2) of the Marine Insurance Act 1906, count against the uniform adoption of merger as applying automatically to all general insurance, or material damage insurance policies. Here, the policy was on terms that the insurer could conceptually be liable for later claims in respect of subsequent happenings, even after a constructive total loss.The approach of the Court of Appeal[32] The Court of Appeal, having decided the case on the basis of the construction of the policy, did not express an opinion on the merger argument.24The marine insurance cases[33] The first of the relevant cases is Livie v Janson.25 The policy in question was in respect of the ship Liberty for a voyage from New York to London. The insured warranted that the ship and cargo were free from American condemnation by reason of an embargo imposed by the United States on trade with the United Kingdom and France during the Napoleonic Wars.26 Shortly after setting sail, the ship came into contact with floating ice and grounded, suffering serious damage as a result. It was then seized by American customs officials and condemned for breach of the embargo. Prior to condemnation, the ship and cargo had undoubtedly been damaged as a result of the perils of the sea and the owner sued on the policy. In delivering thejudgment of the Court of King's Bench, Lord Ellenborough CJ said that the question in the case was:27 whether the total loss by subsequent seizure and condemnation takesaway from the assured the right to recover in respect to the previous partial loss by sea-damage?24 Ridgecrest (CA), above n 2, at [53].25 Livie v Janson, above n 20.26 The warranty just meant that condemnation was not an insured risk as the premium reflected therisks which the ship's crew would be likely to take to avoid detection.27 At 653.The question was answered in the affirmative:28Considering the deterioration of the ship and cargo we think we may layit down as a rule, that where the property deteriorated is afterwards totally lost to the assured, and the previous deterioration becomes ultimately a matter of perfect indifference to his interests, he cannot make it the ground of a claim upon the underwriters.The Court later commented:29There may be cases in which, though a prior damage be followed by a total loss, the assured may nevertheless have rights or claims in respect of that prior loss, which may not be extinguished by the subsequent total loss. Actual disbursements for repairs in fact made, in consequence of injuries by perils of the seas prior to the happening of the total loss, are of this description; unless indeed they are more properly to be considered as covered by that authority, with which the assured is generally invested by thepolicy, of "suing, labouring, and travailing, [etc], for, in, and about the defence, safeguard, and recovery of the property insured:" in which case the amounts of such disbursements might more properly be recovered as money paid for the underwriters under the direction and allowance of this provisionof the policy At the end of the judgment, there is this comment:30 there never existed a period of time, prior to the total loss, in which the assured could have practicably called on the underwriters for an indemnityagainst the temporary and partial injury sustained by the property insured Livie v Janson did not involve precisely the situation envisaged by s 77 of the Marine Insurance Act as the seizure and condemnation of the Liberty (which brought about the total loss) was not covered by the policy. As will become apparent, the other leading cases to which we will refer also concerned factual situations which were only analogous to those envisaged by s 77.[34] The judgment of the House of Lords in British and Foreign Insurance Co, Ltd v Wilson Shipping Co, Ltd31 concerned a vessel which was chartered to the British Admiralty on terms which required the Admiralty to assume responsibility for war risks. The vessel was also insured against marine risks under a time policy. During the currency of the policy, the vessel was damaged by marine28 At 654.29 At 655.30 At 656.31 British and Foreign Insurance Co, above n 19.risks. Temporary repairs were carried out but before permanent repairs – estimated to cost £1,770 – were effected, the vessel was torpedoed by a German submarine andbecame a total loss. The Admiralty paid to the owner the agreed "sound" value ofthe vessel less £1,770, reflecting its damaged condition when lost.[35] The question for the House of Lords was whether the owner was entitled to recover from the insurer the estimated cost of the unexecuted repairs of £1,770, which was the amount by which it was out of pocket. This turned on whether the principle established by Livie v Janson was confined to circumstances where, by reason of the subsequent total loss, the insured was unaffected by the earlier lesser damage. In the High Court, Bailhache J held that the principle was not so confined and he found for the insurer. The Court of Appeal disagreed but, on further appeal, the House of Lords restored the judgment of Bailhache J.[36] In his speech, Lord Birkenhead LC reviewed the cases, starting with Livie v Janson. He saw that decision as being "clearly right".32 Otherwise, the owner of a vessel which met and suffered damage from successive gales before being finally lost would be entitled to more than the value of the vessel and thus make a profit from its loss. As to the principle underlying Livie v Janson, he referred to a number of cases in which Livie v Janson had been applied, including Lidgett v Secretan,33which we discuss later in the reasons. Lord Birkenhead then adopted the following passage from the judgment given by Bailhache J at first instance:34Whether an underwriter is or is not liable for unrepaired damage cannot be ascertained until the expiration of the policy. If before the expiration of the policy there is a total loss he is not liable to pay for the earlier unrepaired damage sustained during the currency of the same policy, and it makes no difference whether the total loss falls upon him or is due to an excepted peril against which the owner is insured or uninsured. The true doctrine is that thesmaller merges in the larger . The question in every case must be, did thetotal loss happen before the underwriter's liability for the unrepaireddamage accrued? If yes, he is not liable; if no he is liable.[37] Viscount Finlay's speech is relevant to the present case primarily in respect ofhis approval of the judgment in Lidgett v Secretan.35 Lord Shaw expressed his entire32 At 194.33 Lidgett v Secretan (1871) LR 6 CP 616 (Comm Pleas).34 British and Foreign Insurance Co, above n 19, at 198 (emphasis added).35 At 202–203.agreement with the judgment of Bailhache J (and thus with the passage set out above).36 He adopted with approval remarks by Maule J in Stewart v Steele37 to the effect that a claim in respect of unrepaired damages vests only at the expiry of the policy.38 However, in a passage which provides some support for IAG, he then went on to say:39If once the principle of indemnity were extended the length of permitting a mere right of action to be the ground or the measure of the indemnity, then the reality of the case is lost sight of, for the right of action is a mere abstraction and represents in truth a right to be recouped for expenses whichcan never be laid out, and the word "indemnity," or the word "recouping,"fails on account of sheer self-inconsistency. One finds oneself not in the region of indemnity against loss, but in the region of profit-earning.The other speeches do not throw any light on the issue which we must determine.[38] In Lidgett v Secretan40 the owners had insured their vessel "at and from London to Calcutta, and for thirty days after arrival". By a separate policy, they insured the ship "at and from Calcutta to London". Both were valued policies. Thevessel was damaged on the outward voyage and on arrival in Calcutta she was taken into a dry-dock for repair. After the expiry of the first policy and within the scope of the second, the vessel, which was still in the dry-dock, was destroyed by fire. In issue was whether the owners could claim for the vessel's depreciation resultingfrom the unrepaired damage suffered on the outward voyage. The Court held thatthe owners were entitled to recover for both the vessel's depreciation in value andunder the second policy for a total loss at the agreed value and thus with no deduction for the depreciation.[39] The final marine insurance case which we should mention is Kastor Navigation Co Ltd v AGF MAT.41 As a result of a fire in the engine room, the Kastor Too became a constructive total loss. Some 15 hours later the vessel sank and thus became an actual total loss. The sinking was mysterious and the trial Judge36 At 205.37 Stewart v Steele (1842) 5 Scott NR 927 (Comm Pleas).38 British and Foreign Insurance Co, above n 19, at 206.39 At 207.40 Lidgett v Secretan, above n 33.41 Kastor Navigation Co Ltd v AGF MAT [2002] EWHC 2601 (Comm), [2003] 1 Lloyd's LR 296(QB) [Kastor (HC)]; Kastor Navigation Co Ltd v AGF MAT [2004] EWCA Civ 277, [2004] 2 Lloyd's LR 119 (CA) [Kastor (CA)].(Tomlinson J) was not satisfied that the sinking was caused by a peril covered by the policy. The case was not within s 77(2) of the Marine Insurance Act 1906 (UK) –corresponding to s 77(2) of the New Zealand Marine Insurance Act – because it involved a constructive total loss (and not a partial loss) followed by a total loss. The question for Tomlinson J was whether the constructive total loss claim was extinguished by the later sinking. After referring to s 77(2), the Judge said:42It may be that the common law rule is wider than the sub-section. Thus inBritish and Foreign Insurance Co Ltd v Wilson Shipping Co Ltd it was held that when a vessel insured against perils of the sea is damaged by one of the risks covered by the policy and before that damage is repaired she is lost, during the currency of the policy, by a risk which is not covered by the policy, in fact in that case by an excluded risk, the insurer is not liable for the unrepaired damage. However it should be remembered that both at common law and under the 1906 Act the time to estimate the loss where the party is put to no expense, as in the case of unrepaired damage, is at the expiration of the risk – see Livie v Janson and Marine Insurance Act, 1906, s 69(3). See also per Lord Birkenhead, LC in Wilson at pp 194–195. Wilson is no authority for the proposition that an accrued cause of action can be defeated by a subsequent loss.[40] The Court of Appeal upheld the judgment of Tomlinson J. In doing so, it referred to Wilson Shipping, noting that the case was "unfortunate" for the owner.43The Court then went on to say:44On the other hand the rule which had been established in marine insurance was that there was no claim for unrepaired partial loss until the end of the policy period. We suppose a different rule might have been established, but the rule makes good sense in a relationship where a ship is usually insured at an agreed value which is paid on total loss whatever might be her true value and there is therefore the danger that an owner will be overcompensated if he can claim both for unrepaired partial loss and for a total loss all within the policy period.Although the Court's specific reasons for rejecting the merger argument did notexplicitly proceed on the basis that a cause of action had accrued to the owner before the sinking, much of the judgment is directed to whether a cause of action did accrue before the sinking. There is nothing in the Court of Appeal judgment that is inconsistent with the remarks of Tomlinson J which we have cited.42 Kastor (HC), above n 41, at [11] (emphasis added, citations omitted).43 Kastor (CA), above n 41, at [70].44 At [70].Authorities as to merger in non-marine insurance contexts[41] In Wright, Stephenson, and Co, Ltd v Holmes,45 the insured's car had been damaged in an accident on 30 January 1931. It was then towed to a garage and an estimate of the cost of repairs (at £122) was prepared. These repairs were never completed because on 3 February 1931 the garage was damaged by the Napier earthquake and subsequently completely destroyed by a resulting fire. As a result the car was damaged beyond repair. Under the policy, the insurer was required toindemnify the insured in respect of damage caused by "accidental external means"and undertook to: repair, reinstate, or replace any motor-car or any part thereof destroyed or damaged, or may settle for the damage in money.There was a specific exclusion in respect of earthquake damage.[42] The insured unsuccessfully sought to recover the estimated cost of repairs. The Court of Appeal treated the policy as amounting to a contract to repair the car if damaged.46 This conclusion is debateable as it was still open to the insurer to indemnify the owner for the damage incurred by either replacing the vehicle or, andmore plausibly, settling "for the damage in money". What is more significant forpresent purposes, however, are comments made by the Court in relation to theinsurer's argument that the damage to the car "became merged in the total loss":47It remains to consider the last point made by Mr Evans for the underwriters –that the partial damage to the car became merged in its total loss, within the doctrine promulgated by the House of Lords in British and Foreign Insurance Co v Wilson Shipping Co. To this argument Mr Cooke pointed out, in reply, that the actual decision in the Wilson Company's case applied in terms to marine-insurance policies only. In view of the opinion hereinbefore expressed regarding the construction of the present policy and its frustration,it is not necessary to decide this precise question now. It is difficult,however, to read the broad statement of the law by Lord Birkenhead, LC, in the Wilson Company's case without feeling that His Lordship's remarks therewere of wide application, and probably would be held to extend to all contracts of insurance by way of indemnity, of which the present case is a typical instance.45 Wright, Stephenson, and Co, Ltd, above n 21.46 At 822.47 At 825–826 (citations omitted).[43] In Crystal Imports Ltd v Certain Underwriters of Lloyds of London,48 a judgment delivered after that of the Court of Appeal in the present case and addressing successive losses associated with the Christchurch earthquakes in the context of a policy which provided for automatic reinstatement of the sum insured after each loss, Cooper J declined to follow the approach of Dobson J on the merger issue and applied that proposed in Wright, Stephenson, and Co, Ltd v Holmes.49[44] In Re Earthquake Commission,50 yet another case arising out of the Christchurch earthquakes, the High Court was required to deal with a conceptually similar issue as to the liability of the Earthquake Commission in relation to successive losses. The Court held, accepting the arguments of the insurers (including IAG), that the cover provided by the Earthquake Commission continued at the statutory limit after each event. The case was decided by reference to the interpretation of the relevant statutory provisions and merger principles were not discussed.[45] The view taken by Dobson J as to merger has been criticised by the authors of Colinvaux's Law of Insurance51 in the current supplement52 and slightly more elaborately in a Canterbury Law Review article by Professor Robert Merkin (who is the editor of Colinvaux).53 Professor Merkin considered that the approach taken by Dobson J was unsupported by authority and incorrect as a matter of principle.54 Inhis view, a marine insurer's liability accrues on the date of the insured peril and notat the end of the policy year and there is no practical difference between the position of a marine and non-marine insurer in respect of a total loss. The doctrine of merger applies to marine and non-marine insurance unless it is ousted by agreement. Hewas also sceptical of the argument that the "happening" nature of the coverage wascontrolling:5548 Crystal Imports Ltd v Certain Underwriters of Lloyds of London [2013] NZHC 3513.49 At [110].50 Re Earthquake Commission [2011] 3 NZLR 695 (HC).51 Robert Merkin (ed) Colinvaux's Law of Insurance (9th ed, Sweet & Maxwell, London, 2010).52 Robert Merkin (ed) Colinvaux's Law of Insurance: Second Supplement to the Ninth Edition(Sweet & Maxwell, London, 2013) at [10-005].53 Robert Merkin "The Christchurch Earthquake Insurance and Reinsurance Issues" (2012)18 Canta LR 119.54 At 129–130.55 At 130. it might be thought that the wording is not sufficiently clear to oust the indemnity principle and to allow the assured to recover for notional rather than actual losses.[46] Professor Merkin's analysis of the position as to when a cause of actionaccrues did not engage with the passages which we have quoted at [33], [36], [37],[39] and [40] above, which address, with more or less specificity, the narrow question of when a cause of action arises under a marine policy in respect of unrepaired damage. And we will come later to discuss the indemnity principle.56[47] A somewhat different approach has been taken by Kelly and Ball.57 Under the heading "[m]erger of partial loss in total loss", they say:58Whether the doctrine of merger applies to non-marine insurance has not been finally decided. The view that it does not apply to non-marine insurance is generally based on the fact that, in the case of marine insurance, the insurer'sliability for a partial loss normally arises only when the ship is repaired or the period of the policy ends ; whereas, in non-marine insurance, theinsurer's liability arises when the damage is caused.They were of the view that the doctrine of merger does not apply to non-marine insurance as they concluded this section by saying:The application of the marine doctrine to cases of non-marine insurance would be unfortunate. In particular, the rule that merger takes place even in the case of a subsequent uninsured total loss only makes sense (if at all) in cases where the insurer only becomes liable to the insured in respect of an unrepaired partial loss at the end of the period of insurance. It has nothing to commend it in the case of non-marine insurance, where it would have the undesirable effect of depriving the insured of recovery for an unrepaired partial loss for which the insurer had become liable at the time when that loss occurred.This passage plainly supports Ridgecrest's argument. But the authors also discussthe reasons given by Dobson J for rejecting the merger argument in the present case and the judgment of Cooper J in Crystal Imports Ltd declining to follow the reasoning of Dobson J.59 They saw the result in the latter case as "surely correct"and, at least by implication, that the reasoning of Dobson J was wrong. This was onthe basis of the indemnity principle, as it would be "wrong to allow the insured to56 See [53]–[61] below.57 David St L Kelly and Michael L Ball Principles of Insurance Law (looseleaf ed, Butterworths).58 At [12.0120.45].59 Crystal Imports Ltd v Certain Underwriters of Lloyds of London, above n 48.recover a windfall in respect of a loss for which he or she has already beenreimbursed as part of the indemnity provided for a total loss".60Discussion[48] Under a policy of marine insurance, a cause of action for unrepaired damage arises only when the risk expires. Although Dobson J suggested in his judgment that this rule might be a consequence of, rather than the reason for, the development of the merger principle,61 we see the cause of action rule and the merger principle as correlatives of each other. This is apparent from Lidgett and Kastor. In Lidgett, the partial loss claim did not merge into the total loss because a cause of action in respect of the partial loss arose at the end of the first policy period. And in Kastor, where there was only one insurance period, the cause of action had accrued when the vessel became a constructive total loss and was not displaced when it later sank.[49] More generally, the merger principle as applied in marine insurance cases is not a free standing legal rule which operates independently of the contractual relationship agreed upon between insurer and insured. Instead, it defines the extentof the insurer's liability in the particular circumstances of a partial loss followed by atotal loss within a single insurance period. It is thus just a component of the particular contractual relationship. If the insurance policy in Wilson Shipping had provided that, following any partial loss, the insurer was to pay the estimated costs of repair or diminution in value to the vessel resulting from the loss, a cause of action would have accrued and on the basis of the reasoning of Bailhache J, which was approved in the House of Lords, the result of the case would have been different.[50] Accordingly, we should point out the material respects in which the present policy differs from the marine insurance policies in issue in the merger cases:(a) The policy provides for both indemnity and replacement cover. It is therefore quite possible for an insured to make a profit, in the sense of60 Kelly and Ball, above n 57, at [12.0120.45].61 Ridgecrest (HC), above n 1, at [49].recovering (on a replacement basis) more than the actual (that is the indemnity) value of the building.(b) The policy does not operate on the basis of a loss assessed at the end of the risk period. Rather, it applies happening by happening.(c) Under cl C1, the insurer may be required to pay before any repairs are effected and the liability to pay is unaffected even if such repairs are not effected.(d) A cause of action in respect of the losses caused by each of the earlier earthquakes accrues immediately.(e) The liability limit is reset after each happening.62[51] The conceptual basis of the merger rule in marine insurance is that the smaller partial loss is overtaken by, and subsumed in, the subsequent larger total loss. The underlying premise is that there is no independent cause of action for the partial loss. That is a consistent feature of the marine insurance cases. To develop thistheme, we can see no basis upon which Ridgecrest's accrued causes of action inrelation to the earlier earthquakes could have been lost by reason of later events. As explained, the insurer's liability was not subject to repairs being – or being able to be – effected. The fact that the building was later in effect destroyed by the final earthquake is of no moment. Let us assume that the pre-September 2010 indemnity value of the building was $3.5m, the diminution in value caused by the earlier earthquakes was $1.75m and that the indemnity value of the building immediately preceding the final earthquake was thus $1.75m. Because the liability cap was reset after each happening, the indemnity value was, at this stage, less than the liability limit. On this set of assumptions, it is clear that Ridgecrest would be entitled to recover the aggregate of the losses associated with the earlier earthquakes ($1.75m) and the final earthquake ($1.75m). There could be no question of applying the merger principle as that would be flatly inconsistent with the scheme of the policy, under which the limit was reset after each happening.62 See [7] above.[52] The damage caused by the earlier earthquakes (and the associated diminutionin value) does not merge with Ridgecrest's entitlement under cl C2 in relation to thefinal earthquake. This is simply a consequence of the policy (a) resetting after the earlier earthquakes in relation to the building in its damaged state and (b) providing replacement cover for the building in that state. In this respect the position is analogous to that in Lidgett, where depreciation in value resulting from the partial damage to the vessel did not come into play in relation to its later total loss because the vessel had been insured on a valued basis.Is Ridgecrest's claim precluded by the indemnity principle?[53] Under the indemnity principle, an insured cannot recover more than the loss under the insurance policy. It will be recalled that both Professor Merkin and Kellyand Ball have expressed the view that Ridgecrest's claim falls foul of the indemnityprinciple. This proposition was not developed in any detail by Professor Merkin (albeit that he recognised that application of the indemnity principle might arguably not be compatible with the resetting of the liability cap after each event). The point is discussed in more detail by Kelly and Ball. The authors assumed that it was Ridgecrest that had expended money on partial repairs and, more importantly, thatthe claim in respect of the earlier earthquakes was for the amount that "would have been spent on repairs if the premises had not become a total loss".63 There was, however, no reference to the terms of the policy under which there was a limit of liability but where this limit was reset after each happening. As well, the authors did not address the case on the basis that after the earlier earthquakes, the building, in its damaged state and thus diminished value, remained insured and that the reset liability limit applied to the building in that state.64 That said, the indemnity principle question does warrant careful consideration.[54] We accept that the indemnity principle (which is a slightly awkward phrase in the context of a replacement policy) would preclude the recovery of more than the replacement value of the property insured. As we understand it, however, Ridgecrest is not seeking to recover more than the replacement value of the building. If so,63 Kelly and Ball, above n 57, at [12.0120.45].64 Thus the line of argument set out in more detail at [51] above is not addressed at all.there is no double counting in that respect. We accept that this principle might also apply more broadly, for instance in relation to any separately identifiable building component which is first damaged and then destroyed by successive events. In thatsituation, the insurer's liability would be confined to the replacement cost of that building element so that any diminution in value to that element resulting from an earlier event could not be separately recovered.[55] Leaving aside the possibility of double counting of the kind just discussed, the indemnity principle is only engaged if $1,984,000 is deemed to be the replacement value of the building so that payment by IAG of that amount (less some deductions) to Ridgecrest is to be regarded as the equivalent of payment of the replacement value of the building.[56] The only cases of which we are aware which bear directly on this issue comefrom America and they do not support IAG's position.[57] The first of these is Palilla v St Paul Fire and Marine Insurance Co.65 The insured had insured the contents of a building for $45,000. There were two fireswhich caused total losses of $62,500. The policy provided that "[a]ny loss hereunder shall not reduce the amount of this policy". Under this clause, the Courtconcluded:66 the only logical interpretation that we can arrive at is that where thereare successive losses, each loss will be treated independently under the policy – full recovery may be had for each separate loss up to the full amount of the coverage under the policy. To construe it otherwise would reduce the amount of the policy after each loss which would be in violationof the loss clause.[58] The second case, O'Bryan v Columbia Insurance Group,67 involved the insured house being damaged by two fires in a single policy period. Under the policy, there was a liability cap of $40,000. Following the first fire, the insured was paid $37,105.50. Not long afterwards, there was a second fire. None of the damage caused by the first fire had been repaired. The Supreme Court of Kansas construed65 Palilla v St Paul Fire and Marine Insurance Co 322 So 2d 46 (Fla DCA 1975).66 At 48.67 O'Bryan v Columbia Insurance Group 56 P 3d 789 (Kan SC 2002).the policy cap as applying on an occurrence by occurrence basis and in holding for the insured held that it was irrelevant that the damage caused by the first fire had not been repaired. Although the replacement value of the house is not specifically referred to in the case, it would appear to have been greater than $40,000 and there is no suggestion that the insured would receive more than the house was worth. The case thus seems to be practically on all fours with this case.[59] Also closely on point is the third case, Young v American Security Insurance Co.68 The facts were similar to those in O'Bryan: house insurance, a liability cap (in this instance $75,000) and two fires. In this case, the house was destroyed by the second fire before the damage caused by the first had been repaired. The policy provided replacement cover which was not to exceed the policy cap. There was a loss clause similar to that in Palilla. The losses caused by the fires exceeded $75,000 and the fact pattern generally seems to be the same as the present case. The Court of Appeals of Michigan, by a majority, held that the insured was entitled to make separate claims, each up to the policy limit in respect of each fire,albeit that it held that the claim in respect of the second loss was "limited by thediminished value of the property after the first loss".69 The dissenting judgedisagreed on the basis that the insured "suffered but a single loss the destruction of his dwelling by fire".[60] The judgments referred to are all comparatively brief. What they support, however, is an approach based firmly on the policy wording as to the resetting of liability limits.[61] As we have already noted, the indemnity principle is only engaged if $1,984,000 is deemed to be the replacement value of the building so that payment by IAG of that amount (less some deductions) to Ridgecrest is to be regarded as the equivalent of payment of the replacement value of the building. We are of the view that this would not be an appropriate approach for us to take in respect of this policy. This is for four reasons:68 Young v American Security Insurance Company (unreported) Mich Ct App 11 January (2007 WL 81804) .69 The judgment does not explain how this worked given the replacement value of the policy.(a) First and foremost, IAG has never presented its case on this basis, which, if it were correct, would be decisive of the case.(b) Such an approach would result in the liability cap being applied to more than one happening, which is inconsistent with that cap resetting after each event and would involve a very substantial rewriting of the liability cap provision in favour of IAG.(c) The reason why payment of full replacement value would have mopped up the liabilities in respect of the earlier earthquakes is because full replacement under cl C2 would necessarily not onlydischarge IAG's liability for the final earthquake but would also coverall losses caused by the earlier earthquakes. In the present case, however, the payment made was not enough to do so.(d) This approach would thus treat the liability cap as if it represented an agreed replacement value of the building when that is not the way the policy is structured. This is not a valued policy (as marine insurance policies often are). There is no indication in the policy that the liability cap was intended to represent the replacement value of the building. Indeed, it was well understood by the parties that the liability cap was not based on, and was less than, the replacement value of the building.Conclusion[62] Before we close, it is desirable to bring together, in summary form, the key points we have determined. The rights of Ridgecrest under its policy with IAG are subject to three limits: first, there can be no double counting; secondly, each happening gives rise to a separate limit in respect of which the contractual limit of $1,984,000 applies; and thirdly, the total of all claims cannot exceed the replacement cost of the building.[63] For those reasons, and thus with those caveats, the preliminary question is answered in the affirmative with the result that the appeal is allowed. The appellantis awarded costs of $25,000 together with reasonable disbursements to be fixed by the Registrar in relation to the appeal. The orders for costs in the High Court and Court of Appeal are set aside and the respondent is to pay the appellant costs in those courts to be fixed by those courts.Solicitors:Parry Field Lawyers, Christchurch for AppellantFortune Manning, Auckland for Respondent