PacLII Home | Databases | WorldLII | Search | Feedback

Supreme Court of Fiji

You are here:  PacLII >> Databases >> Supreme Court of Fiji >> 2023 >> [2023] FJSC 30

Database Search | Name Search | Recent Decisions | Noteup | LawCite | Download | Help

  Download original PDF


The New India Assurance Co Ltd v Kalabo Investment Ltd [2023] FJSC 30; CBV0012.2021 (31 August 2023)

IN THE SUPREME COURT OF FIJI
[APPELLATE JURISDICTION]


CIVIL PETITION NO: CBV0012 OF 2021
Court of Appeal No. ABU 0091 of 2017


BETWEEN:


THE NEW INDIA ASSURANCE CO LTD
Petitioner


AND:


KALABO INVESTMENT LTD
Respondent


Coram: The Honourable Mr. Justice William Calanchini
Judge of the Supreme Court
: The Honourable Mr. Justice Terence Arnold
Judge of the Supreme Court
: The Honourable Mr. Justice Filimone Jitoko
Judge of the Supreme Court


Counsel: Mr. R. R. Gordon for the Petitioner
: Mr. C. B. Young for the Respondent


Date of Hearing : 4 August 2023
Date of Judgment : 31 August 2023


JUDGMENT

Calanchini, J


[1] I have read in draft form the judgment of Arnold J and agree with his reasoning and conclusions.
Arnold, J


Introduction


[2] The Respondent, Kalabo Investment Ltd (Kalabo), owned supermarkets in Nadi, Ba, Lautoka, Tavua and elsewhere. On 30 March 2012, refrigerated goods in three of its supermarkets were damaged as a result of power failure that resulted from a flood (the March flood). Later that year, on 17 December, refrigerated goods in three other supermarkets owned by Kalabo were damaged when the power supply failed as a result of Cyclone Evan (the December cyclone).


[3] Kalabo held a material damage and business interruption insurance policy with the Petitioner, The New India Assurance Company Limited (New India).[1] The effect of the policy was to insure Kalabo against loss or damage caused to refrigerated goods in the supermarkets as a result of power failures caused by floods and cyclones. The policy specified that New India’s liability in relation to damage to refrigerated goods would be limited to $150,000 for “any one loss”.[2] The policy also contained two clauses which aggregated losses or events for certain purposes, most importantly a clause referred to as the “72-hour” clause.


[4] The principal issue before the Court concerns how the insurance policy applies to the losses to refrigerated goods at Kalabo’s supermarkets as a result of power failures resulting from, first, the March flood and, second, the December cyclone. The parties agree that damage to refrigerated goods is covered under the policy. The dispute concerns the extent to which the policy limits New India’s liability, specifically through the application of the 72-hour clause.


[5] The difference between the parties is significant. In relation to the March floods, the damage to refrigerated goods at Kalabo’s three supermarkets totalled $504,213.84. This total was made up of losses at the three individual supermarkets of $157,525.97, $183,476.95 and $163,210.92 respectively. New India argues that the effect of the 72-hour clause is that the losses should be aggregated and treated as one loss for the purpose of the clause limiting New India’s liability to $150,000 for “any one loss”. Accordingly, on New India’s interpretation, Kalabo is entitled to receive $150,000 in total for these losses.


[6] For its part, Kalabo argues that the 72-hour clause does not apply in the way that New India contends. It argues that under the policy there were three losses, one at each supermarket, so that each is independently subject to the $150,000 limitation. Consequently, Kalabo argues, it is entitled under the policy to receive $450,000 for its losses from the power failures following the March floods.[3]


[7] The same arguments are made in connection with the refrigerated goods losses at the three other supermarkets following the power failure resulting from the December cyclone. There, the total refrigerated goods losses were $285,050.60, comprising $151,121.51 at the first supermarket, $82,284.55 at the second and $51,644.54 at the third. On Kalabo’s interpretation of the relevant clauses, it would receive almost the full amount of its losses ($283,929.09); on New India’s interpretation, Kalabo would receive $150,000.00.


[8] New India offered to pay Kalibo the $300,000 that it accepted was owed under the policy (i.e., $150,000 for each set of losses), but its offer was conditional on Kalabo giving a discharge acknowledging that the $300,000 was in full and final settlement of its claims. Kalabo refused to agree to this and issued proceedings against New India.


Kalabo’s Proceedings


[9] In its proceedings, Kalabo sought payment of $789,264.44 (or as much of that sum as was found to be due and payable), plus interest at 10% per annum compounded under the s 34 of the Insurance Law Reform Act 1996.[4] In addition, it applied for summary judgment for the sum that New India accepted was payable under the policy, ie, $300,000 (less deductible), together with interest on that sum at 10%. The Master granted the application for summary judgment, with the result that New India paid that amount plus interest at 10%. This left the matter to go to trial on the remaining amount.


[10] Before trial, the parties agreed that no witnesses would be called. Rather, the trial would be conducted on the basis of an agreed statement of facts and the documentary evidence before the Court. In the result, the High Court considered that Kalabo’s interpretation of the policy was correct and awarded judgment for the amount sought, together with interest at 10%. This decision was upheld in the Court of Appeal. New India now seeks leave to appeal to this Court.


The Application for Leave


[11] As set out in its petition for leave, New India’s grounds are extensive. For present purposes, however, I summarise them as follows:


(a) The Court of Appeal was wrong to hold that, in terms of the insurance policy, the refrigerated goods losses at each of Kalabo’s three supermarkets following the March flood were independent losses for the purpose of the policy limit of $150,000. Similarly, it was wrong when it found that the refrigerated goods losses at Kalabo’s three other supermarkets following the December cyclone were independent losses for the purpose of the $150,000 limitation.

(b) The Court of Appeal was wrong to uphold the High Court’s award under s 34 of the Insurance Law Reform Act of interest at 10% compounding annually on the sum remaining to be paid at the time of trial (ie, excluding the $300,000 paid by New India following summary judgment).

(c) The Court was wrong to reject New India’s argument based on estoppel by convention.

[12] Under s 7(3) of the Supreme Court Act 1998, the Court may not grant leave to appeal unless the case raises a far-reaching question of law, a matter of great general or public importance or a matter that is otherwise of substantial general interest to the administration of civil justice.


[13] In my view, leave to appeal should be granted on the first two matters set out in paragraph [11], but not on the third.


[14] In relation to the first issue, the interpretation of this policy is likely to have wider application than simply between the two parties before the Court. Fiji is prone to heavy rain that results in widespread flooding and to other forms of destructive natural disasters. How the clauses at issue in this case operate in these circumstances is a matter of more general interest than simply these parties. As neither of the Courts below undertook a thorough analysis of the policy wording, I consider that this Court must do so. In relation to the second issue, this Court has recently considered the application of s 34 of the Insurance Law Reform Act: see New India Assurance Co Ltd v Punjas & Sons Ltd.[5] However, I would grant leave in the present case given that the relevant principles were not set out in the judgments of the Courts below and can usefully be restated in the context of this appeal.


[15] The reason that I would deny leave on the “estoppel by convention” issue is that it is based on the settlement of a claim under an earlier insurance policy between the parties, referred to as the 2009 policy. This, it is argued, established a course of dealing between the parties sufficient to establish an estoppel by convention. However, that policy is not in evidence before the Court.[6] It is not referred to in Kalabo’s affidavit of documents, but it is referred to in that of New India. Given its importance to the estoppel argument, it should have been in evidence. Other evidence would also have been needed. Mr Gordon tried to address this evidentiary deficiency by relying on the pleadings. However, I do not see how the Court can address the estoppel by convention issue appropriately on basis of the material before us. Accordingly, I would not grant leave on that issue.


The Policy Wording


[16] As a consequence of the agreement made between the parties before trial, I must proceed on the basis of the agreed statement of facts and the documentary evidence. The critical documentary evidence is the insurance policy,[7] the relevant provisions of which I will now describe.


[17] The policy was entered into by way of a placing slip prepared by Kalabo’s broker, AON. The resulting policy, which was also prepared by AON, provided what was, in effect, “all risks” cover for material damage and for business interruption. The policy consisted of a title page, a Schedule, which set out some important terms of cover; separate Sections dealing with material damage and with business interruption; Conditions, which contained conditions applicable to both material damage and business interruption as well as conditions applying to each separately; and Memoranda, again both common and individual.


[18] The material damage section provided cover against “all risks of physical loss or damage – unintended and unforeseen by the insured − not otherwise excluded by the policy”. I highlight four features of the material damage cover.


(i) Refrigerated goods cover

[19] First, the policy provided refrigerated goods cover, as follows:


Refrigerated Goods:

This Policy extends to cover:

(a) damage to goods stored in in refrigerated cabinets or chambers arising from accidental stoppage or malfunction of refrigerating plant from any cause whatsoever except as provided below.
(b) ...

Special Provisions


  1. The words ‘goods stored’ are deemed to include goods anywhere on the premises at the time of the loss causing event and that, would but for the event, have been placed in the refrigerated cabinets or chambers.
  2. In respect of loss due to the disconnection of the public electricity supply, this extension only covers the loss where the electricity supplier has given less than 24 hours prior notice of the disconnection.”

[20] Accordingly, the policy provided cover where there was an accidental stoppage of the refrigerating plant “from any cause whatsoever” (subject to an irrelevant exception). Special provision 2 makes it clear that losses resulting from interruption of the public power supply fall within the clause (where the electricity supplier has given less than 24 hours’ notice of disconnection). As noted, the parties agreed that Kalabo’s cover under this extension applied to the refrigerated goods losses resulting from the power failures resulting from the March flood and the December cyclone.


(ii) Liability limits

[21] Second, the policy limited New India’s maximum liability for, among other things, flood and refrigerated goods claims:


Special Limit:


The Company’s maximum liability for any one loss (or otherwise as applicable) under the following Specifications is limited as shown:

...

Water including Flood – (MD&BI)

per event per location and $200,000 per location

in annual aggregate and during the policy period

$350,000 per location and in annual aggregate

...

Refrigerated Goods − Limit any one loss .................. $150,000

...”


There does not appear to be any limitation of liability clause in relation to tropical cyclone cover.


[22] Whereas the refrigerated goods special limit applies to “any one loss”, the flood damage limit has several elements - a “per event, per location” limit, and a location-based annual aggregate limit. Looking at the other Special Limits provided for in the Schedule, most apply to “any one loss” but one applies to “any one location”. As is clear from the Schedule of Locations/Sums Insured, the reference to “location” is a reference to the individual supermarkets insured under the policy. The different limits in the clause indicate that specific consideration was given to how the various limits were to be formulated.


(iii) Deductibles

[23] Third, the policy’s provisions in relation to the deductibles for two perils and one type of damage − tropical cyclones, flood and refrigerated goods − were as follows:


Deductibles:

Applicable to each and every loss or series of losses arising out of one event... From each adjusted loss the amount specified below will be deducted.


A series of events arising from any one cause during any period of 72 consecutive hours will be treated as one event for the purpose of applying the Deductible.

...

Tropical Cyclone 10% of final adjusted loss or F$1,000 whatever greater but not exceeding 10% of Sum Insured as stated in the within policy

Flood $15,000 any one location any one loss


Refrigerated Goods $1,500”

[24] For the purposes of the discussion which follows, four features of this provision are noteworthy:


(a) The clause distinguishes between a “loss”, a “series of losses”, an “event”, a “series of events” and a “cause”.

(b) The aggregation provision relating to a series of events arising from one cause during any 72-hour period means that several events will be treated as one event for the purpose of applying the deductible.


(c) The tropical cyclone deductible must not exceed “10% of the Sum Insured as stated in the within policy”. The reference to “sum insured” must be a reference to the sum insured for the particular location suffering cyclone damage. The sums insured for the individual insured locations are set out in a schedule.


(d) Whereas the flood deductible refers to a loss at a location, the refrigerated goods deductible is simply an amount.


(iv) 72-hour clause

[25] Fourth, the policy contained a 72-hour clause which limited New India’s liability in certain circumstances:


“72-hour clause:

All insured losses that occur during a period of 72 hours consecutive hours caused by:


(a) Earthquake, earth tremor, seaquake tidal wave or any other loss from seismic activity insured under this Insurance
(b) Volcanic eruption
(c) Hurricane, typhoon, tornado, windstorm wind driven water or other wind peril insured under this Insurance
(d) Flood

shall be deemed a single loss occurrence for the purposes of this insurance.


Any such event, which continues for a period exceeding 72 consecutive hours shall be, deemed two or more events.


The insured may choose the date and time when such loss period of 72 hours shall commence provided that:


(i) This is not earlier than the first recorded loss sustained by the insured.
(ii) The date of commencement falls within the period of this Insurance.
(iii) No two or more periods of 72 hours shall overlap.”

Again, this clause is a form of aggregation clause. Its meaning is at the heart of the present dispute.


Approach to interpretation of insurance policies


[26] Section 29 of the Insurance Law Reform Act 1996 sets out the rules of construction to be observed in the interpretation of insurance policies. It provides:


“29 Notwithstanding any law or agreement to the contrary, the following rules of construction shall be observed in the interpretation of any proposal for insurance or any policy of insurance or endorsement on a policy of insurance—


(a) the intention of the parties, ascertained from the face of the documents, documents incorporated therewith and surrounding circumstances, shall prevail;

(b) the whole of a document shall be looked at and not a particular clause;

(c) written words shall ordinarily be given more effect than printed words;

(d) wherever possible, the grammatical construction shall be adopted, but the intention of the parties shall be of paramount consideration;

(e) words shall be construed in their plain, ordinary, popular, commonsense and natural meaning except that terms of art or technical words shall be understood in their strict, technical and proper sense unless the context controls or alters the meaning;

(f) the meaning of a word is to be ascertained with reference to its context and may be restricted or modified thereby, and where, from the context, it appears that the parties intended to use the word in a special and peculiar sense, and not in a meaning which it might otherwise bear, the word shall be construed in accordance with their intention;

(g) subject to the precise terms, subject matter and context of a clause, where specifications of particular things belonging to the same genus precede a work of general signification, the latter word of general signification, shall be confined in its meaning to things belonging to the same genus and shall not include things belonging to a different genus;

(h) where a word of general signification is followed by words of limitation or definition, which introduce words of narrower signification, the first word shall not be taken in its full sense but shall be construed as limited by and applying only to the particulars specified;

(i) words shall be construed to mean what they say, unless there is some strong ground for placing a different construction on the words from what they naturally import;

(j) words shall be construed liberally so as to give effect to the real intention of the parties and the document shall not be so construed as to defeat the object of the transaction or as to render it illusory;

(k) in any case of ambiguity, where words are capable of more than one construction, the reasonable construction shall be taken to represent the intention of the parties;

(l) the language of a document shall not be strained in favour of or against any party but if there is any ambiguity, the ambiguity shall be resolved in favour of the person insured;

(m) every effort shall be made to reconcile inconsistencies, but where there is an inconsistency between the wording of a policy and that in the proposal or any earlier document, the policy shall be regarded as expressing the true intention of the parties in the absence of sufficient evidence to the contrary;

(n) an express term shall override any implied term inconsistent with it.”

[27] Notably, s 29 does not provide for the operation of the contra proferentem rule (i.e., the rule stating that where a clause is ambiguous, it should be interpreted against the interests of the party responsible for including it in the contract). Rather, s 29(l) provides that in cases of ambiguity, the ambiguity will be resolved in favour of the insured.


Analysis


[28] As I have said, the policy comprises a title page, Schedule, two Sections dealing Material Damage and Business Interruption, Conditions and Memoranda, some common to both material damage and business interruption cover, others applying only to one or other. The way such policies are put together can lead to difficulties of interpretation, sometimes requiring, as one Judge put it, “some mental gymnastics”[8] to see how the agreed terms operate and which clauses prevail over others. This is a feature of the present case.


[29] Aggregation clauses take many guises, appearing in different insurance and reinsurance settings and often create difficulties of interpretation.[9] Sometimes they work to the benefit of an insured, sometimes to the benefit of an insurer. So, in the present case, one of the aggregation provisions affects deductibles in a way that favours the insured, whereas the other limits liability in a way that favours the insurer. In the context of aggregation clauses in professional indemnity policies, the United Kingdom Supreme Court has said:[10]


“Because [aggregation] clauses have the capacity in some cases to operate in favour of the insurer (by capping the total sum insured), and in other cases to operate in favour of the insured (by capping the amount deductible per claim), they are not to be approached with a predisposition towards either a broad or a narrow interpretation.”

I agree.


[30] In relation to the refrigerated goods damage at the three supermarkets following the March flood, Mr Young for Kalabo argued that the 72-hour clause did not apply to limit Kalabo’s recovery to $150,000 for two reasons:


(a) First, he argued that the purpose of the clause was to aggregate events which caused losses, rather than to aggregate losses arising from a single event.

(b) Second, he argued that the cause of the damage at Kalabo’s three supermarkets was the power failure, not the flood.

Similar arguments were made in the case of the losses resulting from the December cyclone. I deal with each argument in turn.


Is the purpose of the 72-hour clause to aggregate events, not losses?


[31] Mr Young put his first argument as follows in his written submissions:


“all losses occurring within 72 hours which are caused by two or more events stated in (a) to (d) are treated as caused by one event for the purpose of applying Deductibles or special limit per event provided in the policy. The clause aggregates events causing the losses for the purpose of Deductible or per event limit.” [Emphasis omitted].


[32] Mr Young stressed the use of the word “occurrence” and said that it meant “event”. He suggested that the word “loss” in the phrase was “surplusage”. He drew support for this from the opening words of the next sentence in the provision − “Any such event” − which he argued referred back to “occurrence”. He also drew support from the Deductibles clause, which says:


A series of events arising from any one cause during any period of 72 consecutive hours will be treated as one event for the purpose of applying the Deductible.


[33] Mr Young is correct that the word “occurrence” in insurance policies is generally synonymous with “event”. Often it will refer to something which happens at a particular time in a particular way at a particular place. But that is not always the case. As Lord Briggs said in Financial Conduct Authority v Arch Insurance (UK) Ltd:[11]


“Depending upon context, the word “occurrence” can properly be applied to happenings which do not take place at a single specified time, in a particular way and at a particular location. Thus a hurricane, a storm or a flood may properly be described as an occurrence even though each may take place over a substantial period of time, and over an area which changes over time.”


[34] I do not accept Mr Young’s argument that the purpose of the 72-hour clause was to aggregate events rather than losses, as I now explain.


[35] The 72-hour clause can be broken down into six elements:


(1) all insured losses
(2) that occur during a period of 72 consecutive hours
(3) caused by a named peril
(4) shall be deemed (ie, treated as)
(5) a single loss occurrence
(6) for the purposes of the insurance policy.

[36] If one asks “what is deemed to be a single loss occurrence?”, the answer is all insured losses occurring during a 72-hour period caused by any of the named perils. As this indicates, the clause aggregates losses, not events. There is no wording suggesting that the purpose of the clause is to aggregate two or more events. For example, the clause simply says “caused by”, not “caused by any two or more of”, the named events.


[37] Accordingly, in my view, the provision means that all insured losses occurring within a 72-hour period arising from a named event are to be treated as the occurring of a single loss for the purpose of the policy. The special limits clause is, of course, an important component of the policy.


[38] This interpretation of the phrase “a single loss occurrence” is supported by the remainder of the 72-hour clause. When the clause goes on to refer to “Any such event”, the word “event” is not, as Mr Young contended, referring back to the word “occurrence”, but rather back to whichever of the named perils caused the losses. Moreover, the clause states that when one of the named events exceeds 72 hours, it can be treated as two or more events. In addition, the clause allows the insured to choose when the 72-hour period starts, provided that it is not earlier that the “first recorded loss sustained by the insured”. This again indicates that the 72-hour clause is intended to aggregate losses, not events.


[39] This interpretation also derives some support from the only other instance in the policy where the phrase “loss occurrence” is used. As I discuss further when I address Mr Young’s second argument, the policy insured against damage “directly caused by water”, including flood water. The policy went on to provide:


“With respect to flood, loss occurrence shall mean all losses, wherever occurring, which arise between the time of movement of water into, onto, or over the property insured hereunder and the receding of the same, regardless of the time so embraced; EXCEPT, no loss occurrence shall be deemed to commence earlier than the date and the time of the happening of the first recorded individual loss to the Insured in that occurrence during the period of this insurance, nor to extend beyond three days after the expiry of this Insurance.” [Emphasis added.]


Here, “loss occurrence” aggregates a series of losses from a single event (a flood) – it does not aggregate events.


[40] There is one further aspect of the 72-hour clause that I mention now but will discuss later in these reasons, which is whether the clause aggregates losses at different locations where damage is suffered during the 72-hour period.


Were the losses caused by power failure, not by flood?


[41] Mr Young’s second argument was that the refrigerated goods losses in March 2012 were caused by power failure, not by flood, so that the 72-hour clause did not apply. Mr Young said that the flood had facilitated the loss, but did not “cause” it, i.e., the flood was not the proximate (or efficient) cause of the losses. Similarly, the refrigerated goods losses following the December cyclone were caused by power failure, not Cyclone Evan.


[42] The flooding occurred on 29-30 March 2012, when a tropical depression produced very heavy rainfall over a short period and resulted in widespread flash-flooding, especially in the Western Division of Viti Levu, causing loss of life and substantial damage to infrastructure and other property. It took some time to restore the power supply.


[43] The starting point for analysis is that the policy provided cover for refrigerated goods losses caused by (among other things) a power supplier’s decision to discontinue electricity supply (providing the supplier gave less than 24 hours’ notice of the discontinuance). Obvious situations where a power distributor in Fiji would discontinue supply are where flooding and/or tropical cyclones create conditions which make it dangerous to continue supply, whether from the viewpoint of customers or of the network. The refrigerated losses clause would also provide cover where the power supply network failed because widespread flooding and/or a tropical cyclone damaged key equipment, even if that flooding or tropical cyclone did not enter or damage Kalabo’s insured premises. So, the refrigerated goods losses that Kalabo suffered were contemplated in terms of the policy.


[44] The 72-hour clause deals with perils that take place over a period of time (including floods and tropical cyclones) and cause insured losses. As the refrigerated goods losses were contemplated in terms of the policy, and floods and/or tropical cyclones contributed in a causative sense to those losses, on the face of it the 72-hour clause would apply. But Mr Young argues that there is no legally effective causal connection between the flood and cyclone and the losses at issue.


[45] Lord Wright famously said in Yorkshire Dale Steamship Co Ltd v Minister of War Transport that the choice of the real or efficient cause from out of the whole complex of facts must be made applying common sense standards – “causation is to be understood as the man in the street, and not as either the scientist or metaphysician, would understand it”.[12] However, this does not mean that it is “a matter of choosing a cause as proximate on the basis of an unguided gut feeling”.[13] A principled approach must be taken, which involves a careful consideration of the language of the policy, interpreted objectively, and an assessment of the facts against the policy wording.


[46] In considering this issue, I begin with the agreed statement of facts. Replicating the Kalabo’s statement of claim, it says:


Whilst the policy was current, [Kalabo’s] refrigerated goods were damaged:


(a) in its stores at Market Subdivision, Ba; Sahu Khan St, Nadi and Lodhia Street as a result of FEA’s power failure caused by floods on 30 March 2012;[14] and

(b) in is stores at Yasawa St, Lautoka; Market Rd, Nadi and Main St, Tavua as a result of FEA’s power failure caused by Cyclone Evans on 17 December 2012.” [Emphasis added.]

[47] This description of the power failures being “caused” by the flood and Cyclone Evan respectively occurs throughout the parties’ correspondence concerning Kalabo’s claims, as Mr Gordon emphasised in his submissions. But that does not necessarily answer the point of interpretation before the Court.


[48] Focusing on the issue of flooding, the Agreed Statement’s use of the description “as a result of [the Fiji Electricity Authority’s] power failure caused by floods” is unhelpful. It could mean that:


(a) the power supply failed because flash flooding affected the operation of the network (by knocking out key components, for example); or

(b) the FEA deliberately discontinued the power supply because of concerns about the effect on the network of widespread flooding, actual or anticipated; or (possibly)

(c) flooding on the insured’s premises either resulted in equipment damage which caused the power supply to fail or caused Kalabo’s employees to shut down the power supply at the premises for safety reasons.[15]

The precise reason for the power failure matters in terms of the operation of the policy. Given the description in the Agreed Statement of Facts, the first of the options is the most likely, but the parties should have provided more detailed information on this aspect in the Agreed Statement of Facts.


[49] The policy provides specific cover for flood damage. The cover provided is for damage to the insured’s property “directly caused by water” including flood water, subject to several irrelevant exclusions. The use of the term “directly” is significant. It can be contrasted with the broader language used in other clauses, for example, that dealing with land movement. There, the policy says it does not insure “against loss or damage directly or indirectly caused by or resulting from landslip, subsidence or erosion of land”. [Emphasis added.]


[50] The flood cover provision then gives a wide definition of “flood”, as follows:

Definition of Flood:


“If Flood is insured under this Section, flood shall mean rising water; surface water; waves; tidal waves or tidal water; overflow of streams, rivers, lakes, ponds or other bosies of water; or spray from any of the foregoing; all whether driven by wind or not.


With respect to flood, loss occurrence shall mean all losses, wherever occurring, which arise between the time of movement of water into, onto, or over the property insured hereunder and the receding of the same, regardless of the time so embraced; ....”


Although it is headed “Definition of Flood”, this clause really does two things. First, it provides a definition of “flood”; second, it provides a definition of what “loss occurrence” means in the context of the flood cover.


[51] What emerges from this is that there are two elements to the policy’s flood coverage, which reinforce each other – first, that the insured damage must be directly caused by water and second, that the damage must occur while the water is on the insured’s property. In short, the policy provides what is sometimes referred to as “on premises” flood cover.


[52] As already noted, the policy contains a provision limiting New India’s liability in relation to flood damage on a “per event per location” basis, with a further annual aggregated location-based limit. Similarly, the deductible applicable to flood damage is for “any one location any one loss”. This differs from the special limit applying to damage to refrigerated goods, which is simply a “per loss” limit.


[53] How does the 72-hour clause operate in respect of damage caused by flooding? It is useful to begin by considering the position under the clause in relation property damage at one of Kalabo’s supermarkets other than damage to refrigerated goods. To meet the words of the 72-hour clause – “any insured losses ... caused by ... Flood” − the property damage would have to meet the requirements of being directly caused by water while water was on the insured’s property. If those elements were met and the flooding persisted for 72 hours, the losses caused over that period would be treated as “a single loss occurrence” under the 72-hour clause.


[54] The Special Limits clause in relation to flood damage would then come into play. It will be recalled that the clause provides:


“The Company’s maximum liability for any one loss (or otherwise as applicable) under the following Specifications is limited as shown:

...

Water including Flood – (MD&BI)

per event per location and $200,000 per location

in annual aggregate and during the policy period

$350,000 per location and in annual aggregate”


The application of the Special Limit would depend on where the particular supermarket was located. But in principle, there would be no difficulty with aggregating the losses caused by flooding under the 72-hour clause at that supermarket for the purpose of the Special Limit.


[55] I now consider the situation where the flood caused property damage other than refrigerated goods damage at three of Kalabo’s supermarkets rather than at one. How would the 72-hour clause apply then? In particular, could the losses at the three locations be aggregated on the basis that they are insured losses caused by flood?


[56] In my view, the losses across the three supermarkets could not be aggregated because:


(a) it is inconsistent with the “per location” element of the Special Limit and the deductible provision; and

(b) it would be impossible to apply the Special Limit’s annual aggregated limits.


It is difficult to see how losses aggregated across several locations and deemed to be “a single loss occurrence” could be allocated to particular locations appropriately for the purpose of the Special Limits. It is also unclear how the deductible provision would operate in those circumstances.


[57] This indicates that the 72-hour clause does not operate to aggregate losses across different insured locations in respect of flood damage. If the clause does not operate to aggregate losses across different insured locations in relation to flood damage, it is difficult to see how it could operate to do that in the context of the other named perils as that would require giving the words “all insured losses” in the 72-hour clause a different meaning for different perils. Making it clear that the clause was intended to aggregate losses across locations would have been straightforward.


[58] The foregoing discussion has addressed a situation where flooding caused property damage at insured locations other than damage to refrigerated goods. I now turn to the situation where flooding has caused the power supply to fail[16] at three of Kalabo’s insured locations and that has led to damage to refrigerated goods at each location. (For these purposes, I am assuming that there was no cover under the flood provisions because their causal and loss occurrence requirements were not met.) The question is how the 72-hour clause applies in that situation.


[59] The first issue is whether the clause applies at all. This depends on what is meant by “All insured losses ... caused by ... Flood”. I make three points.


[60] First, although the insured loss in this situation would not meet the definition of “loss occurrence” under the flood cover, there would be an “insured loss” under the refrigerated goods cover. So, the “insured loss” element of the 72-hour clause would be met.


[61] Second, there is the issue raised by Mr Young, ie, whether that loss was “caused” by flood, given that flooding was not the direct cause but triggered other events which led to the refrigerated goods damage. In this context, I reiterate that the 72-hour clause refers to losses “caused by” flooding rather than using the broader language found elsewhere in the policy of “directly or indirectly caused by or resulting from” flooding, which would clearly cover storm-related power failures leading to refrigerated goods losses.


[62] There is a clause in Business Interruption cover provided under the policy that is potentially relevant here. The clause is headed “Off Premises Utility Services”. It provides Business Interruption cover for:


“... loss resulting from necessary interruption of business caused by Direct Physical Loss or Damage, as covered by the Insurance to which this extension is attached, to gas, electric, water and telephone supplying the Insured premises. Limited to within one statute mile of the Insured premises”.


This clause recognises that the Material Damage section of the policy may provide cover where an insured peril causes off-site power failures that damage the insured’s property. It might be thought to be relevant to the issues in relation to the 72-hour clause. However, I do not think the clause helps, given that the refrigerated goods cover is part of the material damage section of the policy and clearly provides cover where losses result from off-site power failures. Accordingly, I do not see this clause as suggesting any particular interpretation of the 72-hour clause.


[63] On the meaning of the phrase “caused by” in the 72-hour clause, then, I incline to the view that it requires a direct causative connection to the loss rather than being the trigger for other processes which result in loss. However, this is not an issue I need to resolve because of my view on the next point.


[64] Third, even if the refrigerated goods losses could be regarded as “caused by” flooding for the purposes of the 72-hour clause, I do not accept that the 72-hour clause contemplates aggregating losses across different insured locations, for the reasons given above.

[65] The final point I make is that, to the extent that there is any relevant ambiguity in the policy, s 29(l) of the Insurance Law Reform Act requires that it be resolved in favour of the insured (even though, in this case, the policy is that of the insured’s broker).


Summary of reasoning


[66] Given the complexity of the foregoing reasoning, I give the following brief summary:

(a) The factual material before the Court does not indicate precisely how the power disruptions occurred. I have assumed from the way the Agreed Statement of Facts is expressed that either the network failed as a result of the storm events or the Fiji Electricity Authority (as it then was) closed the network down anticipating that, otherwise, the storm events would cause the network to fail. This means I have assumed that it was not the impact of flood water on the individual insured premises that caused their power supply to fail. The Agreed Statement of Facts should have given greater detail about this.

(b) Contrary to Kalabo’s submissions, the 72-hour clause aggregates losses, not events. This is clear from the way the clause is expressed.

(c) In respect of flood damage as defined in the policy (ie, damage directly caused by flood water while flood water was on the premises), the 72-hour clause aggregates losses occurring at individual insured locations during a 72-hour period. But it does not aggregate losses occurring at multiple insured locations during that period. This is because the limits of liability for flood damage are expressed to be “per-event per location” and the deductible is expressed to be “any one location any one loss”. It is difficult to see how these provisions could operate if losses were aggregated across locations, as argued by New India.

(d) If the 72-hour clause does not allow the aggregation across insured locations of losses from flood damage, it is difficult to see how it does so for losses from the other perils identified in the clause.

(e) Given the assumptions I have made based on the Agreed Statement of Facts about the cause of the power failures, the damage to Kalabo’s refrigerated goods does not constitute flood damage as defined in the policy because it was not directly caused by flood water while flood water was on the insured’s premises. It does, however, fall within the refrigerated goods cover, which contemplates power failures within the network.

(f) In relation to the refrigerated goods losses and the 72-hour clause, there are two issues to be considered. The first concerns whether the refrigerated goods losses were “caused by” the flood within the meaning of the 72-hour clause. This question arises because the flood did not directly cause the refrigerated goods losses but rather triggered events which caused them. Given that the policy uses the broader expression “directly or indirectly caused by or resulting from” in other contexts, it is arguable that the losses were not “caused by” the flood. However, it is unnecessary that I determine that issue. This is because I consider that even if the losses were “caused by” the flood, in the context of this particular insurance policy, the 72-hour clause does not permit aggregation of losses across different insured locations.

Result


[67] Accordingly, I consider that the $150,000 special limit for refrigerated goods losses of $150,000 applies at each of Kalabo’s three affected supermarkets. As a consequence, Kalabo is entitled to $450,000 (less deductible), rather than $150,000, as argued by New India.


Does the same analysis apply in the case of the refrigerated goods damage after Cyclone Evan?


[68] Turning to the damage caused by Cyclone Evan, that was a category 4 tropical storm which caused widespread damage to infrastructure and property in the Northern and Western Divisions of Fiji. As a consequence, the Government declared a state of natural disaster in those Divisions.


[69] As indicated by the Deductibles provision, liability for cyclone damage falls within the policy’s coverage. However, it only applies to some of Kalabo’s supermarkets. The list of locations insured under the policy identifies which of the supermarkets are covered for cyclone damage and which are not. None of the supermarkets which suffered refrigerated goods losses from the power failures resulting from Cyclone Evan were covered for cyclone damage. This does not, of course, affect the refrigerated goods coverage because that applies where there is a breakdown of the refrigeration units “from any cause whatsoever”.


[70] The 72-hours clause’s reference to “All insured losses ... caused by ... other wind peril insured under this Insurance” includes insured losses from tropical cyclones. Here, however, the supermarkets affected by the failure of the power supply were not covered for cyclone damage, although their losses did fall within the refrigerated goods cover. In these circumstances, what is the position under the 72-hour clause in respect of the losses at the three supermarkets?


[71] This raises two of the issues discussed above in the context of floods – did Cyclone Evan “cause” the refrigerated goods losses given that it simply triggered events which ultimately led to those losses? does the clause permit the aggregation of losses across different insured locations? In my view, even if Cyclone Evan did “cause” the losses, the 72-hour clause does not permit the aggregation of losses across insured locations, for the reasons already discussed.


Result


[72] In the result, I consider that Kalabo is entitled to receive the full amount of the refrigerated goods losses at two of the supermarkets (ie, $82,284.55 and $51,644.54) and $150,000 for the refrigerated goods damage at the third, subject to application of the deductible.


The issue of interest


[73] Under 34 of the Insurance Law Reform Act 1996, New India is required to pay interest on the amounts payable under the policy “for the period commencing the day as from which it was unreasonable for the insurer to have withheld payment of the amount and ending on the day on which payment is made”.[17] The amount set in Reg 2(1) of the Insurance Law Reform (Interest Rates) Regulations 2004 for interest under s 34 is 10% per annum.


[74] As previously noted, New India paid Kalabo $300,000 (less deductible) and interest at 10% following the summary judgment proceedings. New India does not challenge that this was appropriate. Both the High Court and the Court of Appeal considered that interest at 10% on the outstanding balance was also appropriate but did not explain why it was unreasonable for the insurer not to have paid the additional amounts earlier. Interest on the unpaid balances was to run from 1 July 2012 in relation to the losses following the March 2012 flooding, where the unpaid balance was $300,000, and from 15 April 2013 for the losses following Cyclone Evan, where the unpaid balance was $132,429. Mr Gordon challenged the interest assessments, arguing that there was no evidence that New India had acted unreasonably in withholding payment.


[75] The position adopted in Australia in relation to an identical provision is that if a claim is incorrectly declined, the insurer will be liable to pay interest, even where it declined payment on a bona fide and reasonable basis. In LCA Marrickville Pty Ltd v Swiss Re International[18] Derrington and Colvin JJ said:


“What is required is a determination of the day on which a reasonable insurer would have paid out the claim on which the insured did succeed, assuming the insurer reached the factual conclusions ultimately found by the court and

otherwise adopted the correct view as to its legal position.”


[76] The underlying philosophy is that, whether understandable or not, the insurer has wrongfully declined liability, thus keeping the insured out of their money. The insurer will be given a reasonable period to investigate the claim, which will vary according to the nature and circumstances of the claim. But after that period for investigation has expired, the insurer’s obligation is to make payment of what is owed under the policy even if that has ultimately to be determined by a court. The fact that the insurer has an arguable but ultimately losing argument on liability or quantum will generally be irrelevant.


[77] In the present case, the policy at issue is that of the insured’s broker, rather than of the insurer, and it is a policy which is difficult to interpret. I have rejected Kalbo’s interpretation of the 72-hour clause as aggregating events, not losses, but have ultimately concluded that Kalabo is entitled under the policy to a greater level of compensation than New India offered. The interest award allowed a sufficient time for New India to investigate Kalabo’s claims. In those circumstances, I consider that the interest award of 10% must stand.


Disposition


[78] In the result I would:


(a) Grant New India leave to appeal on the questions set out in paragraphs 11(a) and (b) above;

(b) Dismiss New India’s appeal on both issues; and

(c) Order that New India pay Kalbo costs of $5,000 in relation to this appeal.

Jitoko, J


[79] I have had the advantage of reading in draft the judgment of Arnold J. I agree with his reasoning and conclusions and with the orders proposed.


[80] Orders
1. The Petitioner is granted leave to appeal against:


  1. The Court of Appeal’s finding that the refrigerated goods losses at the Respondent’s three supermarkets following the March 2012 floods and the refrigerated goods losses at three other supermarkets following Cyclone Evan in December 2012 were independent losses for the purposes of the “per loss” policy limit of $150,000; and
  2. The Court of Appeal’s award of interest of 10% on the outstanding sums under s 34 of the Insurance Law Reform Act 1996.
  3. The Petitioner’s appeal on both issues is dismissed and the orders of the Court of Appeal (as clarified in its decision of 17 December 2021) are affirmed.

3. The Petitioner must pay costs of $5,000.00 to the Respondent.


.........................................................
The Honourable Mr. Justice William Calanchini
Judge of the Supreme Court


............................................
The Honourable Mr. Justice Terence Arnold
Judge of the Supreme Court


...........................................
The Honourable Mr. Justice Filimone Jitoko

Judge of the Supreme Court



[1] There are two relevant policies – one applying to the March floods (the 2011 policy) and one applying to the December cyclone (the 2012 policy). Because there is no material difference between them, we will refer simply to “the policy” in the judgment.
[2] The amounts quoted in this judgment are in Fijian dollars.
[3] The figures referred to would be subject to the deductible of $1,500.
[4] There was also a claim for consequential damages, but I can put that to one side.
[5] New India Assurance Co Ltd v Punjas & Sons Ltd [2022] FJSC 49 at paras [34]-[49].
[6] Apparently, the placing slip was in evidence in the Courts below, but it alone is not sufficient.
[7] This wording is taken from the 2012 policy. As noted in fn 1 above, there is no material difference in wording from the 2011 policy.
[8] Star Entertainment Group Limited v Chubb Insurance Australia Ltd [2022] FCAFC 16, (2022) 400 ALR 25 at [6].
[9] See, for example, the discussion in Kelly & Ball Principles of Insurance Law at [8.0130.5] and following.
[10] AIG Europe Ltd v Woodman [2017] UKSC 18 at para [14].
[11] Financial Conduct Authority v Arch Insurance (UK) Ltd [2021] UKSC 1, at [323].
[12] Yorkshire Dale Steamship Co Ltd v Minister of War Transport [1942] AC 691 at 706.
[13] Financial Conduct Authority v Arch Insurance (UK) Ltd, above fn 11, per Lord Hamblen and Lord Leggatt (with whom Lord Reed agreed) at para [168].
[14] FEA stands for Fiji Electricity Authority, now Fiji Electricity Ltd.
[15] Generally, a reasonable human intervention of this sort will not break the chain of causation in these circumstances: see, for example, Financial Conduct Authority v Arch Insurance (UK) Ltd, above n11, per Lord Hamblen and Lord Leggatt (with whom Lord Reed agreed) at para [168]. Moreover, the insurance policy contained a provision requiring the insured to take prompt and reasonable steps to protect the insured property upon becoming aware of an event which was likely to give rise to a claim
[16] Either because the power supplier has shut the network down or because network infrastructure has suffered storm damage.
[17] Section 57 of the Insurance Contracts Act 1984 (Cth) is to the same effect. There is a very helpful discussion of the effect of that section in Kelly & Ball Principles of Insurance Law at [8.0170.05].
[18] LCA Marrickville Pty Ltd v Swiss Re International [2022] FCAFC 17; (2022) 401 ALR 204 at [248].


PacLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.paclii.org/fj/cases/FJSC/2023/30.html