Terrorism Where are we now? Insurance and Reinsurance Forum

Transcription

1 Terrorism Where are we now? Insurance and Reinsurance Forum Wednesday, 7 July 2004 John K Morgan, Partner Malcolm Stephens, Senior Associate 1. Introduction In this forum we intend to: give an overview as to where the legislative changes introduced by the Terrorism Insurance Act 2003 (the Act) are today; discuss broadly the way in which the Act may be expected to operate in the event of a terrorist incident and the terrorism exclusions currently in the marketplace; look more closely at one relatively unexplored area of the Act and that is the application of the legislation to liability policies; and and then to ask Mike Pennell of the Australian Reinsurance Pool Corporation (ARPC) to comment on some practical issues relating to the application of the Act particularly in the event that there is a terrorist incident. 2. Overview I thought firstly we should revisit the essential structure of the legislation to remind you how it works, to identify the issues that appear to have received most consideration to date and then to move to more fully discuss how the Act will operate if there is a terrorist incident. As most of you are aware, the Act applies to eligible insurance contracts which are defined in sub-section 7(1) to be contracts that provide insurance cover in relation to: loss of, or damage to eligible property that is owned by the insured; business interruption and consequential loss arising from loss of, or damage to or inability to use eligible property that is owned or occupied by the insured; and liability of the insured that arises out of the insured being the owner or occupier of eligible property. Eligible property is property located in Australia but the only property which is actually caught within the definition is: buildings (including fixtures) or other structures or works on, or under land; tangible property that is located in or on property to which the first limb applies; and jkms S v Page 1

2 any other property described by the regulations which to date have only extended the definition to tangible property that is on, in or under the seabed. Reinsurance as opposed to insurance is not covered and the Regulations operate to exclude various types of eligible insurance contract. There have been a number of problems in relation to the application of the Regulations. Sub-section 7(2) says that a contract covered by sub-section 7(1) is not an eligible insurance contract to the extent to which it is prescribed by the regulations. The regulations take various forms including: in some cases they cover insurance which would not caught in any case under the core definition and are introduced for the purposes of creating certainty; some regulations appear to exclude particular types of contracts of insurance in totality so the characterisation of the whole contract is important; some regulations expressly exclude insurance only to the extent that the policy provides certain types of cover; and other regulations exclude insurance providing cover to the Commonwealth and to certain State or Territory bodies. This had led to a number of difficulties in applying the provisions to particular types of policy. The correct characterisation of a policy or part of it as an eligible contract is important for the purposes of the reinsurance contract with ARPC. But that reinsurance contract significantly reduces the risk of an insurer making an incorrect decision. Under clause 2(b) of the standard reinsurance contract, if an insurer makes an error because it was not obvious or apparent that the insurance was an eligible insurance contract then the insurer is still reinsured so long as the unpaid reinsurance premium is paid. However, if there is a declared terrorist incident, the issue of what is and what is not an eligible insurance contract will not be able to be avoided. Insureds who have suffered loss will have a strong incentive to succeed in arguing that their insurance is an eligible insurance contract and that any terrorism exclusion is written back by the Act. I do not intend today to revisit all of various issues that have arisen in relation to the application of the regulations except to remind everyone that the first step in determining whether the Act applies is to consider whether or not the contract of insurance is an eligible insurance contract under the core definition in sub-section 7(1) and only after having determined that should you move to the Regulations to determine whether the contract is excluded or partly excluded. There has been a terrible tendency to just jump into the Regulations and assume if no regulation applies then the contract of insurance is an eligible contract of insurance. The Act operates by writing back terrorism exclusions. This write back occurs in the event that there is a declared terrorist incident and as a consequence the policy of insurance will respond as if the terrorist exclusion was not in the original policy. In other words, it will continue to be subject to the ordinary limits and terms that the policy would otherwise have and as a consequence even though the terrorism exclusion may be written back the policy may or may not still respond to the particular loss that arises. jkms S v Page 2

3 A very good example of this is a contract that has both a terrorist exclusion and separately and distinctly a general exclusion for biological or chemical loss or a nuclear exclusion. Notwithstanding that there is a declared terrorist incident that has resulted in a terrorist exclusion being removed, if the loss is caused by a biological or chemical act or comes within the nuclear exclusion, then the policy will still not respond to losses excluded by the exclusion. If the policy does respond after removal of the terrorist exclusion, then the insurer is liable but the insurer may have access to the reinsurance arrangements provided by the ARPC. If those arrangements have been put in place with respect to the relevant policies then the insurer will have a retention determined in accordance with the legislation and the reinsurance contract. Thereafter the claims will be met by the ARPC. The ARPC or more particularly, the Minister has the power to determine a reduction percentage. The reduction percentage is likely to arise if the event is one where the Commonwealth will be exposed to a liability of more than $10 billion Australian dollars. If the insurer has been reinsured with ARPC then the loss arising from the imposition of a reduction percentage does not fall on the insurer but rather it falls on the insured. The reduction percentage mechanism does not apply where the insurer has chosen not to insure with ARPC and in that event the insurer will be exposed to whatever liability it has under the policy on the basis that a terrorism exclusion has been excised. Having provided that short introduction I would now like to introduce Malcolm Stephens, a Senior Associate of the firm. What Malcolm will consider are the problems that will actually arise if there is a terrorist incident in Australia and how it may be expected that the Act will actually operate in those circumstances. Following that I will then return to make some short comments in relation to liability insurances. 3. The effect of the Act Section 8 of the Act deals with the effect of terrorism exclusions in eligible insurance contracts. The key provision is that a terrorism exclusion in an eligible insurance contract has no effect in relation to a loss or liability to the extent to which the loss or liability is an eligible terrorism loss. That then calls for consideration of two concepts - what is an eligible terrorism loss and what is a terrorism exclusion. 4. Eligible terrorism losses The only loss recoverable under the Act is an eligible terrorism loss. That term is defined as follows: Eligible terrorism loss means a loss or liability arising from a declared terrorist incident, but does not include a loss or liability arising from the hazardous properties (including radioactive, toxic or explosive properties) of nuclear fuel, nuclear material or nuclear waste. jkms S v Page 3

4 Of course, the eligible terrorism loss will be further constrained by the actual terms and conditions and limits in the policy. The important point to note here is notwithstanding that the policy may not have a nuclear exclusion or the nuclear exclusion may be relatively narrow, no losses of the type described in the definition which are within the nuclear exclusion in this definition will be payable. 5. Terrorism Exclusion Sub-section 8(2) defines a terrorism exclusion in the following words: terrorism exclusion means an exclusion or exception (however described) for: (a) (b) acts that are described using the word "terrorism" or "terrorist" or words of similar effect; or other acts (however described) that are substantially similar to terrorist acts as defined in section 5. Therefore, the test will depend on the use of terminology that is the words "terrorism" or "terrorist" in the exclusion or if those words are not found, the similarity of the exclusion to the matters defined as terrorist acts in the legislation. This in turn will lead to some interesting issues relating to certain types of exclusion. To take a simple example, if an exclusion related to losses arising from acts done with the intention of advancing a political cause and with the intention of coercing government, then such an exclusion will clearly fall within the bounds of section 5 and will be a terrorist exclusion although it does not use the relevant terminology. More difficult questions may arise in relation to exclusions of that sort which relate to industrial action or the like. However, there would be many exclusions which do not use terrorism language and do not fall clearly within the characteristics set out in section 5. These exclusions might include the following: riot; nuclear risk; biological and chemical risk, particularly where those exclusions do not, in their terms, refer to terrorism or a terrorist activity and do not pick up any of the other key elements of the definition of "terrorist act" found in section 5. The analysis of the exclusion by reference to section 5 will also need to have regard to the inclusions and the exclusions which we have described above. For example, if the exclusion does not use the term "terrorism" or "terrorist" and relates solely to a matter excluded from the meaning of terrorist act under section 5, then it could not be a terrorism exclusion for the purposes of the legislation. We should also note that a terrorist exclusion may be not an exclusion but an "exception". This raises an interesting issue because it may be suggested that an insurer could avoid the application of the Act by imposing a sub-limit of liability of say $1 in respect of losses jkms S v Page 4

5 arising from terrorism. In our view, although this may not be an exclusion, it is an exception and as such will be similarly ousted by the operation of the legislation. Therefore, as a consequence of the declaration of a terrorist incident, the Act would operate to remove the sub-limit of liability that related to terrorism. Many of the standard definitions use terms such as terrorism or terrorist in describing the acts which are excluded. Once the exclusion is determined to be in those terms, then that exclusion will be classified as a terrorism exclusion. But if that is not the case, then further investigation is needed as we have indicated to compare the exclusion to section 5. Another issue that arises is the degree to which the exclusions in the policy are either broader than section 5 or narrower than section 5. At the end of the day, this does not matter insofar as the legislation is concerned because, if the exclusion is characterised as a terrorism exclusion, then it is an exclusion to which the Act applies. However, the broadness or narrowness of the exclusion may impact on the operation of the policy once the declared terrorist event has occurred. For example, if an exclusion was broken into two components, one being the terrorism exclusion strictly limited to matters of the type covered by the Act and a further exclusion was then introduced into the policy, then that further exclusion may not be a terrorist exclusion and would continue to operate notwithstanding the application of the Act. As we have already indicated, this could occur for example where a terrorism exclusion was reasonably narrowly drawn and there was a separate exclusion for losses arising from a biological or chemical event. 6. The Terrorism Insurance Act: Is there a "declared terrorist incident" If the Minister, after consulting the Attorney General is satisfied that a terrorist act has happened in Australia (or two or more related acts have happened in Australia) then the Minister must declare the act or the acts together constitute a declared terrorist incident. Acts of war are excluded. Of course this raises an interesting issue in that if there are two acts of terrorism which are treated by the Minister as a single declared terrorist incident. It would follow that, for the purposes of the insurer's retention under the reinsurance arrangements and the application of any reduction percentage, the two acts would be treated as one event. However, that would not necessarily mean that for the purposes of the application of the underlying policy that there would be one or two events. That will still be determined in accordance with the terms of the underlying policy and, for example, there may be one declared terrorist incident but because of the way in which the policy defines a claim, there may be two deductibles. Section 6(3) specifies that a threat of action is only taken into account if the Minister is satisfied that the threat resulted in economic loss to a person. In other words, the loss must have occurred. Given that section 7(1)(a) and (b)(i) only makes an insurance an eligible insurance contract if there is damage to eligible property, a threat would only be relevant where: jkms S v Page 5

6 There was a business interruption and consequential loss arising from an inability to use eligible property that would not be covered because of a terrorist exclusion; or Less likely, a liability of the insured covered by the policy that arose because of the threat. We have already mentioned the reduction percentage the percentage of the total of all losses which will not be covered by the reinsurance. Once specified, it cannot be increased but can be reduced. This suggests that the Minister will need to: have a good feel for the size of loss; act conservatively when first fixing a reduction percentage. After that, adverse loss development is at the risk of the Commonwealth. Division 6 of the Act deals with Finances. Section 35 provides a guarantee by the Commonwealth and ARPC and section 37 provides the appropriation from the Consolidated Revenue Fund of moneys borrowed by the Corporation from the Commonwealth and moneys payable under the guarantee. The management of ARPC are vested in the Chief Executive subject to the direction of and in accordance with the policies of the Board. The Corporation is subject to Ministerial direction under section 38. These directions relate to payments to the Commonwealth, borrowings, premiums and insurer retentions. Importantly, in the case of a declared terrorist incident, the Minister has no power to direct the Corporation as to how to pay claims or what claims to admit or deny. For there to be a declared terrorist incident, there must be a terrorist act. A terrorist act is defined in section 5 of the Act. 7. Definition of terrorist act The definition of "terrorist act" in the Terrorism Insurance Act is partly an inclusive definition and partly an exclusive definition. 7.1 The Inclusive Definition The inclusive definition looks at both the intention of the act and its effect. In relation to the intention of the perpetrator, an act will only be a terrorist act if it is made with the intention of: (a) (b) advancing a political, religious or ideological cause; and either: (i) (ii) intimidating a section of the public; or coercing or intimidating a government. jkms S v Page 6

7 For reasons explained above, the requirement that a terrorist act be intended to advance a political, religious or ideological cause is inconsistent with the common understanding of what constitutes a terrorist act. The definition does not, for example, apply to acts motivated by hatred of a particular group, rather than the intention of advancing one of the specified causes. A further, practical problem is that the intention of an act will often not be known unless and until the perpetrator is found. If a large bomb exploded in Sydney, without the perpetrators being caught or there being any verifiable claim of responsibility, the act could never be a declared terrorist incident. There are also difficulties with the requirement that there be an intention of intimidating a section of the public or influencing a government. This would generally exclude, for example, revenge terrorist attacks. The definition of "Terrorist Act" also looks to the effect of the act, as well as the intention behind it. It includes any act that causes serious damage to property or serious harm (including death) to a person. It also includes acts which create a serious risk to the health of safety of a section of the public or which interfere with an electronic system. There is no requirement that these consequences be intended 1. There is also no requirement that the act cause "terror", which would normally be regarded as a basic requirement for there to be a "terrorist act". 7.2 The exclusive definition Section 5(3) of the Terrorism Insurance Act provides that an act will not be a "terrorist act" if: (a) (b) it is advocacy, protest, dissent or industrial action; and it is not intended to cause serious harm or to create a serious risk to the health or safety of a section of the public. One possible concern about the exclusive part of the definition is the qualification for creating a serious risk to the health or safety of a section of the public. There would appear to be a risk that industrial action by people working in the health care industry or in the emergency services could, as a result of this qualification, find themselves being deemed to have committed terrorist acts The Scope of Standard Terrorism Exclusions Whether or not a particular act falls within a tourism exclusion is very important for at least the following two reasons: An insurer will not be entitled to recover from the ARPC in respect of a declared terrorist incident if that incident falls outside the wording of its terrorism exclusion; 1 Such acts may, however, be excluded by the exclusive part of the definition in section 5(3) of the Terrorism Insurance Act. 2 There will clearly be room to argue that whether industrial action for done for a "political" cause and whether it was intended to "coerce" or "influence by intimidation". jkms S v Page 7

8 an insured might suffer a loss as a result of an act which is not a declared terrorist incident but falls within an insurer's terrorism exclusion. Whether or not a terrorism exclusion applies will depend on the particular wording of the exclusion. Many standard terrorism exclusions used in Australia will apply if there is an act committed: by an individual or an organisation; for a political, ideological or religious purpose; and which may, but not necessarily, be intended to influence a government. There are many difficulties with the exclusions commonly used in the Australian market. These include the following. (a) (b) (c) (d) The exclusions focus on the purpose of the act, rather than its consequence. For example, setting off a bomb in a cafe will only be a terrorist act if it is committed for one of the specified purposes. If a bomb is set off purely for the purpose of "terrorising", then it will not be terrorist act. This would appear to conflict with the common (and dictionary) understanding of a terrorist act, being an act which inspires terror. The possible requirement that an act be committed alone or by an organisation also appears inconsistent with the common understanding of what constitutes a terrorist act. For example, if a person acting alone or as part of an organisation explodes a bomb, then that may be a terrorist act. If two people who do not belong to a organisation cause the bomb to go off, then that might not be a terrorist act. It is difficult to understand the rationale behind this distinction. Many definitions of terrorism include acts which inspire terror, but are not limited to such acts. By the liberal use of the words "including" or "not limited to", there are a number of exclusions in the Australian market which effectively define any political, religious or ideological act as a terrorist act (perhaps with the qualification that it be committed alone or by an organisation). The effect of a number of terrorism exclusions, therefore, is that attending church alone is a terrorist act 3. Wordings currently used in the Australian market do not normally refer expressly to "declared terrorist incidents". This is surprising, as it is clearly in the interests of insurers that, having paid premiums to the ARPC, they do not miss out on the reinsurance cover because their terrorism exclusion does not apply. The ARPC is not liable under its reinsurance contract unless the insurer is entitled to deny liability by relying on a terrorism exclusion. In summary, therefore, the terrorist exclusions commonly be found in the Australian market: cover many acts which would not normally be regarded as terrorists acts; and fail to cover many acts which would normally be regarded as terrorists act. 3 In practice, of course, a terrorism exclusion could not in fact be used to exclude attending a church. jkms S v Page 8

9 It is very difficult to propose an adequate definition of the word "terrorism". It may therefore be a word that is better left undefined. Although each person may have a different opinion as to precisely what acts constitute "terrorism", these differences will generally be smaller than the difference between the common understanding of "terrorism" and the definitions that are used in the market. 9. Summary of issues relating to terrorist exclusion wordings 9.1 Insureds Insureds face a real risk of having a terrorism exclusion apply to an event which, in ordinary parlance, would never be considered a terrorist act. In practice, insureds will often have limited room to negotiate as underwriters may not have any discretion to vary these wordings (because, for example, of the wording of their reinsurance policies). In so far as there is room to negotiate, however, insureds might seek to have any terrorism exclusion limited to declared terrorist incidents. This will maximise the prospect of the insured being entitled to recover (as a result of the Terrorism Insurance Act) but, assuming the insurer reinsures with the ARPC, will involve a comparatively small concession on the part of the insurer. 9.2 Insurers As shown in the following flow chart, the worst outcome for an insurer would be: to pay premiums for reinsurance cover with the ARPC; for their to be a declared terrorist incident; but for its own terrorist exclusion not to apply. Any terrorism exclusion put forward by an insurer should therefore include, at least, declared terrorist incidents. Insurers also need to be careful that, in providing for exclusive definitions 4 of terrorism, they do not inadvertently fail to include acts which would, in the absence of such a definition, be regarded as terrorist acts. For example, the common focus of definitions on the purpose of the act, rather than its effect, sits uncomfortably with the commonly accepted view that a terrorist act is an act which inspires terror. 9.3 ARPC The Terrorism Insurance Act does not give the government any discretion in determining whether or not there has been a terrorist act. Court action might therefore compel the declaration of a declared terrorist incident in circumstances where the conduct would not normally be regarded as a terrorist act. 4 A definition which begins: "terrorist act means " rather than "terrorist act includes, without limitation, ". jkms S v Page 9

10 10. Flow Chart The diagram below summarises the outcomes, described in the previous sections, depending on whether or not an act is a terrorist act or a declared terrorist incident 5. Incident Primary Policy: Does a terrorism exclusion apply? No Yes Insurer Liable Was the loss caused by a "declared terrorist incident? ARPC not liable No Yes Insurer bears loss Insurer not liable Insurer Liable * Insured bears loss ARPC Liable * ARPC bears loss * * Subject to Reduction Percentage 5 This flow chart does not take into account issues such as a reinsured's retention, the possibility of a "reduction percentage", or the consequence of delays in declaring a "terrorist incident" (or such a declaration subsequently being overturned or compelled to be made). jkms S v Page 10

11 11. Liability Insurance Under paragraph 7(1)(c) of the Act, a contract of insurance is an eligible insurance contract to the extent that it provides insurance cover for liability of the insured that arises out of the insured being the owner or occupier of eligible property. The regulations specifically exclude contracts of professional indemnity insurance (Item 10 of Schedule 1), product liability insurance to the extent that it does not provide cover for public liability (Item 16), certain aviation liability covers (Item 23), directors and officers liability policies, employment practices policies and trustee liability policies (Items 28, 29 and 30) or compulsory classes of liability insurance (Item 18 and 22). It may be that a liability policy will fall within one of the other provisions within the regulations. Importantly a contract of insurance that provides cover only for loss or liability arising from terrorism is also excluded (Item 38) although that contract may cover certain types of terrorism loss (eg. fire) but not others. However this does not address the important question which is, what circumstances, such as under a public liability policy, can it be said that the "liability of the insured arises out of the insured being the owner or occupier of eligible property." The first thing to note that eligible property is defined in the manner which I have outlined earlier, that is, it only relates to certain items in or under land or in or under the sea bed and other tangible property located in or on that property and the property must be located in Australia. Also, the terrorist act might be in Australia. The term "own" in relation to eligible property is defined to include: owning the property together with another person or other persons; and having an insurable interest in the property. If Australian law is the proper law of the contract then the insurable interest will include economic and financial interests in the property. This would include, for example, a lessee or mortgagee. The term occupier is not defined but is used in conjunction with the term owner and to the extent that an owner is also an occupier then no issue arises. What is more important is who is an occupier where that person is not within the definition of ownership. Clearly, it may include a right which is less than ownership which is created by law such as the creation of the rights of a lessee under a lease or it may arise from actual possession of the particular property. However, most people with these types of rights will have an insurable interest and the extension by use of the term "occupier" is not needed for them. The type of issue that will arise, for example, is whether a property manager is an occupier of the relevant property. And if that person is an occupier what part of the property is the manager an occupier of or if a property owner engages a security firm to provide security services in respect of the property under a contact for that purpose and the security personnel are given access to the security centre within the property and have rights to jkms S v Page 11

12 allow entry into or to eject persons from the property does that make that person an occupier. At the end of the day these types of questions can only be answered by looking closely at the particular agreements that have been entered into, the rights that are created under those agreements and the particular facts that arise and the degree to which the various people who may fall into these categories can be said to be in occupation as opposed to being merely present within the building performing a particular service on behalf of the owner. The matter will not be easy to determine and there is no simple answer which will apply to all situations. There is a high level of uncertainty and so the problems of quantifying the likely loss from the incident may be significant. This could directly impact the Commonwealth in relation to the reduction percentage. Some assistance can be obtained from case law to give guidance as to the approach courts may take to the question of what is occupation but care needs to be taken in relation to those cases because often they are decided within the context of particular situations and within a particular legislative framework. For example, under the Pastures Protection Act 1902 of New South Wales, occupier was defined as being a person entitled to a possession of a holding or land and included a resident manager of the occupier where the occupier does not reside on the holding or land. Palmer v Crisp (1907) 7 CLR 612 concerned a boundary rider charged with the duty of looking after sheep depasturing in a paddock. It was alleged that he was the occupier. No one lived in the paddock and the boundary rider was charged with failing to take appropriate steps to reduce the rabbit population. The boundary rider was responsible for the sheep upon the land and he had keys of the only paddock which was locked. He boarded at a house on adjoining land and daily went on to the paddock to perform his tasks. The boundary rider obtained his instructions from the company who leased the land and he placed poisons when directed to do so. The Magistrate refused to convict the boundary rider and that view was upheld by the Supreme Court of New South Wales. The High Court refused leave to appeal. What this case demonstrates is the point that I have made about the need to look carefully at the interlocking relationships of the parties in order to determine whether a person is an occupier. The types of criteria which the courts seemed to have looked to are as follows: the taking of possession; the degree of control; whether or not others are in occupation; the right to exclude others; the degree to which the occupation is transient; even where there is no legal possession, the degree of actual physical possession. A particularly interesting case is that of Gas and Fuel Corporation of Victoria v City of Williamstown (1978) VR677 which concerned the imposition of rates under the Local Government Act. The issue related to gas mains and whether or not the owner of those jkms S v Page 12

13 mains was subject to Council rates as an occupier. In the context of the Terrorism Act, the mains would be property under land and therefore eligible property. In the context of the Local Government legislation under consideration, the court held that the gas authority was an occupier of the gas mains in that it passed gas through those mains and by doing so occupied them. So if a pipeline is owned by A, then it may be that B who owns the liquid or gas passing through that pipeline is an occupier of the pipeline for the purposes of the legislative test. It is then another matter as to whether or not they have incurred a liability that arises because of that occupation. The next point that needs to be made is that the relevant owner/occupier must be an insured under the relevant insurance policy. Clearly this will include a named insured and probably also extends to other persons referred to in the policy but there may be issues as to whether or not for the purposes of this Act an insured includes a person whose entitlement to claim merely arises under Section 48 of the Insurance Contracts Act The next issue that we need to confront is when can it be said that the liability arises out of the insured being the owner or occupier. The requirement involves a consideration of causation but it also needs a careful consideration of what it is that is relevant cause. It is not merely the fact of ownership or occupation but rather, that it has to arise out of "the insured being the owner or occupier". The words "arising out of" and similar terms tend to be interpreted as imposing a simpler and less stringent test of causation than other terms such as "caused by" or "proximately caused by". In other words, the issue is whether, as a matter of common sense and practicality, can it be said that the insured's ownership and occupation was the cause of the liability. In the context here, clearly courts will have regard to the public policy and purpose of the legislation and given that the purpose is to protect insureds in the event of terrorism, where there has been a declared terrorist act, there must be a likelihood that courts will give a generous interpretation to this provision. Clearly, in circumstances where a person was injured following a terrorist act because exit doors were blocked or fire escape was blocked, a very strong argument can be made that that is a liability that arises out of the insured being an owner or occupier. However, more difficult questions arise where a person is injured by a terrorist act or property is lost o damaged by a terrorist act and an argument is run that the loss, damage or injury would not have occurred or would have been significantly less if the owner or occupier had taken certain precautions and had failed to do so. The question will then be, did the failure to take the precautions involve a breach of duty and did that breach arise out of the insured being the owner or occupier. Let us say, for example, that given the nature of threats which were made and the known risks, a person should have provided barriers or security arrangements to reduce the likelihood of terrorist attack, can it then be said that that failure arises out of the insured's ownership or occupation. I think that a reasonably good argument may be able to be made in those circumstances that if there is such a breach of duty and a breach of the relevant jkms S v Page 13

14 Conclusion standard of care, then that breach may be one which arises out of the ownership or occupation if it is the owner or occupier who actually had the duty which was breached. There is a degree of circularity here but it seems likely if an owner or occupier has a duty which is breached, then the resultant liability is very likely to arise out of the insured ownership or occupation. We therefore conclude that there are a number of potential of significant issues that will arise on the occurrence of a terrorist act. They range from whether or not it is a declared terrorist incident, how the declaration is made, the manner in which the Commonwealth imposes or does not impose a reduction percentage and more particularly, the application of policies which are eligible contracts of insurance in the context of the actual losses that arise. In many instances, this will require a very close examination of the underlying facts and circumstances and there is significant potential for dispute. Once the insurer's retentions have been eroded fully, the issue is going to be fundamentally between ARPC and the claimants under the policies and will need to give close consideration as to how it administers those claims as reinsurer. It will have an interest in determining: whether the loss is one which is recoverable under the Act and the policy; more particularly, whether there are other exclusions or limits in the policy which will limit or exclude the loss. There are obviously going to be very difficult questions that may arise under this provision and there is a high level of uncertainty as to what liability is covered or not covered under it. jkms S v Page 14