Income tax: captive insurance companies - deductibility of premiums and the appropriate basis of assessment

This Ruling explains: (a) what will constitute a captive insurance company for income tax purposes; (b) whether premiums paid by a company to a captive insurance company will be deductible for income tax purposes; (c) what the appropriate basis of assessment is for a non-resident captive insurance company in respect of premium income derived; (d) in what circumstances the Commissioner is likely to apply the anti-avoidance provisions of the income tax law to deny a deduction for the premium paid to captive insurer; and (e) the application of Division 13 to circumstances where the parties are not dealing with each other at arm's length.

For taxation purposes a captive insurance company is a company that is either directly or indirectly controlled by an Australian resident. Such a company will be a closely held company whose insurance business is principally that of providing indemnity for insurance risks of the parent and/or associated companies.

For an insurance premium to be deductible it is essential that the risk of loss from the occurrence of contingent events is transferred from the insured to the insurer. That is, where the arrangement does not transfer the risk of loss from the insured to an insurer, the premiums paid to the insurer are not deductible as an insurance expense. The transfer of risk however, is not the only factor which determines the deductibility of insurance premiums. The following factors will also need to exist before it will be accepted that the premium has been incurred as part of an acceptable insurance arrangement for taxation purposes: • there is a risk; • there is a transfer of that risk; • there is a distribution of that risk; • the premium must be commercially realistic given the risk insured; and • there must be a promise from the insured to indemnify the insurer.

Division 15 of Part III of the Income Tax Assessment Act 1936 (ITAA) governs the assessment of non-resident insurers. Where an insurer is a captive of an Australian resident, the captive's profit or loss - in relation to premiums deemed to have an Australian source by virtue of section 142 - are to be calculated, subject to the provisions of the Act, by reference to receipts and expenditure taken into account in calculating the captive's profit or loss.

Where a non-resident insurer is not a captive of an Australian resident and the Australian resident insured may not have sufficient access to the insurer's records to enable it to calculate the insurer's income, section 143 will apply to deem the insurer to have derived a taxable income equal to 10 per cent of such premiums that are deemed to have an Australian source. Section 143, however, only applies to genuine insurance arrangements.

The Commissioner will apply Part IVA to deny a deduction to the company paying a premium to its captive insurer where it can be concluded that, having regard to the available evidence, the dominant purpose of entering into the arrangement with the captive insurer was to provide a tax benefit.

Under Division 13, the Commissioner will deem an arm's length consideration for an insurance premium where: • the insured and the captive are not dealing with each other at arm's length in relation to the insurance arrangement; • the premium paid by the insured exceeds an arm's length consideration; or • it is not possible or practicable for the Commissioner to ascertain the arms length consideration.

This Ruling applies to years commencing both before and after its date of issue. However, the Ruling does not apply to taxpayers to the extent that it conflicts with the terms of a settlement of a dispute agreed to before the date of issue of the Ruling (see paragraphs 21 and 22 of Taxation Ruling TR 92/20).

A captive insurance company is a wholly-owned insurance company subsidiary established to underwrite all or a selection of the risks of the parent and its associates.

Captives need not be owned by a single parent but may be owned by a number of entities. Captives can be a financially efficient method of allowing a number of business enterprises, organisations or professional people, facing a common Risk Financing problem, to co-operate in a common solution. The most common feature of a captive insurance company is that it enters into contracts, or purported contracts, of insurance.

In our view a captive insurance company will include a company established to underwrite risks of the parent and its associates an which satisfies any of the following conditions: • there is a group of five or fewer Australian entities, each with an associate-inclusive control interest (as defined in Part X of the Act) in the company of at least one per cent, whose aggregate associate-inclusive control interests are not less than 50 per cent; or • there is a single Australian entity whose direct and indirect control interests and whose associate's or associates' direct and indirect control interests are not less than forty per cent, unless the Commissioner is satisfied that control of the company is in the hands of other entities and their associates; or • there is a group of five or fewer Australian entities, either alone or with associates, that may or may not be Australian entities, that in any manner could control the company.

A deduction for an insurance premium will only be allowable if the premium is a loss or outgoing incurred in gaining or producing assessable income or is necessarily incurred in carrying on a business (subsection 51(1) of the ITAA). Whether the outgoing is, for the purposes of subsection 51(1), wholly or partly "incurred in gaining or producing assessable income" is a question of characterisation (see Fletcher & Ors v. FC of T 91 ATC 4950 at 4957; (1991) 22 ATR 613). To characterise an outgoing as having been incurred in gaining or producing assessable income, the outgoing must be incidental and relevant to the derivation of the assessable income (see Ronpibon Tin N.L. and Tongkah Compound N.L. v. Federal Commissioner of Taxation (1949) 78 CLR 47 at p. 56).

For the premium to be incidental and relevant to the derivation of assessable income, it will be necessary for the premium to be paid in respect of an acceptable insurance arrangement. An example of an unacceptable insurance arrangement would be one involving financial insurance. Financial insurance is to be the subject of a separate Ruling on the taxation implications of arrangements known as financial insurance and financial reinsurance.

Insurance is about the spreading of risk based on the 'law of large numbers' - the many paying for the few. A statute of Queen Elizabeth, dated 1601, contained in its preamble a classic definition ofinsurance. The definition is as follows: "By means of a Policy of insurance it cometh to pass that upon the loss or perishing of any ship there followeth, not the undoing of any man, but the loss lighteth rather easily upon many than heavily upon few." We understand that this is the first time that the subject of insurance is referred to in English Law Books, and these few simple words contain the fundamental principles of insurance. Insurance enables insurers to spread the potential loss from a few individuals over many other individuals. Insurance thus involves the transfer of the potential loss from an individual - who may be subject to the loss as a result of the occurrence of an adverse event - to an insurance company.

In Prudential Insurance Company v. Inland Revenue Commissioners (1904) 2 KB 658, the Court said that the following requirements must exist for there to be a contract of insurance: • it must be a contract for some consideration whereby you secure to yourself some benefit upon the happening of some event; • the event should be one which involves some amount of uncertainty, either whether the event will happen at all or as to the time at which it will happen; and • it must be against something which is adverse to the interest of the assured.

These requirements can be summarised in a definition of insurance: "Under a contract of insurance one party, known as the insurer, promises that on the occurrence of an uncertain specified event he will either indemnify the other party, known as the insured or the policy holder, for any financial loss he may sustain, or pay to him a certain sum, and in return the insured agrees to pay the insurer an ascertainable amount known as a premium."

The definition of insurance has also been considered by the United States Courts which, in the decision of Helvering v Le Gierse 312 U.S. 531 (1941), considered that insurance involves risk shifting and risk-distributing, although these terms were not defined.

The main purposes of an insurance arrangement, therefore, is to transfer the risk of loss that may arise from the insured's interest in the subject matter of the insurance to the insurer. Individuals, taking out motor vehicle insurance, for example, transfer the risk of experiencing a loss were an accident to happen, to an insurance company through an insurance policy. Under the insurance policy the insurance company undertakes to indemnify the insured person against such a loss. The consideration for that indemnity is the premium paid by the insured to the insurance company.

The transfer of the risk of loss from the insured to the insurer then exposes the insurer to the possibility of incurring a significant loss under a particular insurance contract.

The insurer, by accepting other policies which are not expected (on the basis of probabilities) to incur a loss, has effectively distributed the risk of loss amongst all the insured parties. The premiums from those parties that do not experience a loss are used to pay for the loss experience of the few. This is the basic concept of the 'law of large numbers' where the probability of insured events occurring is even among all insureds. The greater the number of insureds, the more the risk can be shared (given reasonable loss probabilities).

An essential element that must exist for there to be a contract of insurance is a risk. There must be a hazard which, in the event that it occurs, will cause loss or detriment to the insured.

The U.S. Tax Courts have considered the presence of insurance risk and have held the following: Basic to any insurance transaction must be risk. An insured faces some hazard; an insurer accepts a premium and agrees to perform some act if or when the loss event occurs. If no risk exists, then insurance cannot be present. "Insurance risk" is required; investment risk is insufficient. If parties structure an apparent insurance transaction so as to effectively eliminate the effect of insurance risk therein, insurance cannot be present (see Ocean Drilling & Exploration Co v U.S. 93-1 USTC)

Insurance risk can be defined as the risk arising from uncertainties about both: • the ultimate amount of net cash flows from premiums, commissions, claims and claim settlement expenses paid or incurred under a contract ( underwriting risk ); and • the timing of the receipt or payment of those cash flows ( timing risk ). • 24 Other forms of risk include investment risk, credit risk, and expense risk.

Investment risk is the risk that investment earnings will fall short of projected investment earnings. Investment risk is affected by timing risk as well as market fluctuations.

Credit risk includes the risk that (a) the insured may not pay premiums when due, (b) subrogation rights may not be enforceable or (c) a reinsurer may be unable to pay amounts due under a reinsurance arrangement.

Expense risk is the risk that acquisition and operating expenses may exceed amounts expected when the insurance premium is calculated. Expense risk is primarily a problem of pricing the product.

Arrangements which attempt to cloak the real underlying nature of an arrangement to give the appearance of the existence of insurance risk, where in fact none is present, will not be accepted as valid insurance arrangements for taxation purposes.

A transfer of risk will have occurred if one party has successfully transferred the risk of loss to another. This will happen when policies are written, premiums are paid by the insured to the insurer and the insurer has the capacity to pay the sum insured under policies in the event of claims being made.

In return for the acceptance of a risk, consideration in the form of money (known as a premium) must be paid by the insured to the insurer. The insurer must also be under an obligation to pay a sum of money or its equivalent upon the happening of the event insured. The insured must have a legal right to payment which cannot be at the insurer's discretion (Medical Defence Union Ltd v. Department of Trade (1979) 2 All ER 421 ; Oswald v. Bailey & Ors (1987) 4 ANZ Insurance Cases).

In our view, a transfer of risk will have occurred where the contract between the insured and the captive insurer relieves the insured from suffering an economic loss upon the happening of the insured event. This was stated in the U.S. case of Clougherty Packing Co v Commissioner [87-1 USTC 9204]: 'if the insured has shifted its risk to the insurer, then a loss by or a claim against the insured does not affect it because the loss is offset by the proceeds of an insurance payment'

For a transfer of risk to have occurred from the insured to its captive the captive must be in an economically viable position which will enable it to meet any losses that arise. A captive insurer's ability to pay may be influenced by the extent to which it has reinsured its risk of loss.

In our view a transfer of risk will only have occurred where the contract between the insured and the captive insurer relieves the insured suffering an economic loss upon the happening of the insured event (see paragraphs 43-47).

Accordingly, where there has not been a transfer of risk, the premium will not form part of an acceptable insurance arrangement and a deduction for the premium will be denied under subsection 51(1) of the ITAA.

In the U.S. case of Sears, Roebuck & Co v Commissioner 96 TC 61 (1991), the court found that risk shifting had occurred in both form and substance for the following reasons: • insurance contracts had been written; • premiums had been transferred; • losses had been paid; • the subsidiary insurer had been a separate, viable entity which was financially capable of meeting its obligations; and • the relationship between the subsidiary insurer and its parent was the same as the relationship it had with unrelated insured parties.

Examples of arrangements that have transferred risk are contained at appendices B and C. An example of an arrangement that has not transferred risk is contained at appendix F.

Risk distribution occurs when an insurer pools premiums from many customers to distribute the risks of loss amongst them. In order for a captive to satisfy the risk distribution criterion, it will be necessary for the captive to accept unrelated business or to distribute its risks to other insurers by way of reinsurance with unrelated parties.

In four recent US Tax Cases, the Courts have held that risk distribution had occurred since a significant percentage of each captive subsidiary's business was with companies not related to it (Amerco and Subsidiaries, and Republic Western Insurance Company v. Commissioner of Internal Revenue 96 T.C. No.3; The Harper Group and Includible Subsidiaries v. Commissioner of Internal Revenue 96 T.C. No. 4; and Sears, Roebuck and Co. and Affiliated Corporations v. Commissioner of Internal Revenue 96 T.C. No. 5 and Ocean Drilling & Exploration Co v U.S. 93-1 USTC). In each of the cases more than 30% of the captive's business came from unrelated companies. The actual percentage of unrelated business will not, of itself, be determinative as to whether or not the arrangement will be accepted as an insurance arrangement; however, it will be one factor to be taken into account. It is the scope of the business which demonstrates the distribution of risk. In the Amerco case it was held that risk distribution was present because of the diverse, multifaceted and substantial unrelated business of the taxpayer.

The concept of risk distribution, although not specifically referred to, was considered in the UK case of Department of Trade and Industry v St Christopher Motorists' Association Ltd (1974) WLR 99. The Court when considering what constitutes the carrying on of an insurance business made the following comments at p. 101 "...the object of which is to ensure that when companies take premiums in return for specified obligations, those companies keep in hand in some form or another sufficient moneys to be able to provide a margin of solvency so that, in the public interest, the chances of insurance companies falling on hard times, in a manner which has been painfully familiar in the past, will be eliminated or, at any rate, reduced."

See Appendices B and C for examples where risk distribution has occurred and Appendices D, E and F where it has not occurred.

If on the objective facts presented to the Commissioner it is not possible to ascertain the character of the premium payment, it may be necessary to have regard to the taxpayer's purpose in incurring the payment (see Fletcher & Ors v FC of T 91 ATC 4950; (1991)22 ATR 613).

What is notable about the observations of the High Court in Fletcher's case is the emphasis laid upon a commonsense appreciation of all the relevant facts and upon determining the proportion between the detriment of the outgoing and the benefit thereby obtained. Where the detriment is disproportionate to the benefit, the court says, it may be necessary "to resolve the problem of the characterisation of the outgoing...by a weighing of the various aspects of the whole set of circumstances, including direct and indirect objects and advantages", and that it be done in a commonsense or practical fashion.

The benefit for which premiums are paid, under an acceptable insurance arrangement, is the benefit accruing under the insurance contract entered into between the parties. This benefit can be described as the promise of indemnity, that is, the promise that in the event of an insured loss, the insured party will be reinstated and protected from the economic consequences of the loss by receipts which will restore its net asset position.

In certain captive insurance arrangements however, the insured obtains no such promise from its captive. Where a payment by a captive to the insured serves to diminish the value of the insured's investment in the captive every such payment from the captive to the insured only serves to reduce the insured's net asset position The following example illustrates this point: Company A prior to the establishment of a captive is in the following asset position: BUILDING $10M LIQUID ASSETS $20M NET ASSETS $30M Company A decides to establish a fully capitalised captive to insure its building. Company A's asset position is then: BUILDING $10M LIQUID ASSETS $10M INVESTMENT IN CAPTIVE $10M NET ASSETS $30M The building is subsequently destroyed and Co A makes a claim on its captive. The captive has no reinsurance. Co A's asset position now is: BUILDING NIL LIQUID ASSETS $10M INVESTMENT IN CAPTIVE NIL INSURANCE RECOVERY $10M NET ASSETS $20M The result of this is that Co A has not been indemnified by the insurance arrangement with its captive. The result would be the same as if Co A had put the $10M in a bank to cover the potential loss.

Adopting a 'commonsense' or 'practical' approach to the example in paragraph 44 it can be seen that the parent obtains no benefit from the establishment of the captive that it would not have obtained by simply setting the money aside in its accounts and investing part of it at fixed interest. The example does not constitute an acceptable insurance arrangement as it does not provide for a promise of indemnity.

Had Co A insured with a non-captive insurer, the premiums would be deductible because the net asset position of Company A would be restored by the payment from the insurer.

Similarly, had the captive reinsured rather than having to resort to its capital to fund the claim the result would have been different. The insured would have stood to make a real economic advantage by making a claim on its captive, which would in turn claim upon its reinsurers. The net asset position of the insured would be enhanced by making a claim. Economically speaking, the captive would have acted as a conduit between its parent and the world insurance market; the economic advantage thus obtained gives substance and meaning to the legal relationship between the parent and its captive such as to qualify the outgoing as one incurred in (potentially) gaining future assessable income, or incurred necessarily in carrying on the parents income producing business.

If an outgoing can be characterised as incidental and relevant as incurred in gaining or producing assessable income, it is not for a Court or the Commissioner to say how much a taxpayer should spend in deriving assessable income (see Ronpibon Tin, Cecil Bros. Pty Ltd v. F C of T (1964) 111 CLR 430 at 434 and Fletcher's case at 4957). However, for a premium to form part of an acceptable insurance contract, it must be of an amount which compensates the insurer for accepting the risk. Therefore, the amount of the premium must not exceed that which a willing but not over anxious insurer is prepared to accept in consideration of taking over the risk.

In many cases captive insurance companies are established to insure risks that may not normally be accepted on the general insurance market. If not insured with a captive, the premium required to insure the risk may be higher than the insured is willing to pay. This may be due to the degree of risk involved or to the inclusion of service and administration costs in the premium. In these circumstances it may be difficult to establish what is an acceptable premium. It is expected that the insured would be able to document its attempts to obtain insurance through the open market and be able to demonstrate its reasons for establishing a captive. In establishing whether the premium paid by a captive insurer is at arm's length between the insured and the captive, the Commissioner will examine all the facts on a case by case basis.

Appendix A of this ruling sets out a series of practical working tests which may be applied to determine whether an arrangement will be accepted as an insurance arrangement. These tests are not exhaustive and the weightings applicable to each may vary according to particular circumstances.

The assessment of non-resident insurers is governed by Division 15 of Part III of the ITAA. The Division only applies to genuine insurance arrangements. Thus where a captive arrangement is not accepted as an insurance arrangement for taxation purposes then Division 15 will have no application.

Section 143 provides for the method of taxing non-resident reinsurers and states: "The insurer shall be deemed to have derived in any year, in respect of the premiums paid or payable in that year under such contracts, a taxable income equal to 10% of the total amount of such premiums: Provided that, where the actual profit or loss derived or made by the insurer in respect of such premiums is established to the satisfaction of the Commissioner, the taxable income of the insurer in respect thereof, or the amount of the loss so made by him shall, subject to this Act, be calculated by reference to receipts and expenditure taken into account in calculating that profit or loss."

It has been suggested that the proviso contained in section 143 is only available to the insurer to calculate taxable income on the basis of receipts and expenditure. We do not accept this proposition. We are of the view that where an insurer is able to calculate its assessable income from its Australian business under Australian taxation law then the insurer should return the income on that basis. Where the insurer is a captive of an Australian resident, the Australian resident would have access to the information to enable it to calculate the insurer's income from its Australian business in accordance with Australian taxation law.

In Case S41 (1967) 17 TBRD 215 Taxation Board of Review No. 1 agreed that the Commissioner has power to insist that an income tax return be lodged on behalf of a non-resident insurer on an actual profit or loss basis where the Commissioner is aware that accurate figures as to receipts and expenditure are available.

Nothing appears in the legislative history of section 143 and its antecedent provision, subsection 28B(2) of Act No. 50 of 1930, which indicates that the proviso is only to operate for the benefit of the insurer, that is, where the insurer can establish that there was a loss or a profit less than 10 per cent of actual premiums.

It is our view that the proviso is equally available to the Commissioner provided that the actual profit or loss derived or made can be established to the Commissioner's satisfaction. In our view, it does not matter whether the information is voluntarily provided by the insurer or ascertained by the Commissioner from other sources. The only relevant factor is that the actual profit or loss derived or made by the insurer in respect of such premiums is established to the satisfaction of the Commissioner.

In the case of a captive insurer, it is expected that the Australian resident insured would have sufficient access to the records of the captive for the resident to be able to calculate the captive's assessable income on the basis of Australian taxation law. Where, the non-resident insurer is not a captive of the Australian resident insured, the insured may not have sufficient access to the insurer's records to enable it to calculate the insurer's income in accordance with Australian taxation law. It would be expected, however, that the insured would have sufficient documentary evidence in order to establish that the payments to the non-resident constituted a valid insurance arrangement. In these circumstances the non-resident would be deemed by virtue of the operation of section 143 to have derived a taxable income equal to 10 per cent of the total amount of such premiums paid or payable from Australian residents.

It should also be noted that section 145 of the Act denies a deduction for premiums where arrangements have not been made to the satisfaction of the Commissioner for the payment of tax in relation to those premiums. It would be expected that the insured would have sufficient documentation to establish that payments to a non-resident constituted a valid insurance arrangement.

The extent to which the provisions in Part IVA are to be applied to deny a deduction for a premium paid to a captive insurer will need to be considered in light of the facts relevant to a particular case. Part IVA will apply where there is a 'scheme' which produces a 'tax benefit' and after the Commissioner has had regard to all the factors set out in section 177D(b) it can be concluded that the sole or dominant purpose of entering into the scheme was to obtain the tax benefit. However, in making a decision as to whether the dominant purpose of the arrangement between the company and its captive insurer is to secure a tax benefit, the Commissioner will have regard to whether there were commercial reasons for entering into the arrangement. Where, for example, complex financial arrangements are entered into which effectively result in the premium paid to the captive insurer passing back to the parent or a related or associated company, the arrangement will be one to which the provisions in Part IVA may apply.

The provisions of Part IVA will be applied to circumstances where the arrangement is one which is designed to cloak the actual effect of the arrangement. The application of Part IVA in these circumstances enables the Commissioner to look at the substance and effect of the arrangement when taken as a whole.

The example at appendix G would be the type of arrangement to which Part IVA would be applied.

Subsection 136AD(3) provides that where a taxpayer has acquired property (a term defined in section 136AA as to extend to the benefit of an insurance contract), the Commissioner is satisfied that the parties are not dealing with each other at arm's length in relation to the acquisition, the consideration given by the taxpayer exceeds an arm's length consideration in respect of the acquisition and the Commissioner determines that the subsection should apply, then for all purposes of the Act, a consideration equal to the arm's length consideration shall be deemed to be given by the taxpayer.

Subsection 136AD(4) provides that, for the purpose of the section, where for any reason (including insufficiency of information available to the Commissioner), it is not possible or practicable for the Commissioner to ascertain the arm's length consideration in respect of the supply or acquisition of property, the arm's length consideration in respect of the supply or acquisition shall be deemed to be such amount as the Commissioner determines.

The question of whether premiums paid under a captive insurance arrangement constitute an arm's length premium will be determined by the facts of each particular case. The application of Division 13 is considered in Taxation Ruling TR94/14.