1.4.1 Non-life insurance

Non-life or short-term insurance transactions under local GAAP are typically accounted for on a deferral and matching basis. This means that premiums are normally recognised as revenue over the contract period, usually on a time apportionment basis. Claims are usually recognised on an incurred basis with no provision for claims that have not occurred at the reporting date. Within this basic model differences exist across local GAAPs on various points of detail which include:

• whether or not claims liabilities are discounted;
• the basis for measuring claims liabilities (e.g. best estimate or a required confidence level);
• whether and what acquisition costs are deferred;
• whether a liability adequacy test (see 7.2.2 below) is performed on the unearned revenue or premium balance and the methodology and level of aggregation applied in such a test; and
• whether reinsurance follows the same model as direct insurance and whether immediate gains on retroactive reinsurance contracts are permitted or not (see 7.2.6 below).

1.4.2 Life insurance

Most local GAAP life insurance accounting models recognise premiums when receivable and insurance contract liabilities are typically measured under some form of discounted cash flow approach that calculates the cash flows expected over the lifetime of the contract. In many jurisdictions the key inputs to the calculation (e.g. discount rates, mortality rates) are set by regulators. Key assumptions may be current or ‘locked-in’ at the contract inception. Differences across jurisdictions also include:

• whether or not certain investment-type products are subject to ‘deposit accounting’ (i.e. only fees are recognised as revenue rather than all cash inflows from policyholders);
• if, and how (contracts with) discretionary participation features are accounted for (discussed at 6 below);
• whether and what acquisition costs are deferred;
• how to account for options and guarantees embedded within contracts; and
• the use of contingency reserves or provisions for adverse deviation.

1.4.3 Embedded value

The embedded value (EV) of a life insurance business is an estimate of its economic worth excluding any value which may be attributed to future new business. The EV is the sum of the value placed on the entity's equity and the value of the in-force business. Typically, an embedded value calculation would involve discounting the value of the stream of after tax profits. For insurance liabilities the income stream would normally be calculated by using the income stream from the backing invested assets as a proxy.

EV is used as an alternative (non-GAAP) performance measure by some life insurers to illustrate the performance and value of their business because local accounting is rarely seen as providing this information. This is because local accounting is often driven by what management consider are ‘unrealistic’ regulatory rules and assumptions. [IFRS 4.BC140].

There is no standardised global measure of embedded value and embedded value practices are diverse. For example, in Europe, the European Insurance CFO Forum, an organisation comprising the Chief Financial Officers of Europe's leading life and property and casualty insurers, has published both European Embedded Values (EEV) and Market Consistent Embedded Value Principles (MCEV). Either of these embedded value models can be applied by Forum members.²

The potential use of embedded value under IFRS 4 is discussed at 8.2.4 below.

2.2.1 Definitions

The following definitions are relevant to the application of IFRS 4. [IFRS 4 Appendix A].

An insurer is the party that has an obligation under an insurance contract to compensate a policyholder if an insured event occurs.

An insurance contract is a contract under which one party (the insurer) accepts significant insurance risk from another party (the policyholder) by agreeing to compensate the policyholder if a specified uncertain future event (the insured event) adversely affects the policyholder.

A reinsurer is the party that has an obligation under a reinsurance contract to compensate a cedant if an insured event occurs.

A reinsurance contract is an insurance contract issued by one insurer (the reinsurer) to compensate another insurer (the cedant) for losses on one or more contracts issued by the cedant.

An insured event is an uncertain future event that is covered by an insurance contract and creates insurance risk.

An insurance asset is an insurer's net contractual rights under an insurance contract.

A reinsurance asset is a cedant's net contractual rights under a reinsurance contract.

An insurance liability is an insurer's net contractual obligations under an insurance contract.

A cedant is the policyholder under a reinsurance contract.

A policyholder is a party that has a right to compensation under an insurance contract if an insured event occurs.

Fair value is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. This definition of fair value is the same as in IFRS 13 – Fair Value Measurement.

Guaranteed benefits are payments or other benefits to which a particular policyholder or investor has an unconditional right that is not subject to the contractual discretion of the issuer.

A discretionary participation feature (DPF) is a contractual right to receive, as a supplement to guaranteed benefits, additional benefits:

1. that are likely to be a significant portion of the total contractual benefits;
2. whose amount or timing is contractually at the discretion of the issuer; and
3. that are contractually based on:
   1. the performance of a specified pool of contracts or a specified type of contract;
   2. realised and/or unrealised investment returns on a specified pool of assets held by the issuer; or
   3. the profit or loss of the company, fund or other entity that issues the contract.

A financial guarantee contract is a contract that requires the issuer to make specified payments to reimburse the holder for a loss it incurs because a specified debtor fails to make payment when due in accordance with the original or modified terms of a debt instrument.

2.2.2 Transactions within the scope of IFRS 4

Unless specifically excluded from its scope (see 2.2.3 below) IFRS 4 must be applied to:

1. insurance contracts (including reinsurance contracts) issued by an entity and reinsurance contracts that it holds; and
2. financial instruments that an entity issues with a discretionary participation feature (see 6.2 below). [IFRS 4.2].

It can be seen from this that IFRS 4 applies to insurance contracts and not just to entities that specialise in issuing insurance contracts. Consistent with other IFRSs it is a transaction-based standard. Consequently, non-insurance entities will be within its scope if they issue contracts that meet the definition of an insurance contract.

IFRS 4 describes any entity that issues an insurance contract as an insurer whether or not the entity is regarded as an insurer for legal or supervisory purposes. [IFRS 4.5].

Often an insurance contract will meet the definition of a financial instrument but IAS 39 and IFRS 9 contain a scope exemption for both insurance contracts and for contracts that would otherwise be within its scope but are within the scope of IFRS 4 because they contain a discretionary participation feature (see 6 below). [IAS 39.2(e), IFRS 9.2.1(e)].

Although the recognition and measurement of financial instruments (or investment contracts) with a discretionary participation feature is governed by IFRS 4, for disclosure purposes they are within the scope of IFRS 7. [IFRS 4.2(b)].

Contracts that fail to meet the definition of an insurance contract are within the scope of IAS 39 or IFRS 9 if they meet the definition of a financial instrument (unless they contain a DPF). This will be the case even if such contracts are regulated as insurance contracts under local legislation. These contracts are commonly referred to as ‘investment contracts’.

Consequently, under IFRS, many insurers have different measurement and disclosure requirements applying to contracts that, under local GAAP or local regulatory rules, might, or might not, have been subject to the same measurement or disclosure requirements. The following table illustrates the standards applying to such contracts.

Many local GAAPs and local regulatory regimes prescribe different accounting requirements for life (long term) and non-life (short-term) insurance contracts (see 1.4 above). However, IFRS 4 does not distinguish between different types of insurance contracts.

IFRS 4 confirms that a reinsurance contract is a type of insurance contract and that all references to insurance contracts apply equally to reinsurance contracts. [IFRS 4.6].

Because all rights and obligations arising from insurance contracts and investment contracts with a DPF are also scoped out of IAS 39 and IFRS 9, IFRS 4 applies to all the assets and liabilities arising from insurance contracts. [IAS 39.2(e), IFRS 9.2.1(e)]. These include:

• insurance and reinsurance receivables owed by the policyholder direct to the insurer;
• insurance claims agreed with the policyholder and payable;
• insurance contract policy liabilities;
• claims handling cost provisions;
• the present value of acquired in-force business (discussed at 9.1 below);
• deferred or unearned premium reserves;
• reinsurance assets (i.e. expected reinsurance recoveries in respect of claims incurred);
• deferred acquisition costs; and
• discretionary participation features (DPF).

Receivables not arising from insurance contracts (such as those arising from a contractual relationship with an agent) would be within the scope of IAS 39 or IFRS 9. When an insurer uses an agent, judgement may be required to determine whether insurance receivables payable by an agent on behalf of a policyholder are within the scope of IFRS 4 or IAS 39/IFRS 9. Such receivables might include:

• balances with intermediaries for premiums received from policyholders but not yet remitted to the insurer by the intermediary;
• funds withheld amounts with intermediaries (e.g. claims funds); and
• loans to intermediaries.

Payables and receivables arising out of investment contracts fall within the scope of IAS 39 or IFRS 9 and the capitalisation and deferral of costs arising from such contracts currently fall within the scope of IFRS 15, IAS 38 and IAS 39 or IFRS 9.

2.2.3 Transactions not within the scope of IFRS 4

IFRS 4 does not address other aspects of accounting by insurers, such as accounting for financial assets held by insurers and financial liabilities issued by insurers (which are within the scope of IAS 32, IAS 39, IFRS 7 and IFRS 9), except: [IFRS 4.3]

• insurers that meet specified criteria are permitted to apply a temporary exemption from IFRS 9 for annual periods beginning on or after 1 January 2018 (see 10.1 below);
• insurers are permitted to apply what the IASB describes as the ‘overlay approach’ to designated financial assets for annual periods beginning on or after 1 January 2018 (see 10.2 below); and
• insurers are permitted to reclassify in specified circumstances some or all of their financial assets so that the assets are measured at fair value through profit or loss (see 8.4 below).

In addition, although not mentioned in IFRS 4, IAS 40 – Investment Property – permits an entity to separately choose between the fair value model or the cost model for all investment property backing liabilities that pay a return linked directly to the fair value of, or returns from, specified assets including that investment property (e.g. contracts with discretionary participation features as discussed at 6 below). [IAS 40.32A]. See Chapter 19 at 5.1.

IFRS 4 also describes transactions to which IFRS 4 is not applied. These primarily relate to transactions covered by other standards that could potentially meet the definition of an insurance contract. It was not the intention of the IASB in issuing IFRS 4 to reopen issues addressed by other standards unless the specific features of insurance contracts justified a different treatment. [IFRS 4.BC10(c)]. These transactions are discussed below.

2.2.4 The product classification process

Because of the need to determine which transactions should be within the scope of IFRS 4, and which transactions are not within its scope, one of the main procedures required of insurers as part of their first-time adoption of IFRS 4 is to conduct a product classification review.

Many large groups developed a product classification process to determine the appropriate classification on a consistent basis. In order to ensure consistency, the product classification process is typically set out in the group accounting manual.

The assessment of the appropriate classification for a contract will include an assessment of whether the contract contains significant insurance risk (discussed at 3 below), and whether the contract contains embedded derivatives (discussed at 4 below), deposit components (discussed at 5 below) or discretionary participation features (discussed at 6 below).

The diagram below illustrates a product classification decision tree.

3.2.1 The meaning of ‘significant’

No quantitative guidance supports the determination of ‘significant’ in IFRS 4. This was a deliberate decision because the IASB considered that if quantitative guidance was provided it would create an arbitrary dividing line that would result in different accounting treatments for similar transactions that fall marginally on different sides of that line and would therefore create opportunities for accounting arbitrage. [IFRS 4.BC33].

The IASB also rejected defining the significance of insurance risk by reference to the definition of materiality within IFRS because, in their opinion, a single contract, or even a single book of similar contracts, could rarely generate a loss that would be material to the financial statements as a whole. [IFRS 4.BC34]. The IASB also rejected the notion of defining significance of insurance risk by expressing the expected (probability weighted) average of the present values of the adverse outcomes as a proportion of the expected present value of all outcomes, or as a proportion of the premium. This idea would have required the constant monitoring of contracts over their life to see whether they continued to transfer insurance risk. As discussed at 3.3 below, an assessment of whether significant insurance risk has been transferred is normally only required at the inception of a contract. [IFRS 4.BC35].

The IASB believes that ‘significant’ means that the insured benefits certainly must be greater than 101% of the benefits payable if the insured event did not occur and it expressed this in the implementation guidance as illustrated below. It is, however, unclear how much greater than 101% the insured benefits must be to meet the definition of ‘significant’.

Some jurisdictions have their own guidance as to what constitutes significant insurance risk. However, as with IFRS 4, other jurisdictions offer no quantitative guidance. Some US GAAP practitioners apply a guideline that a reasonable possibility of a significant loss is a 10% probability of a 10% loss although this guideline does not appear in US GAAP itself. [IFRS 4.BC32]. It is not disputed in the basis for conclusions that a 10% chance of a 10% loss results in a transfer of significant insurance risk and, indeed, the words ‘extremely unlikely’ and ‘a small proportion’ (see 3.2 above) suggests to us that the IASB envisages that significant insurance risk can exist at a different threshold than a 10% probability of a 10% loss.

This lack of a quantitative definition means that insurers must apply their own judgement as to what constitutes significant insurance risk. Although the IASB did not want to create an ‘arbitrary dividing line’, the practical impact of this lack of guidance is that insurers have to apply their own criteria to what constitutes significant insurance risk and there probably is inconsistency in practice as to what these dividing lines are, at least at the margins.

There is no requirement under IFRS 4 for insurers to disclose any thresholds used in determining whether a contract has transferred significant insurance risk. However, IAS 1 – Presentation of Financial Statements – requires an entity to disclose the judgements that management has made in the process of applying the entity's accounting policies that have the most significant effect on the amounts recognised in the financial statements (see Chapter 3 at 5.1.1.B). Liverpool Victoria made the following disclosures about significant insurance risk in its 2016 financial statements.

3.2.2 The level at which significant insurance risk is assessed

Significant insurance risk must be assessed by individual contract, rather than by blocks of contracts or by reference to materiality to the financial statements. Thus, insurance risk may be significant even if there is a minimal probability of material losses for a whole book of contracts. [IFRS 4.B25].

The IASB's reasons for defining significant insurance risk in relation to a single contract were that:

1. although contracts are often managed and measured on a portfolio basis, the contractual rights and obligations arise from individual contracts; and
2. an assessment contract by contract is likely to increase the proportion of contracts that qualify as insurance contracts. The IASB intended to make it easier, not harder, for a contract previously regarded as an insurance contract under local GAAP to meet the IFRS 4 definition. [IFRS 4.BC34].

However, where a relatively homogeneous book of small contracts is known to consist of contracts that all transfer insurance risk, the standard does not require that an insurer examine each contract within that book to identify a few non-derivative contracts that transfer insignificant insurance risk. [IFRS 4.B25].

Multiple, mutually linked contracts entered into with a single counterparty (or contracts that are otherwise interdependent) should be considered a single contract for the purposes of assessing whether significant insurance risk is transferred. [IFRS 4.B25fn7]. This requirement is intended to prevent entities entering into contracts that individually transfer significant insurance risk but collectively do not and accounting for part(s) of what is effectively a single arrangement as (an) insurance contract(s).

If an insurance contract is unbundled (see 5 below) into a deposit component and an insurance component, the significance of insurance risk transferred is assessed by reference only to the insurance component. The significance of insurance risk transferred by an embedded derivative is assessed by reference only to the embedded derivative (see 4 below). [IFRS 4.B28].

3.2.3 Significant additional benefits

The ‘significant additional benefits’ described at 3.2 above refer to amounts that exceed those that would be payable if no insured event occurred. These additional amounts include claims handling and claims assessment costs, but exclude:

1. the loss of the ability to charge the policyholder for future services, for example where the ability to collect fees from a policyholder for performing future investment management services ceases if the policyholder of an investment-linked life insurance contract dies. This economic loss does not reflect insurance risk and the future investment management fees are not relevant in assessing how much insurance risk is transferred by a contract;
2. the waiver on death of charges that would be made on cancellation or surrender of the contract. Because the contract brought these charges into existence, the waiver of them does not compensate the policyholder for a pre-existing risk. Hence, they are not relevant in determining how much insurance risk is transferred by a contract;
3. a payment conditional on an event that does not cause a significant loss to the holder of the contract, for example where the issuer must pay one million currency units if an asset suffers physical damage causing an insignificant economic loss of one currency unit to the holder. The holder in this case has transferred to the insurer the insignificant insurance risk of losing one currency unit. However, at the same time the contract creates non-insurance risk that the issuer will need to pay 999,999 additional currency units if the specified event occurs;
4. possible reinsurance recoveries. The insurer will account for these separately; [IFRS 4.B24] and
5. the original policy premium (but not additional premiums payable in the event of claims experience – see Example 55.26 below).

The definition of insurance risk refers to risk that the insurer accepts from the policyholder. Consequently, insurance risk must be a pre-existing risk transferred from the policyholder to the insurer. A new risk, such as the inability to charge the policyholder for future services, is not insurance risk. [IFRS 4.B12]. The following example illustrates this.

It follows from this that if a contract pays a death benefit exceeding the amount payable on survival (excluding waivers under (b) above), the contract is an insurance contract unless the additional death benefit is insignificant (judged by reference to the contract rather than to an entire book of contracts). Similarly, an annuity contract that pays out regular sums for the rest of a policyholder's life is an insurance contract, unless the aggregate life-contingent payments are insignificant. In this case, the insurer could suffer a significant loss on an individual contract if the annuitant survives longer than expected. [IFRS 4.B26].

Additional benefits could include a requirement to pay benefits earlier than expected if the insured event occurs earlier provided the payment is not adjusted for the time value of money. An example could be whole life insurance cover that provides a fixed death benefit whenever a policyholder dies. Whilst it is certain that the policyholder will die, the timing of death is uncertain and the insurer will suffer a loss on individual contracts when policyholders die early, even if there is no overall expected loss on the whole book of contracts. [IFRS 4.B27].

3.5.1 Service contracts

Some fixed-fee service contracts in which the level of service depends on an uncertain event may meet the definition of an insurance contract. However, in some jurisdictions these are not regulated as insurance contracts. For example, a service provider could enter into a maintenance contract in which it agrees to repair specified equipment after a malfunction. The fixed service fee is based on the expected number of malfunctions but it is uncertain whether a particular machine will break down. Similarly, a contract for car breakdown services in which the provider agrees, for a fixed annual fee, to provide roadside assistance or tow the car to a nearby garage could meet the definition of an insurance contract even if the provider does not agree to carry out repairs or replace parts. [IFRS 4.B6].

In respect of the type of service contracts described above, their inclusion within IFRS 4 seems an unintended consequence of the definition of an insurance contract. However, the IASB stresses that applying IFRS 4 to these contracts should be no more burdensome than applying other IFRSs since:

1. there are unlikely to be material liabilities for malfunctions and breakdowns that have already occurred;
2. if the service provider applied accounting policies consistent with IFRS 15, this would be acceptable either as an existing accounting policy or, possibly, an improvement of existing policies (see 8 below);
3. whilst the service provider would be required to apply the liability adequacy test discussed at 7.2.2 below if the cost of meeting its contractual obligation to provide services exceeded the revenue received in advance, it would have been required to apply IAS 37 to determine whether its contracts were onerous if IFRS 4 did not apply; and
4. the disclosure requirements in IFRS 4 are unlikely to add significantly to the disclosures required by other IFRSs. [IFRS 4.B7].

3.7.1 Lapse, persistency and expense risk

Lapse or persistency risk (the risk that the policyholder will cancel the contract earlier or later than the issuer had expected in pricing the contract) is not insurance risk because, although this can have an adverse effect on the issuer, the cancellation is not contingent on an uncertain future event that adversely affects the policyholder. [IFRS 4.B15].

Similarly, expense risk (the risk of unexpected increases in the administrative costs incurred by the issuer associated with the serving of a contract, rather than the costs associated with insured events) is not insurance risk because an unexpected increase in expenses does not adversely affect the policyholder. [IFRS 4.B15].

Therefore, a contract that exposes the issuer to lapse risk, persistency risk or expense risk is not an insurance contract unless it also exposes the issuer to significant insurance risk.

3.7.2 Insurance of non-insurance risks

If the issuer of a contract which does not contain significant insurance risk mitigates the risk of that contract by using a second contract to transfer part of that first contract's risk to another party, this second contract exposes that other party to insurance risk because the policyholder of the second contract (the issuer of the first contract) is subject to an uncertain event that adversely affects it and thus it meets the definition of an insurance contract. [IFRS 4.B16]. This is illustrated by the following example.

3.9.1 Examples of insurance contracts

The following are examples of contracts that are insurance contracts, if the transfer of insurance risk is significant:

1. insurance against theft or damage to property;
2. insurance against product liability, professional liability, civil liability or legal expenses;
3. life insurance and prepaid funeral plans (although death is certain, it is uncertain when death will occur or, for some types of life insurance, whether death will occur within the period covered by the insurance);
4. life-contingent annuities and pensions (contracts that provide compensation for the uncertain future event – the survival of the annuitant or pensioner – to assist the annuitant or pensioner in maintaining a given standard of living, which would otherwise be adversely affected by his or her survival);
5. disability and medical cover;
6. surety bonds, fidelity bonds, performance bonds and bid bonds (i.e. contracts that provide compensation if another party fails to perform a contractual obligation, for example an obligation to construct a building);
7. credit insurance that provides for specified payments to be made to reimburse the holder for a loss it incurs because a specified debtor fails to make payment when due under the original or modified terms of a debt instrument. These contracts could have various legal forms, such as that of a guarantee, some types of letter of credit, a credit derivative default contract or an insurance contract. Although these contracts meet the definition of an insurance contract they also meet the definition of a financial guarantee contract and are within the scope of IAS 39 or IFRS 9 and IFRS 7 and not IFRS 4 unless the issuer has previously asserted explicitly that it regards such contracts as insurance contracts and has used accounting applicable to such contracts (see 2.2.3.D above);
8. product warranties issued by another party for goods sold by a manufacturer, dealer or retailer are within the scope of IFRS 4. However, as discussed at 2.2.3.A above, product warranties issued directly by a manufacturer, dealer or retailer are outside the scope of IFRS 4;
9. title insurance (insurance against the discovery of defects in title to land that were not apparent when the contract was written). In this case, the insured event is the discovery of a defect in the title, not the title itself;
10. travel assistance (compensation in cash or in kind to policyholders for losses suffered while they are travelling);
11. catastrophe bonds that provide for reduced payments of principal, interest or both if a specified event adversely affects the issuer of the bond (unless the specified event does not create significant insurance risk, for example if the event is a change in an interest rate or a foreign exchange rate);
12. insurance swaps and other contracts that require a payment based on changes in climatic, geological and other physical variables that are specific to a party to the contract; and
13. reinsurance contracts. [IFRS 4.B18].

These examples are not intended to be an exhaustive list.

The following illustrative examples provide further guidance on situations where there is significant insurance risk:

3.9.2 Examples of transactions that are not insurance contracts

The following are examples of transactions that are not insurance contracts:

1. investment contracts that have the legal form of an insurance contract but do not expose the insurer to significant insurance risk, for example life insurance contracts in which the insurer bears no significant mortality risk;
2. contracts that have the legal form of insurance, but pass all significant risk back to the policyholder through non-cancellable and enforceable mechanisms that adjust future payments by the policyholder as a direct result of insured losses, for example some financial reinsurance contracts or some group contracts;
3. self insurance, in other words retaining a risk that could have been covered by insurance. There is no insurance contract because there is no agreement with another party (see 3.2.2.A above);
4. contracts (such as gambling contracts) that require a payment if an unspecified uncertain future event occurs, but do not require, as a contractual precondition for payment, that the event adversely affects the policyholder. However, this does not preclude the specification of a predetermined payout to quantify the loss caused by a specified event such as a death or an accident (see 3.7 above);
5. derivatives that expose one party to financial risk but not insurance risk, because they require that party to make payment based solely on changes in one or more of a specified interest rate, financial instrument price, commodity price, foreign exchange rate, index of prices or rates, credit rating or credit index or other variable, provide in the case of a non-financial variable that the variable is not specific to a party to the contract;
6. a credit-related guarantee (or letter of credit, credit derivative default contract or credit insurance contract) that requires payments even if the holder has not incurred a loss on the failure of a debtor to make payments when due;
7. contracts that require a payment based on a climatic, geological or other physical variable that is not specific to a party to the contract. These are commonly described as weather derivatives and are accounted for under IAS 39 or IFRS 9 (see Chapter 45 at 3.3.1); and
8. catastrophe bonds that provide for reduced payments of principal, interest or both, based on a climatic, geological or other physical variable that is not specific to a party to the contract. [IFRS 4.B19].

The following examples illustrate further situations where IFRS 4 is not applicable.

In January 2008, the Interpretations Committee considered a request for guidance on the accounting for investment or insurance policies that are issued by an entity to a pension plan covering its own employees (or the employees of an entity that is consolidated into the same group as the entity issuing the policy). The Interpretations Committee noted the definitions of plan assets, assets held by a long-term employee benefit plan and a qualifying insurance policy as defined by IAS 19 and considered that, if a policy was issued by a group company to the employee benefit fund then the treatment would depend on whether the policy was a ‘non-transferable financial instrument issued by the reporting entity’. Since the policy was issued by a related party, the Interpretations Committee concluded that it could not meet the definition of a qualifying insurance policy as defined by IAS 19. Because of the narrow scope of this issue the Interpretations Committee declined to either issue an Interpretation or to add the issue to its agenda.³

6.3.1 Unallocated DPF liabilities which are negative

Cumulative unallocated realised and unrealised returns on investments backing insurance and investment contracts with a DPF may become negative and result in an unallocated amount that is negative (a cumulative unallocated loss).

Negative DPF is not addressed in IFRS 4 and does not appear to have been contemplated by the IASB at the time the standard was issued.

Assuming an insurer normally classifies contracts with a DPF as a liability, where such amounts become negative we believe that the insurer is prohibited from recognising an asset or debiting the contract liability related to the cumulative unallocated losses (realised or unrealised) on the investments backing contracts that include DPF features except to the extent that:

• the insurer is contractually entitled to pass those losses to the contract holders; or
• the insurer's previous local GAAP accounting for insurance or investment contracts with a DPF permitted the recognition of such an asset or the debiting of the contract liability when the unallocated investment results of the investments backing a DPF contract were negative.

An insurer is permitted to continue existing GAAP accounting for insurance contracts and investment contracts with a DPF as discussed at 7 below. If an existing GAAP accounting policy specifically allows the recognition of an asset or the reduction of contract liabilities related to cumulative unallocated DPF losses, then in our view continuation of that policy is permitted because IFRS 4 specifically excludes accounting policies for insurance contracts from paragraphs 10‑12 of IAS 8. If the existing GAAP accounting policy does not permit the recognition of an asset or the reduction of contract liabilities in such circumstances then the introduction of such a policy would need to satisfy the relevance and reliability criteria discussed at 8.1 below to be permissible.

Examples of situations in which an insurer is contractually entitled to (partially) recover losses can include contracts with a fixed surrender charge or a market value adjustment feature.

6.3.2 Contracts with switching features

Some contracts may contain options for the counterparty to switch between terms that would, prima facie, result in classification as an investment contract without DPF features (accounted for under IAS 39 or IFRS 9) and terms that would result in a classification as an investment contract with DPF features (accounted for under IFRS 4).

We believe that the fact that this switch option makes these contracts investment contracts with a DPF means that the issuer should continuously be able to demonstrate that the DPF feature still exists and also be able to demonstrate actual switching to a DPF in order to classify these contracts as investment contracts with a DPF under IFRS 4.

7.2.1 Catastrophe and equalisation provisions

Catastrophe provisions are provisions that are generally built up over the years out of premiums received, perhaps following a prescribed regulatory formula, until an amount, possibly specified by the regulations, is reached. These provisions are usually intended to be released on the occurrence of a future catastrophic loss that is covered by current and future contracts. Equalisation provisions are usually intended to cover random fluctuations of claim expenses around the expected value of claims for some types of insurance contract (such as hail, credit guarantee and fidelity insurance) perhaps using a formula based on experience over a number of years. [IFRS 4.BC87]. Consequently, these provisions tend to act as income smoothing mechanisms that reduce profits in reporting periods in which insurance claims are low and reduce losses in reporting periods in which insurance claims are high. As catastrophe and/or equalisation provisions are normally not available for distribution to shareholders, the solvency position of an insurer can be improved.

The recognition of a liability (such as catastrophe and equalisation provisions) for possible future claims, if these claims arise from insurance contracts that are not in existence at the end of the reporting period, is prohibited. [IFRS 4.14(a)].

The IASB considers there is no credible basis for arguing that equalisation or catastrophe provisions are recognisable liabilities under IFRS. Such provisions are not liabilities as defined in the Framework because the insurer has no present obligation for losses that will occur after the end of the contract period. Therefore, without the hierarchy exemption discussed at 7.1 above the recognition of these provisions as liabilities would have been prohibited and the requirement described in the paragraph above preserves that prohibition. [IFRS 4.BC90].

The IASB views the objective of financial statements as not to enhance solvency but to provide information that is useful to a wide range of users for economic decisions. [IFRS 4.BC89(d)]. Present imperfections in the measurement of insurance liabilities do not, in the IASB's opinion, justify the recognition of items that do not meet the definition of a liability. [IFRS 4.BC92(a)].

Although the recognition of catastrophe and equalisation provisions in respect of claims arising from insurance contracts that are not in force at the end of the reporting period are prohibited, such provisions are permitted to the extent that they were permitted under previous accounting policies and they are attributable to policies in force at the end of the reporting period.

Although IFRS 4 prohibits the recognition of these provisions as a liability, it does not prohibit their segregation as a component of equity. Consequently, insurers are free to designate a proportion of their equity as an equalisation or catastrophe provision. [IFRS 4.BC93].

When a catastrophe or equalisation provision has a tax base but is not recognised in the IFRS financial statements, then a taxable temporary difference will arise that should be accounted for under IAS 12 – Income Taxes.

7.2.2 Liability adequacy testing

Many existing insurance accounting models have mechanisms to ensure that insurance liabilities are not understated, and that related amounts recognised as assets, such as deferred acquisition costs, are not overstated. However, because there is no guarantee that such tests are in place in every jurisdiction and are effective, the IASB was concerned that the credibility of IFRS could suffer if an insurer claims to comply with IFRS but fails to recognise material and reasonably foreseeable losses arising from existing contractual obligations.

Therefore, a requirement for the application of a ‘liability adequacy test’ (an assessment of whether the carrying amount of an insurance liability needs to be increased, or the carrying amount of related deferred acquisition costs or related intangible assets decreased, based on a review of future cash flows) was introduced into IFRS 4. This assessment of liability adequacy is required at each reporting date. [IFRS 4.BC94].

If the assessment shows that the carrying amount of the recognised insurance liabilities (less related deferred acquisition costs and related intangible assets such as those acquired in a business combination or portfolio transfer discussed at 9 below) is inadequate in the light of the estimated future cash flows, the entire deficiency must be recognised immediately in profit or loss. [IFRS 4.15].

The purpose of this requirement is to prevent material liabilities being unrecorded.

7.2.3 Insurance liability derecognition

An insurance liability, or a part of such a liability, can be removed from the statement of financial position (derecognised) when, and only when, it is extinguished i.e. when the obligation specified in the contract is discharged, cancelled or expires. [IFRS 4.14(c)].

This requirement is identical to that contained in IAS 39 or IFRS 9 for the derecognition of financial liabilities. [IAS 39.39, IFRS 9.3.3.1]. The IASB said it could identify no reasons for the derecognition requirements for insurance liabilities to differ from those for financial liabilities. [IFRS 4.BC105].

Accordingly, insurance liabilities should not normally be derecognised as a result of entering into a reinsurance contract because this does not usually discharge the insurer's liability to the policyholder. This applies even if the insurer has delegated all claims settlement authority to the reinsurer or if a claim has been fully reinsured.

Derecognition should be distinguished from remeasurement. The carrying amounts of many insurance liabilities are estimates and an insurer should re-estimate its claims liabilities, and hence change their carrying amounts, if that is required by its accounting policies. However, in certain situations the distinction between the two concepts can be blurred, for example where there is a dispute or other uncertainty over the contractual terms of an insurance policy.

IFRS 4 contains no guidance on when or whether a modification of an insurance contract might cause derecognition of the assets and liabilities in respect of that contract.

7.2.4 Offsetting of insurance and related reinsurance contracts

IFRS 4 prohibits offsetting of:

1. reinsurance assets against the related insurance liabilities; and
2. income or expense from reinsurance contracts against the expense or income from the related insurance contracts. [IFRS 4.14(d)].

This prohibition broadly aligns the offsetting criteria for insurance assets and liabilities with those required for financial assets and financial liabilities under IAS 32, which requires that financial assets and financial liabilities can only be offset where an entity:

1. has a legally enforceable right to set-off the recognised amounts; and
2. intends to settle on a net basis, or to realise both the asset and settle the liability simultaneously. [IAS 32.42].

Because a cedant normally has no legal right to offset amounts due from a reinsurer against amounts due to the related underlying policyholder the IASB considers a gross presentation gives a clearer picture of the cedant's rights and obligations. [IFRS 4.BC106].

As a result, balances due from reinsurers should be shown as assets in the statement of financial position, whereas the related insurance liabilities should be shown as liabilities. Because of the relationship between the two, some insurers provide linked disclosures in the notes to their IFRS financial statements as discussed at 11.1.2.A below.

The IFRS 4 requirements, however, appear to be less flexible than those in IAS 32 in that they provide no circumstances in which offsetting can be acceptable. So, for example, ‘pass through’ contracts that provide for reinsurers to pay claims direct to the underlying policyholder would still have to be shown gross in the statement of financial position. IAS 32 also does not address offsetting in the income statement.

7.2.5 Impairment of reinsurance assets

If the IASB had required that the impairment model in IAS 36 be applied to reinsurance assets (as proposed in ED 5) many cedants would have been compelled to change their accounting model for reinsurance contracts in a way that was inconsistent with the accounting for the underlying direct insurance liability. This would have required the cedant to address matters such as discounting and risk, together with the attendant systems implications. Consequently, the IASB concluded that the impairment test should focus on credit risk (arising from the risk of default by the reinsurer and also from disputes over coverage) and not address matters arising from the measurement of the underlying direct insurance liability. It decided the most appropriate way to achieve this was to introduce an incurred loss model based on that contained in IAS 39. [IFRS 4.BC107‑108].

Consequently, a reinsurance asset should be impaired if, and only if:

1. there is objective evidence, as a result of an event that occurred after initial recognition of the asset, that the cedant may not receive all amounts due to it under the terms of the contract; and
2. that event has a reliably measurable impact on the amounts that the cedant will receive from the reinsurer. [IFRS 4.20].

Where a reinsurance asset is impaired, its carrying amount should be reduced accordingly and the impairment loss recognised in profit or loss.

IAS 39 provides various indicators of impairment for financial assets, such as the significant financial difficulty of the obligor and a breach of contract, such as a default in interest or principal payments. IFRS 4 does not provide any specific indicators of impairment relating to reinsurance assets. In the absence of such indicators, it would seem appropriate for insurers to refer to those in IAS 39 as a guide to determining whether reinsurance assets are impaired.

The use of this impairment model means that provisions cannot be recognised in respect of credit losses expected to arise from future events.

IAS 39 permits a portfolio approach to determining impairment provisions for financial assets carried at amortised cost. More specifically, IAS 39 permits a collective evaluation of impairment for assets that are grouped on the basis of similar credit risk characteristics that are indicative of the debtors' ability to pay all amounts due according to the contractual terms (for example on the basis of a credit risk evaluation or grading process that considers asset type, industry, geographical location, collateral type, past-due status and other relevant factors). [IAS 39.AG87].

It is questionable whether an insurer's reinsurance assets would normally exhibit sufficiently similar credit risk characteristics to permit such an approach to determining impairment. That said, IAS 39 is clear that impairment losses recognised on a group basis represent an interim step pending the identification of impairment losses on individual assets in the group of financial assets that are collectively assessed for impairment. As soon as information is available that specifically identifies losses on individually impaired assets in a group, those assets are removed from the group. [IAS 39.AG88].

IFRS 9 does not contain any consequential amendments to IFRS 4 in respect of impairment of reinsurance assets. Therefore, if an insurer applying IFRS 4 adopts IFRS 9 it must continue to use an incurred loss impairment model for reinsurance assets notwithstanding the fact that an expected loss model will be used for financial assets within the scope of IFRS 9.

7.2.6 Accounting policy matters not addressed by IFRS 4

8.2.1 Continuation of existing practices

An insurer may continue the following practices but the introduction of any of them after IFRS has been adopted is not permitted because the IASB believes that they do not satisfy the criteria discussed at 8.1 above. [IFRS 4.25].

8.2.2 Current market interest rates

An insurer is permitted, but not required, to change its accounting policies so that it remeasures designated insurance liabilities (including related deferred acquisition costs and related intangible assets) to reflect current market interest rates. Any changes in these rates would need to be recognised in profit or loss. At that time, it may also introduce accounting policies that require other current estimates and assumptions for the designated liabilities. An insurer may change its accounting policies for designated liabilities without applying those policies consistently to all similar liabilities as IAS 8 would otherwise require. If an insurer designates liabilities for this election, it should apply current market interest rates (and, if applicable, the other current estimates and assumptions) consistently in all periods to all those liabilities until they are extinguished. [IFRS 4.24].

The purpose of this concession is to allow insurers to move, in whole or in part, towards the use of fair value-based measures for insurance contracts.

AXA is an example of an insurance group which has used this option for some of its insurance contracts.

Our view is that, where an entity has elected to account for some, but not all, of its insurance products using current market interest rates, or other current estimates and assumptions, it cannot selectively disregard an input variable, such as a change in interest rates, to determine the value of those liabilities. The input variable must be used every time those insurance contracts are valued.

8.2.3 Prudence

In the IASB's opinion, insurers sometimes measure insurance liabilities on what is intended to be a highly prudent basis that lacks the neutrality required by the IASB's Framework. This may be particularly true for insurers who are required under local GAAP to measure their liabilities on a regulatory basis. However, IFRS 4 does not define how much prudence is ‘sufficient’ and therefore does not require the elimination of ‘excessive prudence’. [IFRS 4.BC133]. As a result, insurers are not required under IFRS 4 to change their accounting policies to eliminate excessive prudence. However, if an insurer already measures its insurance contracts with sufficient prudence, it should not introduce additional prudence. [IFRS 4.26].

The liability adequacy test requirements discussed at 7.2.2 above address the converse issue of understated insurance liabilities.

8.2.4 Future investment margins

An insurer need not change its accounting policies for insurance contracts to eliminate the recognition of future investment margins (which may occur under some forms of embedded value accounting). However, IFRS 4 imposes a rebuttable presumption that an insurer's financial statements will become less relevant and reliable if it introduces an accounting policy that reflects future investment margins in the measurement of insurance contracts, unless those margins directly affect the contractual payments.

Two examples of accounting policies that reflect those margins are:

1. using a discount rate that reflects the estimated return on the insurer's assets; and
2. projecting the returns on those assets at an estimated rate of return, discounting those projected returns at a different rate and including the result in the measurement of the liability. [IFRS 4.27].

Such accounting policies are used in some embedded value methodologies. For example, the European Insurance CFO Forum European Embedded Value (EEV) Principles state that the value of future cash flows from in-force covered business should be the present value of future shareholder cash flows projected to emerge from the assets backing the liabilities of the in-force covered business reduced by the value of financial options and guarantees.⁶ The EEV methodology is considered to be an indirect method of measuring the insurance liability because the measurement of the liability is derived from the related asset. In contrast, direct methods measure the liability by discounting future cash flows arising from the book of insurance contracts only. If the same assumptions are made in both direct and indirect methods, they can produce the same results. [IFRS 4.BC138].

The IASB appears to have been concerned that insurers might take advantage of the lack of specific accounting guidance for insurance contracts in IFRS 4 as an opportunity to change their accounting policies to an embedded value basis on the grounds that this was more relevant and no less reliable, or more reliable and no less relevant than their existing accounting policies (possibly prepared on an ‘excessively prudent’ regulatory basis). The use of embedded value measures by insurers is discussed at 1.4.3 above.

The IASB's view is that the cash flows arising from an asset are irrelevant for the measurement of a liability unless those cash flows affect (a) the cash flows arising from the liability or (b) the credit characteristics of the liability. Therefore, the IASB considers that the following two embedded value approaches involve practices that are incompatible with IFRS, namely:

• applying an asset discount rate to insurance liabilities; and
• measuring contractual rights to investment management fees at an amount that exceeds their fair value (see 8.2.1.B above).

However, the IASB concluded that it could not eliminate these practices, where they were existing accounting policies, until IFRS 17 gives guidance on the appropriate discount rates and the basis for risk adjustments and therefore the use of asset-based discount rates for the measurement of insurance liabilities is not prohibited. [IFRS 4.BC142].

In addition, where embedded values are generally determined on a single best estimate basis, the IASB considers that they do not reflect a full range of possible outcomes and do not generally adequately address liabilities arising from embedded guarantees and options. Further, the IASB believes that existing embedded value approaches are largely unregulated and there is diversity in their application. [IFRS 4.BC141].

It is possible for insurers to introduce accounting policies that use an embedded value approach even if that involves the use of asset-based discount rates for liabilities if they can overcome the rebuttable presumption described above. This will be if, and only if, the other components of a change in accounting policies increase the relevance and reliability of its financial statements sufficiently to outweigh the decrease in relevance and reliability caused by the inclusion of future investment margins. For example, suppose an insurer's existing accounting policies for insurance contracts involves excessively prudent assumptions set at inception and a discount rate prescribed by a regulator without direct reference to market conditions, and the assumptions ignore some embedded options and guarantees. The insurer might make its financial statements more relevant and no less reliable by switching to a basis of accounting that is widely used and involves:

1. current estimates and assumptions;
2. a reasonable (but not excessively prudent) adjustment to reflect risk and uncertainty;
3. measurements that reflect both the intrinsic value and time value of embedded options and guarantees; and
4. a current market discount rate, even if that discount rate reflects the estimated return on the insurer's assets. [IFRS 4.28].

In some measurement approaches, the discount rate is used to determine the present value of a future profit margin. That profit margin is then attributed to different periods using a formula. In those approaches, the discount rate affects the measurement of the liability only indirectly. In these circumstances, the IASB has concluded that the use of a less appropriate discount rate has a limited or no effect on the measurement of the liability at inception. However, in other approaches, the IASB considers that the discount rate determines the measurement of the liability directly. In the latter case, because the introduction of an asset-based discount rate has a more significant effect, the IASB believes that it is highly unlikely that an insurer could overcome the rebuttable presumption described above. [IFRS 4.29].

The IASB believes that in most applications of embedded value the discount rate determines the measurement of the liability directly and therefore it is highly unlikely that an insurer could overcome the rebuttable presumption described above if it wanted to change its accounting policies for insurance contracts to an embedded value basis. [IFRS 4.BC144].

8.5.1 Changes to local GAAP

As most entities are applying some form of local GAAP for their insurance contracts under IFRS 4, a common issue is whether an entity is obliged to change its accounting policy when local GAAP changes or whether the decision to change accounting policy for IFRS purposes is one that remains solely with the insurer.

In our view, the decision to change an accounting policy established on the initial adoption of IFRS is at the discretion of the entity. Accordingly, any change in local GAAP for insurance contracts that was used as the basis for the initial adoption of IFRS does not oblige the insurer to change its accounting policies.

Although an entity is not required to change its policies when local GAAP changes, it can make voluntary changes provided the revised accounting policy makes the financial statements more relevant and no less reliable or more reliable and no less relevant, as discussed at 8.1 above.

When a local accounting standard is changed, it is likely that the change is made for a reason. Therefore, there would normally be a rebuttable presumption that any change in local GAAP is an improvement to the existing standard and so is more relevant and no less reliable or more reliable and no less relevant to users than the previous standard would have been.

The fact that an entity can decide whether or not to apply changes in local GAAP has, over time, led to further diversity in practice.

An entity should not state that it fully applies a particular local GAAP for insurance contracts if it no longer complies with that GAAP due to it not implementing a local GAAP accounting policy change.

9.1.1 Practical issues

10.1.1 Activities that are predominantly connected with insurance

An insurer's activities must be predominantly connected with insurance for the temporary exemption from IFRS 9 to be used. An insurer's activities are predominantly connected with insurance if, and only if: [IFRS 4.20D]

• the carrying amount of its liabilities arising from contracts within the scope of IFRS 4, which includes any deposit components or embedded derivatives unbundled from insurance contracts (see 4 and 5 above), is significant compared to the total carrying amount of all its liabilities; and
• the percentage of the total carrying amount of its liabilities connected with insurance relative to the total carrying amount of all its liabilities is:
  • greater than 90%; or
  • less than or equal to 90% but greater than 80%, and the insurer does not engage in a significant activity unconnected with insurance.

It is observed in the Basis for Conclusions that the IASB decided to require that liabilities within the scope of IFRS 4 be significant compared to total liabilities as a condition for use of the temporary exemption in order to prevent entities with very few such contracts qualifying for the exemption. ‘Significant’ in this context is not quantified. The Board acknowledges that determining significance will require judgement but decided not to provide additional guidance on its meaning because this term is used in other IFRSs and is already applied in practice. [IFRS 4.BC258]. For example, IAS 28 – Investments in Associates and Joint Ventures – defines ‘significant influence’ and provides a rebuttable presumption that a holding of 20 per cent or more of the voting power of an investee gives the investor significant influence. [IAS 28.5].

In assessing whether an insurer engages in a significant activity unconnected with insurance, it should consider: [IFRS 4.20F]

• only those activities from which it may earn income and incur expenses; and
• quantitative or qualitative factors (or both), including publicly available information such as industry classification that users of financial statements apply to the insurer.

For this purpose, liabilities connected with insurance comprise: [IFRS 4.20E]

• liabilities arising from contracts within the scope of IFRS 4, which includes any deposit components or embedded derivatives that are unbundled from insurance contracts (see 4 and 5 above);
• non-derivative investment contract liabilities measured at fair value through profit or loss (FVPL) applying IAS 39, including those liabilities designated at fair value through profit or loss to which the insurer has elected to apply the requirements in IFRS 9 for the presentation of gains and losses (see 10.1 above); and
• liabilities that arise because the insurer issues, or fulfils obligations arising from, the contracts noted in the preceding two bullet points. Examples of such liabilities include derivatives used to mitigate risks arising from those contracts and from the assets backing those contracts, relevant tax liabilities such as the deferred tax liabilities for taxable temporary differences on liabilities arising from those contracts, and debt instruments that are included in the insurer's regulatory capital.

Although not specifically mentioned in the standard, the Basis for Conclusions states that other connected liabilities include liabilities for salaries and other employment benefits for the employees of the insurance activities. [IFRS 4.BC255(b)]. Employee benefit liabilities would include, for example, defined benefit pension liabilities.

The Basis for Conclusions observes that, although non-derivative investment contract liabilities measured at FVPL applying IAS 39 (including those designated at fair value through profit or loss to which the insurer has applied the requirements in IFRS 9 for the presentation in OCI of gains and losses arising from changes in the entity's own credit risk) do not meet the definition of an insurance contract, those investment contracts are sold alongside similar products with significant insurance risk and are regulated as insurance contracts in many jurisdictions. Accordingly, the IASB concluded that insurers with significant investment contracts measured at FVPL should not be precluded from applying the temporary exemption.

However, the IASB noted that insurers generally measure at amortised cost most non-derivative financial liabilities that are associated with non-insurance activities and therefore decided that such financial liabilities (i.e. non-derivative financial liabilities associated with non-insurance liabilities measured at amortised cost) cannot be treated as connected with insurance. [IFRS 4.BC255(a)]. In our view, this would not preclude non-derivative financial liabilities measured at amortised cost being included within the numerator provided such liabilities are connected with insurance. Determining whether such liabilities measured at amortised cost are connected with insurance is a matter of judgement based on facts and circumstances.

The 90% threshold for predominance was set to avoid ambiguity and undue effort in determining eligibility for the temporary exemption from IFRS 9. Nevertheless, the IASB acknowledged that an assessment based solely on this 90% threshold has shortcomings. Accordingly, the IASB decided that when an insurer narrowly fails to meet the threshold, the insurer is still able to qualify for the temporary exemption as long as more than 80% of its liabilities are connected with insurance and it does not engage in a significant activity unconnected with insurance. [IFRS 4.BC256].

When a reporting entity's liabilities connected with insurance are greater than 80% but less than 90% of all of its liabilities and the insurer wishes to apply the temporary exemption, additional disclosures are required (see 10.1.5 below). A reporting entity is not permitted to apply the temporary exemption if its liabilities connected with insurance are less than 80% of all of its liabilities at initial assessment (see 10.1.2 below).

The diagram below illustrates the assessment of the temporary exemption.

An example of how the predominance calculation works is illustrated below.

10.1.2 Initial assessment and reassessment of the temporary exemption

An insurer is required to assess whether it qualifies for the temporary exemption from IFRS 9 at its annual reporting date that immediately precedes 1 April 2016. After that date: [IFRS 4.20G]

• an entity that previously qualified for the temporary exemption from IFRS 9 should reassess whether its activities are predominantly connected with insurance at a subsequent annual reporting date if, and only if, there was a change in the entity's activities (as described below) during the annual period that ended on that date; and
• an entity that previously did not qualify for the temporary exemption from IFRS 9 is permitted to reassess whether its activities are predominantly connected with insurance at a subsequent annual reporting date before 31 December 2018 if, and only if, there was a change in the entity's activities, as described below, during the annual period that ended on that date.

The initial assessment date is prior to the issuance of the amendments to IFRS 4 that introduced the temporary exemption from IFRS 9. The Basis for Conclusions explains that this date was selected in response to feedback from respondents who told the IASB that entities would need to perform the assessment earlier than the application date of IFRS 9 because they would need adequate time to implement IFRS 9 if they did not qualify for the temporary exemption. [IFRS 4.BC264].

A reassessment of the predominance criteria is triggered by a change in activities not a change in the predominance ratio. Therefore, an entity whose predominance ratio, say, fell below 80% at the end of a subsequent reporting period (or rose above 80% in a subsequent reporting period) would not reassess its ability to use the temporary exemption unless this was accompanied by a change in its activities. The IASB considered that a change merely in the level of an entity's insurance liabilities relative to its total liabilities over time would not trigger a reassessment because such a change, in the absence of other events, would be unlikely to indicate a change in the entity's activities. [IFRS 4.BC265].

The Basis for Conclusions clarifies that an entity's financial statements reflect the effect of a change in its activities only after the change has been completed. Therefore, an entity performs the reassessment using the carrying amounts of its liabilities at the annual reporting date immediately following the completion of the change in its activities. For example, an entity would reassess whether its activities are predominantly connected with insurance at the annual reporting date immediately following the completion of an acquisition. [IFRS 4.BC266].

A change in an entity's activities that would result in a reassessment is a change that: [IFRS 4.20H]

• is determined by the entity's senior management as a result of internal or external changes;
• is significant to the entity's operations; and
• is demonstrable to external parties.

Accordingly, such a change occurs only when the entity begins or ceases to perform an activity that is significant to its operations or significantly changes the magnitude of one of its activities; for example, when the entity has acquired, disposed of or terminated a business line.

The IASB expects a change in an entity's activities that would result in reassessment to occur very infrequently. The following are examples of changes that would not result in a reassessment: [IFRS 4.20I]

• a change in the entity's funding structure that in itself does not affect the activities from which the entity earns income and incurs expenses; and
• the entity's plan to sell a business line, even if the assets and liabilities are classified as held for sale applying IFRS 5. A plan to sell a business activity could change the entity's activities and give rise to a reassessment in the future but has yet to affect the liabilities recognised on its balance sheet. This means that a disposal must have completed for it to trigger a reassessment.

If an entity no longer qualifies for the temporary exemption from IFRS 9 as a result of reassessment, then the entity is permitted to continue to apply IAS 39 only until the end of the annual period that began immediately after that reassessment. Nevertheless, the entity must apply IFRS 9 for annual periods beginning on or after 1 January 2021 (or 1 January 2022 based on the proposals in the June 2019 ED). This is illustrated in the following example. [IFRS 4.20J].

When an entity, having previously qualified for the temporary exemption, concludes that its activities are no longer predominantly connected with insurance, additional disclosures are required – see 10.1.5.A below.

An insurer that previously elected to apply the temporary exemption from IFRS 9 may always, at the beginning of any subsequent annual period, irrevocably elect to apply IFRS 9 rather than IAS 39. [IFRS 4.20K]. The transitional requirements of IFRS 9 would apply in those circumstances. An insurer ceasing to use the temporary exemption may also elect to use the overlay approach in the first reporting period in which it applies IFRS 9.

10.1.3 First-time adopters

A first-time adopter, as defined in IFRS 1 – First-time Adoption of International Financial Reporting Standards, may apply the temporary exemption if, and only if, it meets the criteria described at 10.1 above. A first-time adopter is required to assess and reassess predominance on the same reporting dates as an existing user of IFRS. This means that a first-time adopter initially performs the predominance calculation (see 10.1.1 above) using the carrying amounts determined applying IFRSs as at its annual reporting date that immediately precedes 1 April 2016. This reporting period is used even if the entity first applies IFRS 1 with a later date of transition to IFRSs (e.g. the period ended 31 December with a 1 January 2017 date of transition). A first-time adopter that qualifies for the temporary exemption at its annual reporting date that immediately precedes 1 April 2016 must reassess eligibility if there is a change in its activities (see 10.1.2 above). A first-time adopter that does not qualify for the temporary exemption on initial assessment is permitted to reassess whether its activities are predominantly connected with insurance at a subsequent annual reporting date before 31 December 2018 if, and only if, there is a change in its activities. [IFRS 4.20L]. The Basis for Conclusions states that the IASB was prepared to allow first-time adopters to use the temporary exemption if, and only if, those first-time adopters met the same criteria as existing users of IFRS (i.e. the initial assessment of eligibility for the temporary exemption was performed for the same reporting period). [IFRS 4.BC282].

Other requirements in IFRS 1, for example the elections available to a subsidiary that becomes a first-time adopter later than its parent or to an entity that becomes a first-time adopter later than its subsidiary (discussed in Chapter 5 at 5.9), do not override the conditions for using the temporary exemption. Therefore, nothing overrides the requirement that a first-time adopter must meet the predominance criteria at an annual reporting date that immediately precedes 1 April 2016, or, in certain circumstances, at a later date, to apply the temporary exemption from IFRS 9. [IFRS 4.20M].

When making the disclosure required by entities using the temporary exemption (see 10.1.5 below), a first-time adopter should use the requirements and exemptions in IFRS 1 that are relevant to making the assessments required for those disclosures. [IFRS 4.20N].

10.1.4 Relief for investors in associates and joint ventures

IAS 28 requires an entity to use uniform accounting policies when applying the equity method to account for interests in associates and joint ventures. Nevertheless, for accounting periods beginning before 1 January 2021 (or 1 January 2022 based on the proposals in the June 2019 ED), an entity is permitted, but not required, to retain the relevant accounting policies applied by the associate or joint venture as follows: [IFRS 4.20O]

• the entity applies IFRS 9 but the associate or joint venture uses the temporary exemption from IFRS 9; or
• the entity applies the temporary exemption from IFRS 9 but the associate or joint venture applies IFRS 9.

These reliefs are intended to reduce the costs of applying the equity method when an entity does not qualify for the temporary exemption from IFRS 9, and thus applies IFRS 9, but one or more of its associates and joint ventures is eligible and chooses to continue to apply IAS 39 (or vice versa). [IFRS 4.BC279]. The effect is that the underlying financial assets and liabilities held by individual associates and joint ventures, which contribute to the equity accounted result, can be valued on a different basis to those of the reporting entity. This election may be applied separately to each associate or joint venture. [IFRS 4.20Q].

The following accounting applies when an entity uses the equity method to account for its investment in an associate or joint venture: [IFRS 4.20P]

• if IFRS 9 was previously applied in the financial statements used to apply the equity method to that associate or joint venture (after reflecting any adjustments made by the entity), then IFRS 9 should continue to be applied; and
• if the temporary exemption from IFRS 9 was previously applied in the financial statements used to apply the equity method to that associate or joint venture (after reflecting any adjustments made by the entity), then IFRS 9 may be subsequently applied.

The elections described above may be applied separately for each associate or joint venture. [IFRS 4.20Q].

This relief is not limited to consolidated financial statements and could also be applied under the temporary exemption in separate or individual financial statements when associates and joint ventures are accounted for using the equity method.

When an entity uses fair value through profit or loss to measure an investment in associate or joint venture, as permitted by paragraph 18 of IAS 28, this election does not apply (as the election applies only to associates and joint ventures accounted for using the equity method).

10.1.5 Disclosures required for entities using the temporary exemption

Insurers that apply the temporary exemption should disclose information to enable users of their financial statements to: [IFRS 4.39B]

• understand how the insurer qualified for the temporary exemption; and
• compare insurers applying the temporary exemption with entities applying IFRS 9.

The standard contains detailed disclosure requirements for each of these principles which are described at 10.1.5.A and 10.1.5.B below. An insurer that uses the temporary exemption from IFRS 9 should use the transitional provisions in IFRS 9 that are relevant to making the assessments required for these disclosures. The date of initial application for that purpose should be deemed to be the beginning of the first annual period beginning on or after 1 January 2018. [IFRS 4.47]. In our view, comparative period disclosures in the first year of application (i.e. a financial year beginning on or after 1 January 2018) are not required since IFRS 9 does not require restatement of prior periods in the year of application. However, comparative disclosures are required for subsequent accounting periods.

There have been no consequential amendments to IAS 34 – Interim Financial Reporting – and therefore the disclosures detailed at 10.1.5.A and 10.1.5.B below are not required specifically in interim financial statements.

10.1.6 EU ‘top-up’ for financial conglomerates

As explained at 10.1 above, the effect of the amendments on financial conglomerates has proved controversial in some jurisdictions. The European Commission considers that the amendments are not sufficiently broad in scope to meet the needs of all significant insurance entities in the Union. Consequently, for those entities that prepare financial statements in accordance with IFRS as adopted by the EU, the following modification applies:

‘A financial conglomerate as defined in Article 2(14) of Directive 2002/87/EC may elect that none of its entities operating in the insurance sector within the meaning of Article 2(8)(b) of that Directive apply IFRS 9 in the consolidated financial statements for financial years the commencement of which precedes 1 January 2021 where all of the following conditions are met:

1. no financial instruments are transferred between the insurance sector and any other sector of the financial conglomerate after 29 November 2017 other than financial instruments that are measured at fair value with changes in fair value recognised through the profit or loss account by both sectors involved in such transfers;
2. the financial conglomerate states in the consolidated financial statements which insurance entities in the group are applying IAS 39;
3. disclosures requested by IFRS 7 are provided separately for the insurance sector applying IAS 39 and for the rest of the group applying IFRS 9’.⁷

The purpose of (a) above is to prevent a group transferring financial instruments between different ‘sectors’ (i.e. between insurance and non-insurance subsidiaries) with the purpose of either avoiding measurement of those financial instruments at fair value through profit or loss in the group financial statements or recognising previously unrecognised fair value gains or losses in profit or loss.

A financial conglomerate (as defined above) which takes advantage of this ‘top-up’ to use a mixed IFRS 9/IAS 39 measurement model for financial instruments in its consolidated financial statements should not make an explicit and unreserved statement that those consolidated financial statements comply with IFRS as issued by the IASB. [IAS 1.16]. Similarly, depending on local regulations, use of the ‘top-up’ may affect the ability of subsidiaries of the financial conglomerate that are parent entities from using the exemption from preparing consolidated financial statements discussed in Chapter 6 at 2.2.1.D.

10.2.1 Designation and de-designation of eligible financial assets

Eligible financial assets can be designated for the overlay approach on an instrument-by-instrument basis. [IFRS 4.35G].

A financial asset is eligible for the overlay approach if, and only if, the following criteria are met: [IFRS 4.35E]

• it is measured at fair value through profit or loss applying IFRS 9 but would not have been measured at fair value through profit or loss in its entirety applying IAS 39; and
• it is not held in respect of an activity that is unconnected with contracts within the scope of IFRS 4. Examples of financial assets that would not be eligible for the overlay approach are those assets held in respect of banking activities or financial assets held in funds relating to investment contracts that are outside the scope of IFRS 4.

The first criterion above limits the application of the overlay approach to those financial assets for which application of IFRS 9 may result in additional volatility in profit or loss. The Basis for Conclusions states that an example of such a financial asset is one that is measured at fair value through profit or loss applying IFRS 9 but that would have been bifurcated into a derivative and a host applying IAS 39. [IFRS 4.BC240(b)(i)].

The second criterion above is expressed in the negative (i.e. an asset is eligible if it is not held in respect of an activity that is unconnected with contracts within the scope of IFRS 4). Logically, this means that a financial asset is eligible for the overlay approach if it is held in respect of a business activity that is connected with contracts within the scope of IFRS 4. In the majority of situations this is likely to be obvious (e.g. the assets are held to back insurance liabilities). In other situations, application depends on the specific facts and circumstances. Although not mentioned in the text of the standard, the Basis for Conclusions clarifies that financial assets held for insurance regulatory requirements (or for internal capital requirements for the insurance business) are eligible for the overlay approach on the grounds that what became IFRS 17 may affect them. [IFRS 4.BC240(b)(ii)].

An insurer may designate an eligible financial asset for the overlay approach when it elects to apply the overlay approach (see 10.2 above). Subsequently, it may designate an eligible financial asset for the overlay approach when, and only when: [IFRS 4.35F]

• that asset is initially recognised; or
• that asset newly meets the criteria above having previously not met that criteria.

For the purpose of applying the overlay approach to a newly designated financial asset: [IFRS 4.35H]

• its fair value at the date of designation should be its new amortised cost carrying amount; and
• the effective interest rate should be determined based on its fair value at the date of designation (i.e. the new amortised cost carrying amount).

An entity should continue to apply the overlay approach to a designated financial asset until that financial asset is derecognised. However, an entity: [IFRS 4.35I]

• should de-designate a financial asset if the financial asset no longer meets the second criterion specified above. For example a financial asset will no longer meet that criterion when an entity transfers that asset so that it is held in respect of its banking activities or when an entity ceases to be an insurer. When an entity de-designates a financial asset in this way it should reclassify from other comprehensive income to profit or loss as a reclassification adjustment any balance relating to that financial asset; [IFRS 4.35J] and
• may, at the beginning of any annual period, stop applying the overlay approach to all designated financial assets. An entity that elects to stop applying the overlay approach should apply IAS 8 to account for the change in accounting policy (i.e. account for the change retrospectively unless impracticable).

An entity that stops using the overlay approach because it elects to do so or because it is no longer an insurer should not subsequently apply the overlay approach. However, an insurer that has elected to apply the overlay approach but has no eligible financial assets may subsequently apply the overlay approach when it has eligible financial assets. [IFRS 4.35K].

10.2.2 First-time adopters

First-time adopters are permitted to apply the overlay approach. A first-time adopter that elects to apply the overlay approach should restate comparative information to reflect the overlay approach if, and only if, it restates comparative information to comply with IFRS 9. IFRS 1 allows a first-time adopter not to apply IFRS 9 to its comparative period if its first IFRS reporting period begins before 1 January 2019. [IFRS 4.35N].

10.2.3 Disclosures required for entities applying the overlay approach

Insurers that apply the overlay approach should disclose information to enable users of the financial statements to understand: [IFRS 4.39K]

• how the amount reclassified from profit or loss to other comprehensive income in the reporting period is calculated; and
• the effect of that reclassification on the financial statements.

To comply with these principles an insurer should disclose: [IFRS 4.39L]

• the fact that it is applying the overlay approach;
• the carrying amount at the end of the reporting period of financial assets to which the entity applies the overlay approach by class of financial assets (‘class’ is explained in Chapter 54 at 3.3);
• the basis for designating financial assets for the overlay approach, including an explanation of any designated financial assets held outside the legal entity that issues contracts within the scope of IFRS 4;
• an explanation of the total amount reclassified between profit or loss and other comprehensive income in the reporting period in a way that enables users of the financial statements to understand how that amount is derived, including:
  • the amount reported in profit or loss for the designated financial assets applying IFRS 9; and
  • the amount that would have been reported in profit or loss for the designated financial assets if the insurer had applied IAS 39;
• the effect of the reclassification and consequential effects (e.g. income taxes) on each affected line item in profit or loss; and
• if during the reporting period the insurer has changed the designation of financial assets:
  • the amount reclassified between profit or loss and other comprehensive in the reporting period relating to newly designated financial assets applying the overlay approach;
  • the amount that would have been reclassified between profit or loss and other comprehensive income in the reporting period if the financial assets had not been de-designated; and
  • the amount reclassified in the reporting period to profit or loss from accumulated other comprehensive income for financial assets that have been de-designated.

If an entity applies the overlay approach when accounting for its investment in an associate or joint venture using the equity method it should disclose the following, in addition to information required by IFRS 12: [IFRS 4.39M]

• the information set out above for each associate or joint venture that is material to the entity. The amounts disclosed should be those included in the IFRS financial statements of the associate or joint venture after reflecting any adjustments made by the entity when using the equity method rather than the reporting entity's share of those amounts (see Chapter 13 at 5.1.1); and
• the quantitative information set out above and the effect of the reclassification on profit and loss and other comprehensive income in aggregate for all individually immaterial associates or joint ventures. The aggregate amounts:
  • disclosed should be the entity's share of those amounts; and
  • for associates should be disclosed separately from the aggregate amounts disclosed for joint ventures.

Designation of financial assets is always voluntary on an instrument-by-instrument basis under the overlay approach. Therefore, an entity could always avoid the need for these disclosures for associates and joint ventures by not designating an investee's financial assets for the overlay approach.

11.1.1 Disclosure of accounting policies

As noted at 11.1 above, IFRS 4 requires an insurer's accounting policies for insurance contracts and related liabilities, income and expense to be disclosed. [IFRS 4.37(a)]. The implementation guidance suggests that an insurer might need to address the treatment of some or all of the following:

1. premiums (including the treatment of unearned premiums, renewals and lapses, premiums collected by agents and brokers but not passed on and premium taxes or other levies on premiums);
2. fees or other charges made to policyholders;
3. acquisition costs (including a description of their nature);
4. claims incurred (both reported and unreported), claims handling costs (including a description of their nature) and liability adequacy tests (including a description of the cash flows included in the test, whether and how the cash flows are discounted and the treatment of embedded options and guarantees in those tests – see 7.2.2 above). Disclosure of whether insurance liabilities are discounted might be given together with an explanation of the methodology used;
5. the objective of methods used to adjust insurance liabilities for risk and uncertainty (for example, in terms of a level of assurance or level of sufficiency), the nature of those models, and the source of information used in those models;
6. embedded options and guarantees including a description of whether:
   1. the measurement of insurance liabilities reflects the intrinsic value and time value of these items; and
   2. their measurement is consistent with observed current market prices;
7. discretionary participation features (including an explanation of how the insurer classifies those features between liabilities and components of equity) and other features that permit policyholders to share in investment performance;
8. salvage, subrogation or other recoveries from third parties;
9. reinsurance held;
10. underwriting pools, coinsurance and guarantee fund arrangements;
11. insurance contracts acquired in business combinations and portfolio transfers, and the treatment of related intangible assets; and
12. the judgements, apart from those involving estimations, management has made in the process of applying the accounting policies that have the most significant effect on the amounts recognised in the financial statements as required by IAS 1. The classification of a DPF is an example of an accounting policy that might have a significant effect. [IFRS 4.IG17].

Because an insurer's accounting policies will normally be based on its previous local GAAP, the policies for such items will vary from entity to entity.

Set out below are the accounting policies for premiums and claims for Aviva. These are based on UK GAAP which has different requirements for recognition of premiums from life and general (non-life) insurance business.

The following example from Allianz illustrates an accounting policy for reinsurance contracts based on US GAAP.

An example of an accounting policy showing a split of contracts with a DPF between liability and equity (AMP) is shown at 6.1 above.

An example of an accounting policy for a liability adequacy test (Allianz) is shown at 7.2.2.A above.

An example of an accounting policy for salvage and subrogation (Royal & SunAlliance) is shown at 7.2.6.E above.

If the financial statements disclose supplementary information, for example embedded value information, that is not prepared on the basis used for other measurements in the financial statements, it would be appropriate to explain the basis of preparation. Disclosures about embedded value methodology might include information similar to that described above, as well as disclosure of whether, and how, embedded values are affected by estimated returns from assets and by locked-in capital and how those effects are estimated. [IFRS 4.IG18].

11.1.2 Recognised assets, liabilities, income and expense

As noted at 11.1 above, IFRS 4 requires disclosure of the recognised assets, liabilities, income and expense (and cash flows if using the direct method) arising from insurance contracts. [IFRS 4.37(b)].

11.1.3 Gains or losses on buying reinsurance

Gains or losses on buying reinsurance may, using some measurement models, arise from imperfect measurements of the underlying direct insurance liability. Furthermore, some measurement models require a cedant to defer some of those gains and losses and amortise them over the period of the related risk exposures, or some other period. [IFRS 4.IG29].

Therefore, a cedant is required to provide specific disclosure about gains or losses on buying reinsurance as discussed at 7.2.6.C and 11.1 above. In addition, if gains and losses on buying reinsurance are deferred and amortised, disclosure is required of the amortisation for the period and the amounts remaining unamortised at the beginning and end of the period. [IFRS 4.37(b)(i)‑(ii)].

11.1.4 Process used to determine significant assumptions

As noted at 11.1 above, IFRS 4 requires disclosure of the process used to determine the assumptions that have the greatest effect on the measurement of the recognised amounts. Where practicable, quantified disclosure of these assumptions should also be given. [IFRS 4.37(c)].

Some respondents to ED 5 expressed concern that information about assumptions and changes in assumptions (see 11.1.5 below) might be costly to prepare and of limited usefulness. They argued that there are many possible assumptions that could be disclosed and excessive aggregation would result in meaningless information, whereas excessive disaggregation could be costly, lead to information overload, and reveal commercially sensitive information. In response to these concerns, the IASB determined that disclosure about assumptions should focus on the process used to derive them. [IFRS 4.BC212]. Further, the standard refers only to those assumptions ‘that have the greatest effect on the measurement of’ the recognised amounts.

IFRS 4 does not prescribe specific assumptions that should be disclosed, because different assumptions will be more significant for different types of contracts. [IFRS 4.IG33].

For some disclosures, such as discount rates or assumptions about future trends or general inflation, it may be relatively easy to disclose the assumptions used (aggregated at a reasonable but not excessive level, when necessary). For other assumptions, such as mortality rates derived from tables, it may not be practicable to disclose quantified assumptions because there are too many, in which case it is more important to describe the process used to generate the assumptions. [IFRS 4.IG31].

The description of the process used to describe assumptions might include a summary of the most significant of the following:

1. the objective of the assumptions, for example, whether the assumptions are intended to be neutral estimates of the most likely or expected outcome (‘best estimates’) or to provide a given level of assurance or level of sufficiency. If they are intended to provide a quantitative or qualitative level of assurance that level could be disclosed;
2. the source of data used as inputs for the assumptions that have the greatest effect, for example, whether the inputs are internal, external or a mixture of the two. For data derived from detailed studies that are not carried out annually, the criteria used to determine when the studies are updated and the date of the latest update could be disclosed;
3. the extent to which the assumptions are consistent with observable market prices or other published information;
4. a description of how past experience, current conditions and other relevant benchmarks are taken into account in developing estimates and assumptions. If a relationship would normally be expected between past experience and future results, the reasons for using assumptions that differ from past experience and an indication of the extent of the difference could be explained;
5. a description of how assumptions about future trends, such as changes in mortality, healthcare costs or litigation awards were developed;
6. an explanation of how correlations between different assumptions are identified;
7. the policy in making allocations or distributions for contracts with discretionary participation features. In addition, the related assumptions that are reflected in the financial statements, the nature and extent of any significant uncertainty about the relative interests of policyholders and shareholders in the unallocated surplus associated with those contracts, and the effect on the financial statements of any changes during the period in that policy or those assumptions could be disclosed; and
8. the nature and extent of uncertainties affecting specific assumptions. In addition, to comply with IAS 1, an insurer may need to disclose the assumptions it makes about the future, and other major sources of estimation uncertainty, that have a significant risk of resulting in a material adjustment to the carrying amounts of insurance assets and liabilities within the next financial year. [IFRS 4.IG32].

Ping An disclose the following assumptions in relation to their insurance liabilities together with further detail about those assumptions.

Some life insurers give details of the mortality tables used for measuring their insurance contract liabilities and changes in those tables during the reporting period. AMP provide an example of the type of disclosures made.

Prudential provides the following disclosures about allocations and distributions in respect of contracts with a DPF.

11.1.5 The effects of changes in assumptions

As noted at 11.1 above, IFRS 4 requires disclosure of the effects of changes in assumptions used to measure insurance assets and insurance liabilities, showing separately the effect of each change that has a material impact on the financial statements. [IFRS 4.37(d)]. This requirement is consistent with IAS 8, which requires disclosure of the nature and amount of a change in an accounting estimate that has an effect in the current period or is expected to have an effect in future periods. [IFRS 4.IG34].

Assumptions are often interdependent. When this is the case, any analysis of changes by assumption may depend on the order in which the analysis is performed and may be arbitrary to some extent. Not surprisingly, IFRS 4 does not specify a rigid format or content for this analysis. This allows insurers to analyse the changes in a way that meets the objective of the disclosure requirement and is appropriate for their particular circumstances. If practicable, the impact of changes in different assumptions might be disclosed separately, particularly if changes in those assumptions have an adverse effect and others have a beneficial effect. The impact of interdependencies between assumptions and the resulting limitations of any analysis of the effect of changes in assumption might also be described. [IFRS 4.IG35].

The effects of changes in assumptions both before and after reinsurance held might be disclosed, especially if a significant change in the nature or extent of an entity's reinsurance programme is expected or if an analysis before reinsurance is relevant for an analysis of the credit risk arising from reinsurance held. [IFRS 4.IG36].

Aviva disclose the impact of changes in assumptions for their insurance business in a tabular format.

11.1.6 Reconciliations of changes in insurance assets and liabilities

As noted at 11.1 above, IFRS 4 requires reconciliations of changes in insurance liabilities, reinsurance assets and, if any, related deferred acquisition costs, although it does not prescribe the line items that should appear in the reconciliations. [IFRS 4.37(e)].

The changes need not be disaggregated into broad classes, but they might be if different forms of analysis are more relevant for different types of liability. For insurance liabilities the changes might include:

1. the carrying amount at the beginning and end of the period;
2. additional insurance liabilities arising during the period;
3. cash paid;
4. income and expense included in profit or loss;
5. liabilities acquired from, or transferred to, other insurers; and
6. net exchange differences arising on the translation of the financial statements into a different presentation currency, and on the translation of a foreign operation into the presentation currency of the reporting entity. [IFRS 4.IG37].

This reconciliation is also required for each period for which comparative information is presented. [IFRS 4.IG38].

The reconciliations given by CNP Assurances for life insurance, non-life insurance and financial instruments with a DPF are shown below.

In the tables the amounts are shown before and after the impact of reinsurance.

A reconciliation of deferred acquisition costs might include:

1. the carrying amount at the beginning and end of the period;
2. the amounts incurred during the period;
3. the amortisation for the period;
4. impairment losses recognised during the period; and
5. other changes categorised by cause and type. [IFRS 4.IG39].

Aviva's reconciliation of deferred acquisition costs is illustrated below.

An insurer may have intangible assets related to insurance contracts acquired in a business combination or portfolio transfer. IFRS 4 does not require any disclosures for intangible assets in addition to those required by IAS 38 (see 9.2 above). [IFRS 4.IG40].

11.2.1 Objectives, policies and processes for managing insurance contract risks

As noted at 11.2 above, IFRS 4 requires an insurer to disclose its objectives, policies and processes for managing risks arising from insurance contracts and the methods used to manage those risks. [IFRS 4.39(a)].

Such disclosure provides an additional perspective that complements information about contracts outstanding at a particular time and might include information about:

1. the structure and organisation of the entity's risk management function(s), including a discussion of independence and accountability;
2. the scope and nature of its risk reporting or measurement systems, such as internal risk measurement models, sensitivity analyses, scenario analysis, and stress testing, and how these are integrated into the entity's operating activities. Useful disclosure might include a summary description of the approach used, associated assumptions and parameters (including confidence intervals, computation frequencies and historical observation periods) and strengths and limitations of the approach;
3. the processes for accepting, measuring, monitoring and controlling insurance risks and the entity's underwriting strategy to ensure that there are appropriate risk classification and premium levels;
4. the extent to which insurance risks are assessed and managed on an entity-wide basis;
5. the methods employed to limit or transfer insurance risk exposures and avoid undue concentrations of risk, such as retention limits, inclusion of options in contracts, and reinsurance;
6. asset and liability management (ALM) techniques; and
7. the processes for managing, monitoring and controlling commitments received (or given) to accept (or contribute) additional debt or equity capital when specified events occur.

It might be useful to provide disclosures both for individual types of risks insured and overall. They might include a combination of narrative descriptions and specific quantified data, as appropriate to the nature of the contracts and their relative significance to the insurer. [IFRS 4.IG48].

The following extract from AMP provides an example of disclosures concerning the management of life insurance risks.

This extract from Beazley plc illustrates the disclosure of non-life insurance and reinsurance risk policies and processes.

11.2.2 Insurance risk – general matters

As noted at 11.2 above, IFRS 4 requires disclosure about insurance risk (both before and after risk mitigation by reinsurance). [IFRS 4.39(c)].

These disclosures are intended to be consistent with the spirit of the disclosures required by financial instruments. The usefulness of particular disclosures about insurance risk depends on individual circumstances. Therefore, the requirements have been written in general terms to allow practice in this area to evolve. [IFRS 4.BC217].

Disclosures made to satisfy this requirement might build on the following foundations:

1. information about insurance risk might be consistent with (though less detailed than) the information provided internally to the entity's key management personnel as defined in IAS 24 – Related Party Disclosures – so that users can assess the entity's financial position, performance and cash flows ‘through the eyes of management’;
2. information about risk exposures might report exposures both gross and net of reinsurance (or other risk mitigating elements, such as catastrophe bonds issued or policyholder participation features). This is especially relevant if a significant change in the nature or extent of an entity's reinsurance programme is expected or if an analysis before reinsurance is relevant for an analysis of the credit risk arising from reinsurance held;
3. in reporting quantitative information about insurance risk, disclosure of the strengths and limitations of those methods, the assumptions made, and the effect of reinsurance, policyholder participation and other mitigating elements might be useful;
4. risk might be classified according to more than one dimension. For example, life insurers might classify contracts by both the level of mortality risk and the level of investment risk. It may sometimes be useful to display this information in a matrix format;
5. if risk exposures at the reporting date are unrepresentative of exposures during the period, it might be useful to disclose that fact; and
6. the following disclosures required by IFRS 4 might also be relevant:
   1. the sensitivity of profit or loss and equity to changes in variables that have a material effect on them (see 11.2.3 below);
   2. concentrations of insurance risk (see 11.2.4 below); and
   3. the development of prior year insurance liabilities (see 11.2.5 below). [IFRS 4.IG51].

Disclosures about insurance risk might also include:

1. information about the nature of the risk covered, with a brief summary description of the class (such as annuities, pensions, other life insurance, motor, property and liability);
2. information about the general nature of participation features whereby policyholders share in the performance (and related risks) of individual contracts or pools of contracts or entities. This might include the general nature of any formula for the participation and the extent of any discretion held by the insurer; and
3. information about the terms of any obligation or contingent obligation for the insurer to contribute to government or other guarantee funds established by law which are within the scope of IAS 37 as illustrated by Example 55.15 at 3.9.2 above. [IFRS 4.IG51A].

An extract of the narrative disclosures provided by Legal & General about the types of life insurance contracts that it issues is shown below.

The following extract from the financial statements of Amlin illustrates a tabular presentation of insurance risk showing information about premiums and line sizes by class of business.

11.2.3 Insurance risk – sensitivity information

As noted at 11.2 above, IFRS 4 requires disclosures about sensitivity to insurance risk. [IFRS 4.39(c)(i)].

To comply with this requirement, disclosure is required of either:

1. a sensitivity analysis that shows how profit or loss and equity would have been affected had changes in the relevant risk variable that were reasonably possible at the end of the reporting period occurred; the methods and assumptions used in preparing that sensitivity analysis; and any changes from the previous period in the methods and assumptions used. However, if an insurer uses an alternative method to manage sensitivity to market conditions, such as an embedded value analysis, it may meet this requirement by disclosing that alternative sensitivity analysis. Where this is done, the methods used in preparing that alternative analysis, its main parameters and assumptions, and its objectives and limitations should be explained; or
2. qualitative information about sensitivity, and information about those terms and conditions of insurance contracts that have a material effect on the amount, timing and uncertainty of future cash flows. [IFRS 4.39A].

Quantitative disclosures may be provided for some insurance risks and qualitative information about sensitivity and information about terms and conditions for other insurance risks. [IFRS 4.IG52A].

Although sensitivity tests can provide useful information, such tests have limitations. Disclosure of the strengths and limitations of the sensitivity analyses performed might be useful. [IFRS 4.IG52].

Insurers should avoid giving a misleading sensitivity analysis if there are significant non-linearities in sensitivities to variables that have a material effect. For example, if a change of 1% in a variable has a negligible effect, but a change of 1.1% has a material effect, it might be misleading to disclose the effect of a 1% change without further explanation. [IFRS 4.IG53].

Further, if a quantitative sensitivity analysis is disclosed and that sensitivity analysis does not reflect significant correlations between key variables, the effect of those correlations may need to be explained. [IFRS 4.IG53A].

If qualitative information about sensitivity is provided, disclosure of information about those terms and conditions of insurance contracts that have a material effect on the amount, timing and uncertainty of cash flows should be made. This might be achieved by disclosing the information discussed at 11.2.2 above and 11.2.6 below. An entity should decide in the light of its circumstances how best to aggregate information to display an overall picture without combining information with different characteristics. Qualitative information might need to be more disaggregated if it is not supplemented with quantitative information. [IFRS 4.IG54A].

QBE provide the following quantitative information about non-life insurance sensitivities in their financial statements:

11.2.4 Insurance risk – concentrations of risk

As noted at 11.2 above, IFRS 4 requires disclosure of concentrations of insurance risk, including a description of how management determines concentrations and a description of the shared characteristic that identifies each type of concentration (e.g. type of insured event, geographical area, or currency). [IFRS 4.39(c)(ii)].

Such concentrations could arise from, for example:

1. a single insurance contract, or a small number of related contracts, for example when an insurance contract covers low-frequency, high-severity risks such as earthquakes;
2. single incidents that expose an insurer to risk under several different types of insurance contract. For example, a major terrorist incident could create exposure under life insurance contracts, property insurance contracts, business interruption and civil liability;
3. exposure to unexpected changes in trends, for example unexpected changes in human mortality or in policyholder behaviour;
4. exposure to possible major changes in financial market conditions that could cause options held by policyholders to come into the money. For example, when interest rates decline significantly, interest rate and annuity guarantees may result in significant losses;
5. significant litigation or legislative risks that could cause a large single loss, or have a pervasive effect on many contracts;
6. correlations and interdependencies between different risks;
7. significant non-linearities, such as stop-loss or excess of loss features, especially if a key variable is close to a level that triggers a material change in future cash flows; and
8. geographical and sectoral concentrations. [IFRS 4.IG55].

Disclosure of concentrations of insurance risk might include a description of the shared characteristic that identifies each concentration and an indication of the possible exposure, both before and after reinsurance held, associated with all insurance liabilities sharing that characteristic. [IFRS 4.IG56].

Disclosure about the historical performance of low-frequency, high-severity risks might be one way to help users assess cash flow uncertainty associated with those risks. For example, an insurance contract may cover an earthquake that is expected to happen, on average, once every 50 years. If the earthquake occurs during the current reporting period the insurer will report a large loss. If the earthquake does not occur during the current reporting period the insurer will report a profit. Without adequate disclosure of long-term historical performance, it could be misleading to report 49 years of large profits, followed by one large loss, because users may misinterpret the insurer's long-term ability to generate cash flows over the complete cycle of 50 years. Therefore, describing the extent of the exposure to risks of this kind and the estimated frequency of losses might be useful. If circumstances have not changed significantly, disclosure of the insurer's experience with this exposure may be one way to convey information about estimated frequencies. [IFRS 4.IG57]. However, there is no specific requirement to disclose a probable maximum loss (PML) in the event of a catastrophe because there is no widely agreed definition of PML. [IFRS 4.BC222].

Brit Limited discloses the potential impact of modelled realistic disaster scenarios (estimated losses incurred from a hypothetical catastrophe).

11.2.5 Insurance risk – claims development information

As noted at 11.2 above, IFRS 4 requires disclosure of actual claims compared with previous estimates (i.e. claims development). The disclosure about claims development should go back to the period when the earliest material claim arose for which there is still uncertainty about the amount and the timing of the claims payments, but need not go back more than ten years. Disclosure need not be provided for claims for which uncertainty about claims payments is typically resolved within one year. [IFRS 4.39(c)(iii)].

These requirements apply to all insurers, not only to property and casualty insurers. However, the IASB consider that because insurers need not disclose the information for claims for which uncertainty about the amount and timing of payments is typically resolved within a year, it is unlikely that many life insurers will need to give the disclosure. [IFRS 4.IG60, BC220]. Additionally, the implementation guidance to IFRS 4 states that claims development disclosure should not normally be needed for annuity contracts because each periodic payment is regarded as a separate claim about which there is no uncertainty. [IFRS 4.IG60].

It might also be informative to reconcile the claims development information to amounts reported in the statement of financial position and disclose unusual claims expenses or developments separately, allowing users to identify the underlying trends in performance. [IFRS 4.IG59].

The implementation guidance to IFRS 4 provides an illustrative example of one possible format for presenting claims development which is reproduced in full below. From this it is clear that the IASB is expecting entities to present some form of claims development table. This example presents discounted claims development information by underwriting year. [IFRS 4.IG61 IE5]. Other formats are permitted, including for example, presenting information by accident year or reporting period rather than underwriting year. [IFRS 4.IG61].

The example appears to be gross of reinsurance but IFRS 4 is silent on whether development information should be given on both a gross basis and a net basis. If the effect of reinsurance is significant it would seem appropriate to provide such information both gross and net of reinsurance.

The illustrative example also provides only five years of data although the standard itself requires ten (subject to the transitional relief upon first-time adoption). Given the long tail nature of many non-life insurance claims liabilities it is likely that many non-life insurers will still have claims outstanding at the reporting date that are more than ten years old and which will need to be included in a reconciliation of the development table to the statement of financial position.

IFRS 4 is also silent on the presentation of:

• exchange differences associated with insurance liabilities arising on retranslation;
• claims liabilities acquired in a business combination or portfolio transfer; and
• claims liabilities disposed of in a business combination or portfolio transfer.

As IFRS 4 is silent on these matters, a variety of treatments would appear to be permissible provided they are adequately explained to the users of the financial statements and consistently applied in each reporting period. For example, exchange rates could be fixed at the date the claims are incurred, the original reporting period dates or amounts could be retranslated at each reporting date. Claims liabilities acquired in a business combination or portfolio transfer could be reallocated to the prior reporting periods in which they were originally incurred by the acquiree or all liabilities could be allocated to the reporting period in which the acquisition/portfolio transfer occurred.

Aviva's loss (claims) development tables are shown below (although, for brevity, the extract illustrates only six years of data). These are presented on an accident year basis. Aviva discloses both gross and net insurance liabilities in this format.

11.2.6 Credit risk, liquidity risk and market risk disclosures

As noted at 11.2 above, IFRS 4 also requires disclosure of information about credit risk, liquidity risk and market risk that would be required by IFRS 7 if insurance contracts were within the scope of that standard. [IFRS 4.39(d)].

Such disclosure should include:

• summary quantitative data about exposure to those risks based on information provided internally to key management personnel; and
• to the extent not already covered by the disclosures discussed above, the information required by IFRS 7.

IFRS 7 allows disclosures about credit risk, liquidity risk and market risk to be either provided in the financial statements or incorporated by cross-reference to some other statement, such as a management commentary or risk report, that is available to users of the financial statements on the same terms as the financial statements and at the same time. This approach is also permitted for the equivalent disclosures about insurance contracts. [IFRS 4.IG62].

To be informative, the disclosure about credit risk, liquidity risk and market risk might include:

1. information about the extent to which features such as policyholder participation features might mitigate or compound those risks;
2. a summary of significant guarantees, and of the levels at which guarantees of market prices or interest rates are likely to alter cash flows; and
3. the basis for determining investment returns credited to policyholders, such as whether the returns are fixed, based contractually on the return of specified assets or partly or wholly subject to the insurer's discretion. [IFRS 4.IG64].

11.2.7 Exposures to market risk from embedded derivatives

As noted at 11.2 above, disclosure is required if there are exposures to market risk arising from embedded derivatives contained in a host insurance contract if the insurer is not required to, and does not, measure the embedded derivatives at fair value. [IFRS 4.39(e)].

Fair value measurement is not required for derivatives embedded in an insurance contract if the embedded derivative is itself an insurance contract (see 4 above). Examples of these include guaranteed annuity options and guaranteed minimum death benefits as illustrated below. [IFRS 4.IG66].

Both of the examples of embedded derivatives above meet the definition of an insurance contract where the insurance risk is deemed significant. However, in each case, market risk or interest rate risk may be much more significant than the mortality risk. So, if interest rates or equity markets fall substantially, these guarantees would have significant value. Given the long-term nature of the guarantees and the size of the exposures, an insurer might face extremely large losses in certain scenarios. Therefore, particular emphasis on disclosures about such exposures might be required. [IFRS 4.IG69].

To be informative, disclosures about such exposures may include:

• the sensitivity analysis discussed at 11.2.6.C above;
• information about the levels where these exposures start to have a material effect on the insurer's cash flows; and
• the fair value of the embedded derivative, although this is not a required disclosure. [IFRS 4.IG70].

An extract of Aviva's disclosures in respect of financial guarantees and options is shown below.

11.2.8 Other disclosure matters