[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries

Deferred tax in embedded values

Deferred tax arises as a result of timing differ-
ences between tax computations and the
recognition of items in company accounts. This may result in the recognition of either an asset or a liability in the accounts. The Accounting Standards Board (ASB) published a Financial Reporting Standard FRS19 in December 2000 and the relationship of this to embedded values is considered later.
In a life insurance company the situation is more complex, since the Inland Revenue normally taxes profits on what is commonly known as the income less expenses (I-E) basis, but this is subject to a notional case I (NCI) test, which determines a minimum level of profits. The NCI test is implemented by restricting the amount of expenses which can be relieved; in effect tax is still levied on an I-E amount, but adjusted so the taxable amount is equal to the NCI profit. Timing differences can apply to either of these bases, or indeed both at once.
We need to be clear as to whether the ‘accounts’ we are talking about are those in the FSA returns or the accounts required under schedule 9A to the Companies Act 1985. Since this article relates deferred tax to embedded values, which usually take FSA returns as a starting point, and the profits in the tax computations also start from there (see Inland Revenue Statement of Practice 4/95), references to accounts are to those in the FSA returns. However, for income protection business where taxation is based on schedule 9A accounts, the Companies Act accounts are the appropriate starting point.

Types of deferral
Timing differences arise between a notional set of fiscal accounts and the actual accounts under consideration. There are a number of timing differences that may occur and the main ones are shown in the table. The items in table 1 are categorised in two ways:
u Do they occur under I-E or one of the profit taxing regimes (for example NCI, actual case I)?
u Do they create an asset or a liability?
If income is recognised in the accounts before it is recognised in the tax computation, a deferred tax liability is created (and vice versa). Likewise, if expense is recognised in the accounts before it is recognised in the tax computation, a deferred tax asset is created (and vice versa).
Case VI losses are relevant for those categories of life business taxed on profits, that is:
u Pension business
u Life reinsurance business
u Overseas life assurance business
u ISA business
Unrealised capital gains or losses should be taken into account when calculating an embedded value, since investments are normally valued at market value and so include unrealised gains on chargeable assets.
The change in the investment reserve item in the FSA returns in any year does not create a deferred tax effect, and so perhaps should not appear in the table. In fact, section 83 of the Finance Act 1989 states that the amount included in the tax computation is the same as is brought into account in the FSA returns implying there is no timing difference. Prior to the enactment of this legislation the situation was more complex, but only realised capital gains were included in the tax computation and even these could be excluded if they were reserved for policyholders. However, when the legislation was introduced (for periods of account ending on or after 1 January 1990) the then existing investment reserve effectively included a latent tax liability. This was because, in order to access the reserve, it has to be transferred to the revenue account (form 40 of the FSA returns), or ‘brought into account’ in the words of the legislation. In an embedded value calculation this is taken into account when it is assumed that all the estate is distributed. At that stage the investment reserve will implicitly be ‘brought into account’, and allowance must be made for the tax on the shareholders’ share of bonuses.
Tax legislation gives relief for capital allowances instead of depreciation, and to the extent that they are an acceleration of the depreciation in the accounts, a deferred tax liability is created and vice versa.

Interaction of I-E and NCI losses
Under current legislation, where there are losses on the NCI basis there are provisions to avoid an effective double-counting of tax relief from using the NCI losses. These require that the amount of any NCI loss used (ie by loss or group relief) is determined. The case VI losses are then reduced by such amount, or proportionately if they are greater in aggregate. If the NCI loss used exceeds the total case VI losses, then the excess is used to reduce the expenses of management in the I-E computation.
Hence any deferred tax asset in respect of tax losses should be consistent with this approach to relieving losses.

The ASB published FRS19 relating to deferred tax in December 2000; it replaces the previous standard Statement of Standard Accounting Practice SSAP15. Although the replacement standard applies to the main published accounts of companies, it is important to consider how closely current embedded value methodology for deferred tax conforms to FRS19, and what changes might be necessary. There are three main areas to consider.

Partial vs full provisioning
The new standard changes the method used from partial to full provisioning. The former method reflects the fact that existing timing differences will not reverse since they will be replaced by timing differences on future transactions and this method usually results in a lower provision for deferred tax. The full provision method just considers the tax consequences of transactions that have occurred by the balance sheet date. The ASB decided to reject partial in favour of full provisioning on a number of grounds including the following:
u Partial provisioning is subjective and inconsistently applied.
u Partial provisioning takes account of future transactions when measuring an existing liability.
u Both the Financial Accounting Standards Board (ASB) and the International Accounting Standards Committee (IASC) were moving to a full provisioning approach.
It is usual when calculating an embedded value to adopt a closed fund approach to valuing deferred tax and this is consistent with full provisioning.

Recognition criteria
FRS19 describes two approaches to recognition:
u the incremental liability approach; and
u the valuation adjustment approach.
The former recognises deferred tax when it can be considered as an asset or liability in its own right, while the latter considers it as an adjustment to the value at which other assets or liabilities are recognised, and therefore recognises it even if it does not meet the strict recognition criteria as an asset or liability in its own right. The ASB adopted the incremental liability approach.
The key difference is that, under the incremental liability approach, if a company is not obliged to pay the tax at the balance sheet date as a result of past events then it need not recognise the tax. FRS19 gives two examples of this:
u revaluation of a fixed asset, since there is no tax payable until the asset is sold; and
u unremitted earnings of overseas subsidiaries, since no tax is payable until the earnings are remitted and the company is not obliged to remit them.
However, the ASB did require deferred tax to be recognised in the case where an asset is continuously revalued to fair value, with changes in fair value being recognised in the profit and loss account. This covers the case of investments of life companies and so tax on unrealised capital gains should continue to be provided as at present.
The valuation adjustment approach essentially ensures that assets are not valued at more than their post-tax economic value.
The ASB decided to adopt the incremental liability approach since it is more consistent with other accounting standards, does not rely on theoretical models (eg on when assets are sold or earnings remitted), and the valuation adjustment approach might not reflect the eventual payment of tax. Actuaries consider embedded values from a valuation approach and probably reflect that in their treatment of deferred tax. The use of embedded value in reporting the profits of life businesses does, however, involve a process of continuous revaluation reflected in the profit and loss account. The related taxation can therefore be recognised consistently with FRS19.

Companies are permitted, but not required, to discount deferred tax assets and liabilities. In deciding whether or not to discount they should consider materiality, benefits to users of accounts, and industry practice. Given that in calculating an embedded value, any deferred tax will allow for discounting, it appears that current practice can continue. However, FRS19 states that deferred tax balances should be discounted at post-tax risk-free rates, whereas embedded values will normally use a risk discount rate. It would seem to be inconsistent to discount all other cashflows at a risk discount rate and cashflows relating to deferred tax at a risk-free rate.
The inconsistency may be resolved by analysing the calculation as the determination of a discounted value of future non-tax cash flows which are then subjected to tax. The key point is that the application of tax to timing differences where discounting has already been used is not the discounting of the tax itself. Finally, FRS19 states that when scheduling the reversals (ie estimating when the deferred tax items impact on the tax computation), no account should be taken of timing differences expected to arise on future transactions, or of future tax losses. In other words, a closed fund approach consistent with full provisioning.