1993 SNA Update Information - Consistency Issues
Annuities | | | Description | One sort of life insurance takes the form of transferring one set of assets for an annuity. This is an amount of money, payable annually until the death of the person involved, The sum may be fixed or may be indexed but the amount payable is predetermined, The uncertainty comes from the length of time the annuity is to be paid. For a number of annuities, taken together, a reasonable estimate can be made of the likely liability for the whole though every single one may be subject to considerable uncertainty. (Here again actuarial estimates are needed and it might be more accurate to call the liabilities of the insurance funds “provisions”.)
The SNA does not discuss how annuities should be recorded. They are a common form of private pension provision, however, and some guidance would be helpful. In terms of the 1993 SNA, private pension provision is not treated as a social insurance schemes. Premiums paid are not recorded as current transfers but as a use of saving and net worth. Payments are taken as a run down of net worth and not as any form of current income. Once it is recognised that some form of property income is payable on the remaining liabilities, it becomes clear that this recording in incomplete.
For any company offering annuities, there following evolution of the balance sheet occurs.
There is a reduction in the liabilities at the start of the year (LS) due to the payments made to existing subscribers (A);
There is a reduction in the future liabilities to existing subscribers of (B);
There are new enrolments whereby new subscribers contribute new capital (C);
This represents payment in the current year of D and future liabilities of E (thus C=D+E)
There is investment income arising which should led to a property income distributed to existing and new subscribers (F);
The end of year liabilities (LE) stand at LS – A – B – D + E + F
A simplifying assumption would be that the level of liabilities is in steady state, thus LE = LS because the composition of the cohort of beneficiaries does not change so B = E. Thus we would have F = A + D, or property income in the year exactly matches liabilities paid in the year. This could be recorded in the accounts as
Property income payable to subscribers of F,
“Premium supplements” payable to subscribers by the companies of F,
Adjustment item in the use of income account of F
The change in the level of liabilities ( F + E – D – B – A) is zero by assumption
This has the attraction of recording the income from even a private pension fund in the current accounts but, like pension funds, the amounts of inflows and outflows each year may not match and in the case of outflows exceeding inflows may indicate a situation which is not sustainable in the long term. A more rigorous recording would show the first three items as above but there would be no assumption that the change in the level of liabilities was zero. A positive increase in liabilities would mean that more people were taking out annuities than dying or that the actuarial estimates of life length were too optimistic; , a negative figure would mean that people were living longer than expected and/or that too few new subscribers were being recruited to make up the short fall. |
|
| | Next Steps | | The AEG agreed that it was desirable to clarify the transactions and asset and liabilities to be recorded in the case of annuities. A paper has been invited making proposals. As soon as it is available and posted, comments on it would be welcome. |
|
| | Comments | | Number of comments: 14 | | Date posted | Source | Comment |  | | 10/4/2006 | Jan Heller | agreement | | 10/4/2006 | Irena Tvarijonaviciute | Agree | | 10/4/2006 | Robin Lynch | Agree | | 10/2/2006 | Ole Berner | Agree | | 10/2/2006 | Brent Moulton | I agree with the proposed approach. | | 9/30/2006 | Peter van de Ven | Agree with single comprehensive discussion of life insurance. | | 9/29/2006 | Bank of Korea | We are generally in agreement with the recommendations of the AEG. However, In case of Korea we think that many specific and practical examples need to be shown in the long term for the judgement of the actual estimation. | | 9/29/2006 | A.C. Kulshreshta | I agree with the proposed approach | | 9/29/2006 | Johan Prinsloo | Yes, the suggestion seems reasonable. | | 9/28/2006 | Reimund Mink | I am generally content with the proposed procedure. Shouldn't this section on life insurance and annuities in the cross-cutting chapter cover all insurance? | | 9/28/2006 | Peter Harper | Agree | | 9/27/2006 | Mariam Cover | I agree with the approach proposed on the issue. | | 9/26/2006 | Heidi Arboleda | I agree with the approach proposed on annuities | | 9/15/2006 | Johann Prinsloo | I found the issue on annuities quite puzzling. It is possible to have a similar situation within a defined contribution pension scheme and in addition, it is probably important to distinguish between different types of annuities. There are for example “conventional” and”living” annuities where the first one refers to a guaranteed fixed income with no value left to the estate – with adjustments related to performance and compensation for inflation possible, providing prescribed circumstances. In the case of living annuities the income is flexible, while capital will be available to the estate. Due to a lack of more clearness, I would like to pass on a final answer to this issue. |
|
|
|
| | Conclusion | | The BEA investigations show that, although life insurance typically works as many small payments preceding a single large payment and annuities often function as a single large payment preceding a stream of smaller payments, the mechanism of having property income accruing to the policy holders/annuitants and a fee being payable out of premiums and property income is similar. Thus it is concluded that a single comprehensive discussion of life insurance (to replace the current annex on insurance), to be placed in the chapter on cross-cutting issues, will cover both life insurance and annuities. |
|
|
|
