Insurance contracts: ED/2013/7

Comments from ACCA to the International Accounting Standards Board (IASB), October 2013.

Comments from ACCA to the International Accounting Standards Board (IASB), October 2013.


ACCA agrees with the ED’s main proposals. There is a significant demand from preparers and users for a complete standard from the IASB on insurance contracts. Ideally the same standard would be applicable to US insurers as well and we hope that as much as possible should be done to try to achieve that goal. In a similar way Islamic financial institutions should be able to use the new standard as much as practicable. In this case guidance on any presentational adjustments would be useful.  

We support the model for insurance accounting as sound in principle – the measurement of liabilities using the building block approach to estimating future cash flows and the avoidance of day 1 profits, as well as the appropriate simplifications of the premium allocation approach. 

There are concerns however around the complexity the difficulties of the application in practice of certain parts of the model and also of the extensive disclosures proposed. Concerns about both of these will be best met by field-testing the various requirements, including the transitional provisions with both preparers but also with the users of the financial information. We would encourage further testing to be done and for the results to be published.


Q1. Adjusting the contractual service margin (CSM) for some changes in expected cash flows, rather than recognition in profit & loss

Given that the CSM represents a difference between values of estimated cash inflows and outflows, we do not see why that difference should be fixed at inception of the contract. If those estimates need to change then it seems right in principle that for some changes the remaining CSM might be adjusted rather than just the results for the period when the re-estimation occurs.

However we have some concerns over how consistently the distinction will be able to be made in practice between changes that do, or do not, relate to future service and the consequential potential for manipulation. 

The other matter concerning the recognition of margins we have noticed concerns reinsurance. In some cases the reinsurance is not a separate risk management transaction but is entered into as part of entirely coherent design of the product being sold. In these cases the CSM should include the effects of the reinsurance which should be recognised as the overall service is delivered.

Q2. Contracts which require the entity to hold underlying items and specify a link to returns on those underlying items

We agree with the proposal here. There are some concerns that the definition of the linkage might have been drawn too strictly leaving some products which are in substance linked to an underlying investment outside of this treatment.

Q3. Presentation of insurance contract revenue and expenses

It will be helpful to have a structure for the P&L of revenue, expense and interest rather than simply reporting the net margin. However it could be clearer from either the paragraphs 56 to 60 or from the examples (especially example 7) how all of these requirements would fit together. 

Though in many cases this will be a new presentation it seems likely to be helpful in identifying and presenting separately an underwriting margin and the effects of interest and of other revisions to estimates. Given its relative novelty the disclosure requirement reconciling insurance revenue with premiums received will be a helpful link to what in some cases is a current benchmark.

We recognise that the definition of insurance revenue is trying to be as consistent as possible with the definition of revenue in the new general standard. In principle this must be desirable, but not at the price of excluding performance measures that are currently proving useful. We consider the needs to make comparisons between insurance companies are significantly greater than across sectors. There will be significant analysis of the revenue and the costs in the further disclosure requirements which are extensive.

The presentation of insurance revenue and expenses is another area where field-testing with insurers and with their investors will help to confirm the practicality and usefulness of this presentation and the detail of the disclosures provided.

Q4. Interest expense in the P&L and OCI

We note that the main intention of the proposal is to match the treatment of an insurer’s assets and liabilities to avoid undue volatility in P&L that comes from changes in interest rates, on the assumption that the insurance liabilities are matched by bonds which will be accounted for under IFRS9 either

  • At amortised cost, or
  • Fair value through OCI

This is likely to be generally but not always the case.

We also note that this treatment of changes in interest rates is not consistent with other accounting standards, for example IAS37 where discount rates are updated and all changes go through P&L. On the other hand for pension obligations under IAS19 discount rates are current but changes will be in the actuarial gain and losses taken to OCI. In IFRS9 interest on other financial liabilities are at historical rates, or if at fair value then changes in own credit effects are in OCI. 

We agree with the proposal on the basis that priority should be given to the link between the treatments in IFRS9 and in these proposals. We note the issue is being covered in the revision of the conceptual framework, and we look forward to some principles that would allow more consistent treatment of items as between P&L and OCI.

Q5. Effective date and transition

We agree that 3 years to allow for the adoption of the standard is reasonable given the potentially new data collection and the different accounting involved.

The practical ability to synchronise the application of IFRS9 and these proposals is highly important to insurance company accounts and IASB should ensure that this is the case if at all possible.

We are concerned that the full retrospective adoption of the standard now required will be difficult in practice. The restatement should be an important part of the field-testing which is needed.