Financial instruments – Amortised cost and impairment

Date recorded:

Discussion on variations of an expected loss approach

The staff presented the Board with a discussion of the variations of an expected loss approach resulting from the received comments as well as output of the Expert Advisory Panel (EAP). The purpose of this session was orientation only. No decisions have been taken.

The staff presented the Board with a work-plan of issues related to variations of an expected loss approach that will be discussed over the following period. The discussion was meant to provide the Board members with overview of issues related to the Expected Loss Approach and its variations as well as interaction among specific features of the project.

The staff noted that constituents overwhelmingly confirmed the Expected Loss (EL) model over any alternative models (e.g. incurred loss as defined in IAS 39 or fair value approach). Also they noted that majority of constituents agreed with the model including Expected losses estimated over the lifetime of the product. The Board agreed with this overall direction.

The staff then continued to explore the need for the recognition threshold of expected losses. The staff noted that any such threshold would be inconsistent with the objective of the measurement method at amortised cost.

The Board then went on to discuss three major features of the expected loss approach: allocation of initial EL estimate, allocation of subsequent changes to EL estimate and the need for a floor for measurement of EL.

The staff provided overview of four different methods of allocation of initial EL estimate, ranging from the spread over the life using integrated calculation of effective interest rate (as proposed in the ED), two variants of such approach based on decoupling, as proposed by the EAP (spread over life using the annuity approach, spread evenly over average life) to allocating all initial EL in the first period. The staff noted that the last method would require a 'full catch-up' for the subsequent changes to EL estimate. The Board will discuss this issue in detail at a future meeting.

On allocation of subsequent changes to EL estimate, several alternatives were presented:

  • full catch-up as proposed in the ED,
  • 'partial catch-up' taking time-proportionate revised EL to date to the profit or loss, with the remaining spread over the remaining life (either by restating EIR considering the EIR over the life, or by spreading the unrecognised EL evenly over the remaining life), and
  • no catch-up that would treat all subsequent changes prospectively.

The Board discussed the concept of 'Good book' and 'Bad book' as the distinction and precise definition may have an impact on the accounting result. The staff noted that 'Bad book' would require a full catch-up whereas 'Good book' could potentially qualify for other treatment. The Board will discuss this issue at a later stage.

The Board also discussed the implications of transfers between the 'Good book' and the 'Bad book'. The Boards asked the staff to provide a numerical example to compare the alternative of transfer of entire or proportionate balance of the allowance from the 'Good book' to the 'Bad book'. One Board member also noted that the definition of bad book would affect the transfer as well as inclusion or non-inclusion of IBNR provision. The staff noted that the definition of IBNR is not consistent (and does not necessarily equal the definition of IBNR losses in IAS 39).

Finally, the Board discussed the floor for measurement of EL, i.e. whether to apply a symmetrical model or an asymmetrical model (that would require floor in the allowance account that would cover all incurred losses). The Board noted that the need for the floor would result from the overall model.

The Board will discuss impairment issue at a following meeting.

Related Topics

Related Meeting Notes


Correction list for hyphenation

These words serve as exceptions. Once entered, they are only hyphenated at the specified hyphenation points. Each word should be on a separate line.