Paragraph B87 introduces the concept of indifferent-ness. The concept of indifferent-ness and its clear use in the derivation of a RA is fundamental and likely to be a particular focus of auditors.
In short, per paragraph B87, a RA is the additional amount that the entity (insurer) requires to take on insurance risk, and other non-financial risks (most-obviously: expense risk, paragraphs B14 and B86). A RA is inward-looking (paragraphs B87 and B88) and takes no account of the expected cost of transferring insurance risk to a third-party.
Paragraph 37 specifies the principle of what is required of a RA. Paragraph 119 specifies that a RA must also be reported as a ‘confidence level’. It is important (I think) that paragraph 119 is included in the Disclosure section of IFRS 17, and that paragraph 37 is included in the Measurement section of IFRS 17. That is, I interpret that paragraph 119 is a secondary (or latter) stage disclosure requirement that follows the first (or earlier) stage measurement requirement.
Many in the actuarial / accounting / IFRS 17 community favour (or advocate) starting with a confidence level and then deriving back to a RA. Mathematically that is certainly feasible, but it appears to me that the missing link is the need to tie the subsequently-derived RA (having started with the confidence level) back to the paragraph 37 principle of what is required of a RA (i.e. indifferent-ness). Beyond the mechanics of ‘getting to a RA’ no-one seems to have thought about how to bridge that gap.
Therefore, starting by deriving a RA per paragraph 37 and then converting that derived RA to a confidence level per paragraph 119 appears preferable. The ordering of the two paragraphs suggest that this approach is obvious, and natural. Having said that, the flexibility afforded by the IFRS 17 Standard does not make this approach a fait accompli.
How should one then go about deriving a RA? The full mechanics of deriving a RA are beyond the scope of this short article, and demand detailed knowledge of the need (or otherwise) to allow for diversification benefits between a given insurer legal entity and the broader insurance group in which that insurer legal entity sits.
However, I posit one possible approach. Allocate capital (however derived; also beyond the scope of this short article) down to a portfolio of insurance contracts, possibly further allocating down to a group of contracts. It is important to differentiate between capital required to supporting Underwriting Risk and capital required to support Reserve Risk. Now determine the insurer legal entity’s ‘hurdle rate’ return on capital. This may reasonably be considered the insurer’s point of indifferent-ness. Multiply the two to derive a RA.
As a general observation the RA does not seem to have received adequate benefit of the actuarial / accounting / IFRS 17 community’s considerable intellectual capabilities to date. It is a subject that is considerably more complicated than it may first appear.
About the Author:
Richard Hartigan is a general insurance actuary working in London for a major global (re)insurer, currently seconded to their IFRS 17 project team. He graduated from Macquarie University (BEc) in 1992, and completed a MBA (with Distinction) at Cornell University’s Johnson School in 2003. He is a member of both UK (FIA) and Australian (FIAA) institutes.
|
