Why the risk adjustment of IFRS 17 only concerns the non-financial risk ?

[Duc Thinh Vu]

Hello everyone,

I would like to ask a small question about the philosophy of the Risk adjustment of IFRS17. Why should the risk adjustment only take into account the non-financial risk and not both non-financial risk AND financial risk ?

As for me, the Risk adjustment is some kind of supplementary amount the insurer needs to hold to compensate the variation of its passif, and the variation can be caused by both financial risk as well as non-financial risk.

Thank you very much for your help!

 

[randomWalker]

My guess is that there are other adjustments/allowances for financial risks. I know there is a separate liquidity premium which is not part of the risk adjustment (at least in the GI space).

 

[Lindsay Smitherman]

It's because the base liabilities are calculated on a market-consistent basis. This automatically takes into account the market's assessment of the 'financial risks', eg by using risk-free discount rates (so basically assuming that all assets are expected to earn the risk-free rate).

 

[Duc Thinh Vu]

It's because the base liabilities are calculated on a market-consistent basis. This automatically takes into account the market's assessment of the 'financial risks', eg by using risk-free discount rates (so basically assuming that all assets are expected to earn the risk-free rate).

Hi, thanks for your reply
Could you explain to me a little furthur why you say that by using the risk free discount rate, we have already taken into account the financial risk ?

In fact, in the calculation of the base liability, i see that we use, for example, mortality table to calculate. For me, the using of mortality table implies that we have already taken into account the insurance risk, but we have always to calculate a Risk Adjustment to "take into account" it. So the same logic should be applied to financial risk, i.e. we have already used the risk free rate to take into account the financial risk, then we should also create a RA to "take into account" the variations in the estimation process of financial risk once more time.

Maybe I have misunderstood the meaning of RA, could you please help me out of it ? Thank you very much!

 

[Lindsay Smitherman]

Hi

Taking your example forward, the base liability would use a best estimate mortality assumption. The risk adjustment is then applied in order to reflect the risk that actual mortality turns out to be different from the best estimate.

Turning to the discount rate, the base liability doesn't use a best estimate investment return for the discount rate, it uses the risk-free rate - which will be (if anything) lower than the best estimate and therefore will result in a higher liability valuation.

We could consider the best estimate investment return to be equal to risk-free rate + risk margin. So discounting at the risk-free rate is equivalent to discounting at 'best estimate return - risk margin' ... which I think would satisfy your logic, as it more clearly shows how an adjustment has effectively been made in relation to investment risk?

The different approaches are taken because we can observe the 'price' (or 'fair value') that the market places on investment risk, and so can allow for it directly through using the risk-free rate. For other types of risk, we cannot observe the 'fair value' of that risk in the same way (at least, there is much less likely to be a deep and liquid market from which such values could be observed). Therefore the risk adjustment is used as an approximation, with the calculation being designed to represent roughly what compensation the 'market' would need to trade in those non-investment risks.

Hope that helps a little more.

 

[Duc Thinh Vu]

Hi

Taking your example forward, the base liability would use a best estimate mortality assumption. The risk adjustment is then applied in order to reflect the risk that actual mortality turns out to be different from the best estimate.

Turning to the discount rate, the base liability doesn't use a best estimate investment return for the discount rate, it uses the risk-free rate - which will be (if anything) lower than the best estimate and therefore will result in a higher liability valuation.

We could consider the best estimate investment return to be equal to risk-free rate + risk margin. So discounting at the risk-free rate is equivalent to discounting at 'best estimate return - risk margin' ... which I think would satisfy your logic, as it more clearly shows how an adjustment has effectively been made in relation to investment risk?

The different approaches are taken because we can observe the 'price' (or 'fair value') that the market places on investment risk, and so can allow for it directly through using the risk-free rate. For other types of risk, we cannot observe the 'fair value' of that risk in the same way (at least, there is much less likely to be a deep and liquid market from which such values could be observed). Therefore the risk adjustment is used as an approximation, with the calculation being designed to represent roughly what compensation the 'market' would need to trade in those non-investment risks.

Hope that helps a little more.

Hi, thanks for your help! It is much much much clearer for me now. Merry christmas and happy new year to you and your family!

 

[Lindsay Smitherman]

And to you! Glad it helped