Risk Neutral Measurement
In this article, I will discuss whether risk-neutral measurement should be allowed in the IFRS 17 final standard.
It is known that IFRS 17 requires splitting cash flows based on their Asset-dependency. According to B74[1]:
Estimates of discount rates shall be consistent with other estimates used to measure insurance contracts to avoid double counting or omissions; for example:
(a) Cash flows that do not vary based on the returns on any underlying items shall be discounted at rates that do not reflect any such variability;
We believe that B77 is the elaboration for B74 (b) (ii) mentioned above in response to industries' backlash on cash flow splitting. According to B77[2]:
IFRS 17 does not require an entity to divide estimated cash flows into those that vary based on the returns on underlying items and those that do not. If an entity does not divide the estimated cash flows in this way, the entity shall apply discount rates appropriate for the estimated cash flows as a whole; for example, using stochastic modelling techniques or risk-neutral measurement techniques.
The implication from B77 is huge. It implicitly allows the use of risk-neutral measurement techniques for all cash flows, with the so called "Adjusted for the effect of that variability ". Consequently, all liability cash flows under the contract will be discounted at risk-free rate.
My view is that, for the valuation of options and guarantees, risk-neutral measurement technique can be applied. However, such technique should not be applied for constructing the whole balance sheet and income statements.
The reasons are listed below. I will go through them one by one:
1. Theoretical Unsoundness
2. Impracticability
3. Inconsistencies
#1 Theoretical Unsoundness
For those studied financial economics, one may know that the theoretical grounds for risk neutral valuation is coming from the Fundamental Theorem of Asset Pricing[3]. It requires a market to be arbitrage-free and complete.
In contrast to some commonly traded assets, insurance contracts are highly illiquid and intransparent. Insurance companies will normally price in margins in the contract, they will also price-in assumed risk premium that is expected to be earned from investment. Insurance contracts by nature is not arbitrage-free, and information is also highly asymmetry hence it is not a complete market.
#2 Impracticability
Even we assumed a perfect Q-measure exists for insurance contract, the cash flows will need to be adjusted significantly to match those under the P-measure.
For example, risk-neutral pricing is required to redetermine the premium. The actual premium, or the real world premium, has priced in risk premium that is expected from investing activities. Directly using the real world premium is a reasonable Q-measure.
Besides, interest margin will vanish under risk neutral measurement, since only risk-free rate is allowed for any kind of investment. The balance sheet and income statement will emerge completely to that may look under P-measure.
#3 Inconsistencies
The introduction of Market Consistent Embedded Value (MCEV) aims at providing greater comparability across (insurance) companies[4]. IFRS is more ambitious that it even wants to create accounting models that are comparable across industries.
The use of risk-neutral measurement technique for insurance will create inconsistencies within the insurance industry, if some companies adopted risk-neutral valuation and some do not. The reason why MCEV can harmonize the industry is that it compulsorily requires risk-neutral technique to be adopted, but not an option to individual companies.
In addition, the use of risk-neutral measurement will create inconsistencies when comparing to other industries as well. Except option pricing (and insurance companies), risk neutral measurement is not commonly used for any other industries.
To sum up, the use of risk-neutral measurement technique should not be allowed for constructing the whole balance sheet / income statement under IFRS 17. While those techniques can be applied for valuation of Options and Guarantees, using it for the whole financial statements is theoretically unsound, create additional works to insurance companies (risk-neutral pricing), and introduce inconsistencies within and across industries, contradicting the IFRS objectives to harmonize accounting models across industries worldwide.
[1] IFRS 17: Paragraph B74
[2] IFRS 17: Paragraph B77
[3] https://en.wikipedia.org/wiki/Fundamental_theorem_of_asset_pricing
[4] https://www.actuary.org/files/MCEV%20Practice%20Note%20Final%20WEB%20031611.4.pdf/MCEV%20Practice%20Note%20Final%20WEB%20031611.4.pdf
In this article, I will discuss whether risk-neutral measurement should be allowed in the IFRS 17 final standard.
It is known that IFRS 17 requires splitting cash flows based on their Asset-dependency. According to B74[1]:
Estimates of discount rates shall be consistent with other estimates used to measure insurance contracts to avoid double counting or omissions; for example:
(a) Cash flows that do not vary based on the returns on any underlying items shall be discounted at rates that do not reflect any such variability;
(b) Cash flows that vary based on the returns on any financial underlying items shall be:
i. Discounted using rates that reflect that variability; or
ii. Adjusted for the effect of that variability and discounted at a rate that reflects the adjustment made.
i. Discounted using rates that reflect that variability; or
ii. Adjusted for the effect of that variability and discounted at a rate that reflects the adjustment made.
We believe that B77 is the elaboration for B74 (b) (ii) mentioned above in response to industries' backlash on cash flow splitting. According to B77[2]:
IFRS 17 does not require an entity to divide estimated cash flows into those that vary based on the returns on underlying items and those that do not. If an entity does not divide the estimated cash flows in this way, the entity shall apply discount rates appropriate for the estimated cash flows as a whole; for example, using stochastic modelling techniques or risk-neutral measurement techniques.
The implication from B77 is huge. It implicitly allows the use of risk-neutral measurement techniques for all cash flows, with the so called "Adjusted for the effect of that variability ". Consequently, all liability cash flows under the contract will be discounted at risk-free rate.
My view is that, for the valuation of options and guarantees, risk-neutral measurement technique can be applied. However, such technique should not be applied for constructing the whole balance sheet and income statements.
The reasons are listed below. I will go through them one by one:
1. Theoretical Unsoundness
2. Impracticability
3. Inconsistencies
#1 Theoretical Unsoundness
For those studied financial economics, one may know that the theoretical grounds for risk neutral valuation is coming from the Fundamental Theorem of Asset Pricing[3]. It requires a market to be arbitrage-free and complete.
In contrast to some commonly traded assets, insurance contracts are highly illiquid and intransparent. Insurance companies will normally price in margins in the contract, they will also price-in assumed risk premium that is expected to be earned from investment. Insurance contracts by nature is not arbitrage-free, and information is also highly asymmetry hence it is not a complete market.
#2 Impracticability
Even we assumed a perfect Q-measure exists for insurance contract, the cash flows will need to be adjusted significantly to match those under the P-measure.
For example, risk-neutral pricing is required to redetermine the premium. The actual premium, or the real world premium, has priced in risk premium that is expected from investing activities. Directly using the real world premium is a reasonable Q-measure.
Besides, interest margin will vanish under risk neutral measurement, since only risk-free rate is allowed for any kind of investment. The balance sheet and income statement will emerge completely to that may look under P-measure.
#3 Inconsistencies
The introduction of Market Consistent Embedded Value (MCEV) aims at providing greater comparability across (insurance) companies[4]. IFRS is more ambitious that it even wants to create accounting models that are comparable across industries.
The use of risk-neutral measurement technique for insurance will create inconsistencies within the insurance industry, if some companies adopted risk-neutral valuation and some do not. The reason why MCEV can harmonize the industry is that it compulsorily requires risk-neutral technique to be adopted, but not an option to individual companies.
In addition, the use of risk-neutral measurement will create inconsistencies when comparing to other industries as well. Except option pricing (and insurance companies), risk neutral measurement is not commonly used for any other industries.
To sum up, the use of risk-neutral measurement technique should not be allowed for constructing the whole balance sheet / income statement under IFRS 17. While those techniques can be applied for valuation of Options and Guarantees, using it for the whole financial statements is theoretically unsound, create additional works to insurance companies (risk-neutral pricing), and introduce inconsistencies within and across industries, contradicting the IFRS objectives to harmonize accounting models across industries worldwide.
[1] IFRS 17: Paragraph B74
[2] IFRS 17: Paragraph B77
[3] https://en.wikipedia.org/wiki/Fundamental_theorem_of_asset_pricing
[4] https://www.actuary.org/files/MCEV%20Practice%20Note%20Final%20WEB%20031611.4.pdf/MCEV%20Practice%20Note%20Final%20WEB%20031611.4.pdf
Comments
Post a Comment