The topic of own credit risk generates perhaps more strongly held views than any other accounting topic. For example, in around 2011, the IASB was developing a discussion paper on insurance contracts. Some of us visited one of the largest insurers in the world. They knew almost nothing about the project. But they had heard one thing: that the IASB was moving towards a position that the measurement of insurance liabilities should reflect own credit risk. Like many others, they hated this idea.
In this post, I provide a brief reminder that the fair value of a liability does reflect own credit risk. I then summarise arguments for and against reflecting own credit risk. Finally, I consider whether own credit risk has a role in measuring liabilities using measurement bases other than fair value.
Own credit risk and non-performance risk
I use the term ‘own credit risk’ in this paper, for brevity. Similar considerations apply to all forms of ‘non-performance risk’.
IFRS 13 Fair Value Measurement defines non-performance risk as ‘the risk that an entity will not fulfil an obligation’. The risk covers not only failure to pay cash but also unintended or wilful failure by a service provider to perform. ‘Non-performance risk’ conveys that scope more clearly than ‘own credit risk’ does.
It is worth remembering that the degree of own credit risk present in a liability does not depend only on the debtor’s financial strength. It also depends on the relative priorities of the various classes of the debtor’s liabilities.
The rest of this post covers
- fair value and own credit risk
- arguments for and against own credit risk in measuring liabilities
- own credit risk in other measurement bases
- some IASB papers on own credit risk
Fair value and own credit risk
Applying IFRS Standards, some financial liabilities are measured at fair value, both initially and subsequently. Other financial liabilities are measured initially at fair value less transaction costs and then subsequently at amortised cost.
I don’t discuss in this post whether and when liabilities should measured at fair value.
The role of own credit risk in fair value seems to be widely understand now. So, I will just provide a brief reminder of:
- what the IASB’s Conceptual Framework for Financial Reporting says on this subject.
- why the fair value of a liability reflects own credit risk.
The Conceptual Framework says:
- the fair value of an asset (or liability) is the price that would be received to sell the asset (or paid to transfer the liability) in an orderly transaction between market participants at the measurement date. (paragraph 6.12)
- when fair value is determined indirectly using measurement techniques, the factors reflected in those techniques include the possibility that the entity may fail to fulfil its liability (own credit risk). (paragraphs 6.14-6.15)
Why the fair value of a liability reflects own credit risk
The fair value of a liability is the price for a transfer that neither increases not decreases the effect of own credit risk. This is because:
- the transferor would not willingly pay the price that a willing transferee would require for a transfer that decreases the effect of own credit risk. To pay for a transfer decreasing that effect, the transferor would in effect be buying a credit upgrade. Yet the transferor itself would not benefit from the upgrade.
- the debtor would not consent to a transfer that increases its effect.
Own credit risk in other measurement bases
This section discuses arguments for and against reflecting own credit risk in liability measurement when the measurement basis is not fair value. Some arguments apply mainly to initial measurement. Others apply mainly to subsequent measurement.
Own credit risk in initial measurement
Some arguments for reflecting own credit risk in measuring liabilities relate mainly to day 1 (initial recognition of the liability). But they may also apply to some extent to subsequent measurement:
- Companies generally measure their debt liabilities at the amount of cash received—assumed to also equal the debt’s fair value in most cases. That initial measurement equals the total contractual cash flows, discounted at the contractual interest rate. Thus, that measurement inherently reflects the own credit risk reflected in the interest rate charged in the debt. There is no obvious reason to adopt a different principle for other types of liability (such as pension liabilities or provisions) on day 1 (at initial recognition).
- Excluding own credit risk ignores those outcomes in which some or all contractual cash outflows do not occur. Ignoring those outcomes is incompatible with measurements based on expected values (ie probability-weighted averages of all possible outcomes). For example, suppose that it is estimated that a company will meet its liabilities in 99% of the outcomes, but will be unable to do so in the other 1% of outcomes. In that example, excluding own credit risk would report the company as expecting to meet all of its liabilities in all outcomes (100% of them).
For discussion of expected values, please see Measurements based on future cash flows – Accounting Miscellany
Some arguments for excluding own credit risk relate mainly or partly to day 1 (initial recognition of the liability). But they may also apply to subsequent measurement:
- Reflecting own credit risk would not reflect a company’s intention to meet all its liabilities in full.
- It may be difficult to measure the effect of own credit risk. This is especially the case when that effect is not readily observable separately in market prices. So, it might be difficult to measure that effect to include it in measuring non-traded liabilities. (On the other hand, it might also be difficult to exclude own credit risk in measuring traded liabilities.)
And some of the measurement difficulties may be particularly severe for liabilities with very long maturities. Discount rates for very long maturities – Accounting Miscellany
Own credit risk in subsequent measurement
Some arguments for reflecting own credit risk in measuring liabilities relate mainly to subsequent measurement (day 2 and after):
- A measurement model is inconsistent if it includes the credit characteristics of liabilities at inception but ignores them later. Such a measurement model is also inconsistent if then ignores subsequent changes in their effect.
- Consider an entity with two liabilities that require identical contractual cash outflows but that were incurred at different times when the entity’s credit standing was different. That entity will have to have to pay two different interest rates. If measurement ignores changes in the effects of own credit risk, the entity will measure the liabilities at different amounts, even though they are economically identical.
Most of the main arguments for excluding own credit risk in measuring liabilities relate mainly or partly to subsequent measurement (day 2 and after):
- Including own credit risk makes it more difficult to use reported numbers in assessing the company’s financial strength and solvency. Moreover, that assessment becomes more difficult at exactly the time when that assessment may become most important—when the company’s financial strength is worsening.
- Reflecting changes in the effect of own credit risk would move the measurement of many liabilities away from their face amount. Responses to several IASB consultations suggest that respondents find it difficult to interpret liability measurements differing from face amount.
- Reflecting changes in the effect of own credit risk would also move the measurement of many liabilities away from the amount that would be payable if the company were to be liquidated immediately. Even if the notes disclose the amount payable on immediate liquidation, some users of financial statements might not be comfortable if companies use a lower measurement in the statement of financial position.
- When liability measurement reflects own credit risk, a company reports a gain when its own credit risk increases. Reporting this gain is counter-intuitive because the company’s financial position has become worse, not better. And although there are valid mechanical and economic reasons for that gain, many people find it difficult to interpret a gain as the direct signal of a deterioration in financial position.
- A decline in a company’s credit standing would normally occur at the same time as an impairment of internally generated goodwill, which is not recognised as an asset. Because that impairment is not recognised as an expense, it would be misleading to recognise a gain when the company’s credit standing declines. And, although some have suggested solving this mismatch problem by recognising and remeasuring internally generated goodwill, recognising such goodwill is beyond the reasonable objective of financial statements.
- Recognising a gain when own credit risk deteriorates might suggest that the company could readily enter into a transaction to realise the gain without becoming insolvent. If such a transaction is unlikely or impossible, reporting the gain may produce irrelevant and misleading information.
- Recognising a gain when own credit risk deteriorates is, some people believe, inconsistent with the going concern assumption, described in paragraph 3.9 of the Conceptual Framework.
- If a company recognises a gain when its own credit risk gets larger, that amount will, if there is no default, reverse in later periods as an expense.
The idea of reflecting own credit risk in liability measurement perplexes many people. The following quotation illustrates this perplexity well. It is from the report of a judicial investigation into the collapse of an Australian insurance company HIH Insurance. (Thanks to Angus Thomson for tracking down this quotation for me.)
The idea that a true and fair view of the value of HIH’s liabilities to its policyholders could be obtained by asking what knowledgeable and willing parties (that is HIH and its policyholders) would pay to settle those liabilities is to my mind bizarre. If the policyholders would rationally accept $0.20 in the dollar because HIH is insolvent, does that mean that HIH should report its liabilities as being a fifth of what they actually are? The proposition has only to be stated to be rejected.” https://parlinfo.aph.gov.au/parlInfo/search/display/display.w3p;query=Id%3A%22publications%2Ftabledpapers%2F19869%22
The failure of HIH Insurance. HIH Royal Commission, 2003, Justice Neville Owen
Changes in own credit and OCI
Many of the arguments listed above show that many people find it difficult to interpret changes in the effect of own credit risk. One possible way to deal with those difficulties is to report those changes in other comprehensive income (OCI). IFRS 9 Financial Instruments already does this in some cases.
Own credit risk in other measurement bases
Some IFRS Standards cover liabilities not measured at fair value but do not specify whether the measurement of those liabilities should reflect own credit risk. Among them are IAS 19 Employee Benefits and IAS 37 Provisions, Contingent Liabilities and Contingent Assets. When those Standards were developed (in the late 90s), there was little focus on own credit risk. If people had been talking about that topic then, I’m sure the Standards would have specified whether to reflect own credit risk.
The IASB now has more tools to address this topic because the Conceptual Framework now contains some discussion of own credit risk and the fulfilment value of liabilities.
Fulfilment value and own credit risk
The Conceptual Framework says:
- the fulfilment value of a liability is the present value of the cash, or other economic resources, that an entity expects to be obliged to transfer as it fulfils the liability. (paragraph 6.17)
- in determining fulfilment value indirectly using measurement techniques, those techniques reflect the same factors as the factors reflected in fair value—but from an entity-specific perspective rather than from a market perspective (paragraph 6.20). Those factors include the possibility that the entity may fail to fulfil its liability (own credit risk). (paragraphs 6.14-6.15)
- cash-flow-based measurement techniques can be used in applying a modified measurement basis—for example, fulfilment value modified to exclude the effect of the possibility that the entity may fail to fulfil a liability (own credit risk). (paragraph 6.92)
- modifying measurement bases may sometimes result in information that is more relevant to users of financial statements or that may be less costly to produce or to understand. But modified measurement bases may also be more difficult for users of financial statements to understand. (paragraph 6.92)
For discussion on selecting a measurement basis, please see Selecting a measurement basis – Accounting Miscellany
In my view, the Conceptual Framework is correct in stating that entity-specific value (as defined in the Framework) conceptually includes own credit risk. A liability is more onerous if the debtor’s credit risk is high than if its credit risk is low.
To take an extreme example, let’s say a company is nearly bankrupt and there is only a 1% chance that it will end up fulfilling the liability. The burden of the liability on the company is only 1%, not 100%. In (an estimated) 99% of the outcomes, it will end up paying nothing. As a result, the company certainly wouldn’t willingly pay more than 1% to be released from the liability—showing that the liability’s entity-specific value is only 1%.
But I think it was also a good idea for the Conceptual Framework to use the following as its example of a modified measurement basis: fulfilment value modified to exclude the effect of the possibility that the entity may fail to fulfil a liability (own credit risk).
The Framework does not require (or even encourage) the IASB to decide to use that particular modified basis. But in view of the arguments cited above, it would not be surprising if the IASB were to decide sometimes to use that modified measurement basis rather than a pure fulfilment value.
Some IASB papers on own credit risk
I include below links to 3 IASB papers discussing own credit risk:
- Discussion Paper Credit Risk in Liability Measurement (2009)
- Discussion Paper Preliminary Views on Insurance Contracts (2008)
- Staff paper for the project Provisions—Targeted Improvements (2022)
Credit Risk in Liability Measurement
In June 2009, the IASB published a Discussion Paper Credit Risk in Liability Measurement Discussion Paper Credit Risk in Liability Measurement (ifrs.org) and accompanying staff paper Staff Paper accompanying Discussion Paper Credit Risk in Liability Measurement (ifrs.org).
Those papers focused on the following questions:
- When a liability is first recognised, should its measurement (a) always, (b) sometimes or (c) never incorporate the price of credit risk inherent in the liability?
- Should current measurements following initial recognition (a) always, (b) sometimes or (c) never incorporate the price of credit risk inherent in the liability?
- How should the amount of a change in market interest rates attributable to the price of the credit risk inherent in the liability be determined?
In October 2009, the Board considered a summary of the responses and decided:
- to stop work on credit risk as a free-standing work stream.
- not to reach a general conclusion on credit risk at this time but instead to incorporate the topic in the Conceptual Framework measurement project;
- not to change the role of credit/performance risk in the definition of fair value as a result of the responses to the DP. The Board had not yet reviewed responses to the exposure draft Fair Value Measurement;
- to consider the application of the fair value definition in every project involving measurements that would otherwise be at fair value; and
- to consider the question of credit risk in every future project involving a current measurement of liabilities that is not fair value.
The IASB’s discussion paper Preliminary Views on Insurance Contracts (2008) discussed own credit risk briefly in paragraphs 229-232 and in more detail in appendix H. That discussion goes into some of the arguments in more detail than is found in this blogpost and than in the Discussion Paper Credit Risk in Liability Measurement.
In that Discussion Paper, the IASB expressed preliminary views that:
- insurers should measure their insurance contracts at current exit value. The Board had not identified current exit value as significantly different from the definition of fair value it was then developing. The Board said that it was not yet in position to determine whether these two notions were the same.
- the current exit value of a liability is the price for a transfer that neither improves nor impairs its credit characteristics.
- in practice, effects of own credit risk on the measurement of insurance contracts would normally be small.
That discussion paper led ultimately to IFRS 17 Insurance Contracts, which requires measurements that do not reflect non-performance risk. The Board concluded that including reflecting own non-performance risk in the measurement of an insurance contract liability would not provide useful information (paragraph BC197 of the Basis for Conclusions accompanying IFRS 17). However, in IFRS 17 there is not much pressure on the question of non-performance risk: including this risk would have simply reduced the ‘fulfilment cash flows’, increasing the ‘contractual service margin’.
A staff paper for the IASB’s October 2022 discusses non-performance risk as part of the IASB’s project Provisions—Targeted Improvements. https://www.ifrs.org/content/dam/ifrs/meetings/2022/october/iasb/ap12a-provisions-discount-rates-nonperformance-risk.pdf
Some IFRS Standards were developed at a time when there was little or no focus on whether liability measurements should reflect own credit risk. There is much more discussion on this topic now, and it seems likely that any major revisions to those or other Standards will reach explicit conclusions on whether to reflect own credit risk.
Fortunately, the Conceptual Framework now provides the IASB with some useful tools. Those tools will will help it decide:
- which measurement basis to adopt for those liabilities.
- whether that measurement basis should reflect own credit risk, or should be modified to exclude (or, indeed, include) own credit risk.