Some time ago, the IASB published a study (link: pdf) examining how terms of likelihood in IFRS were viewed by accounting professionals. While it does provide useful insight, the study only surveyed professional opinions without considering if or how those opinions reflected the guidance.
The list below examines terms of likelihood, adding references to the guidance where pertinent .
The list below is not a verbatim copy. It has been adjusted to better reflect how the terms appear in the guidance.
Also, and not just for the sake of completeness, it compares guidance provided by IFRS and US GAAP.
One side effect of the convergence project is that guidance from one standard can and does influence how guidance in the other is interpreted. This is especially apparent at multi-national entities that often apply one standard at the legal entity (local) level and another at the consolidated/combined (international) level.
|
Term of likelihood: |
IFRS |
US GAAP |
|
Certain |
||
|
While not addressed in any guidance, only past transactions, events or conditions (save one) can ever be 100% likely. Any future occurrence (even taxes), always includes some element of uncertainty. For this reason, the study omitted this term of probability, which is included here for the sake or completeness. While not addressed in any guidance, only past transactions or events (save one) can ever be 100% likely. Any future occurrence (even taxes), always includes some element of uncertainty. Nevertheless, this term of probability is included here for the sake or completeness. |
||
|
Virtually certain |
||
|
Although not a defined term, virtually certain appears at the standard level (i.e. IAS 19, 32 and 37) in that IAS 37.33 provides the most pertinent guidance. IAS 37.33 (edited): ... when the realisation of income is virtually certain, then the related asset is not a contingent asset and its recognition is appropriate. As it represents a level of probability high enough to trigger asset recognition, its likelihood is higher than reasonably certain, which does not.
Hidden
IFRS 3.23A (edited) states:... The acquirer shall not recognise a contingent asset at the acquisition date. However, as IAS 37.33 specifies that a virtually certain asset is not contingent, a virtually certain asset may be recognized in a business combination. Note: the only reason this issue is being brought up is that in the BC to FAS 141r, the FASB states: Accordingly, the IASB concluded that contingent assets should not be recognized, even if it is virtually certain that they will become unconditional or non-contingent. If an entity determines that an asset exists at the acquisition date (that is, that it has an unconditional right at the acquisition date), that asset is not a contingent asset and should be accounted for in accordance with the appropriate IFRS. While correct in stating that contingent assets are not recognized in IFRS, the FASB's description of a contingent asset in IFRS is not consistent with IFRS's description of a contingent asset. While not a defined term, virtually certain appears at the standard level in ASC 450-20-25-3 and the way it is used suggests a likelihood in excess of 90%. ASC 450-20-25-3 (edited) states: The conditions in the preceding paragraph are not intended to be so rigid that they require virtual certainty before a loss is accrued. As ASC 450-20-25-2 requires a loss contingency to be recognized when it is probable, the term virtually certain (certainly) denotes a higher likelihood than probable and (presumably) reasonably assured. Note: while quantified comparably, the result of being virtually certain will likely differ from IFRS. In contrast to IFRS, it makes no difference if a gain contingency is virtually certain, reasonably assured or just plain probable. In no case should it be recognized. IAS 37.33 (edited): ... when the realisation of income is virtually certain, then the related asset is not a contingent asset and its recognition is appropriate. ASC 450-30-25-1: A contingency that might result in a gain usually should not be reflected in the financial statements because to do so might be to recognize revenue before its realization. |
||
| Reasonably certain (assured) | ||
|
While it does show up in the study, the term reasonably assured in not present in any guidance. It does appear in the BCs to IFRS 15 and 17, but those reference are fleeting. It does appear in US GAAP (too many times to count) but the FASB considers, as it explains in ASU 2016-02 BC195, reasonably certain and reasonably assured synonyms. As no IFRS guidance to the contrary exists, there is no reason for this to not be the case in practice. While it figures in the definition of several (lease related) items and appears in IFRS 16 over 10 times, reasonably certain is not itself a defined term. However, since IFRS 16 is a (mostly) converged standard, the conclusion it should be interpreted the same was as under US GAAP (≥ 90%) is justifiable. Nevertheless, the fact that is it not a defined term combined with the results of the study suggest a somewhat more flexible interpretation (≅ 90%) may be warranted under IFRS. The ASC master glossary defines reasonable certainty stating (edited): ... If probabilistic methods are used, there should be at least a 90 percent probability that the quantities actually recovered will equal or exceed the estimate... Consequently, under US GAAP, reasonably certain (assured) should be interpreted as 90% or more likely. Note: ASC 606.BC44 (edited) parenthetically states: ...The term reasonably assured was also used in collectibility thresholds in some parts of U.S. GAAP. However, in this context, the FASB understood that in practice, probable and reasonably assured had similar meanings... This suggests that the standard setters interpret the meaning of the term reasonably assured more flexible that implied by the guidance. Nevertheless, as glossary terms (like the remainder of the ASC) comprise authoritative guidance, they take precedence over views expressed in the BC even if those views are expressed by the standard setters. |
||
|
90% |
90% |
|
|
Why substantially all made the list of terms denoting likelihood is unclear. For example, in IFRS 16 | ASC 842 it is use in the context of risks and rewards, fair value or capacity, while ASC 715-20-55-17 uses it to refer to employees. Nowhere could we find a use case where it denotes probability. Nevertheless, as it did make the list, it may be interesting to examine how practitioners concluded it should represent a likelihood of 83.5% to 90.7%. The most likely reason, IAS 17.10.d and its then counterpart (ASC 840-10-25-1.d) used substantially all and 90% as synonyms. Why substantially all made the list of terms denoting likelihood is unclear. For example, in IFRS 16 | ASC 842 it is used in the context of risks and rewards, fair value or capacity, while ASC 715-20-55-17 uses it to refer to employees. Nowhere could we find a use case where it denoted probability. Nevertheless, as it did make the list, it may be interesting to examine how practitioners concluded it should represent a likelihood 83.5% to 90.7%. The most likely reason, IAS 17.10.d and its them counterpart US GAAP counterpart ASC 840-10-25-1.d used substantially all and 90% as synonyms. Why substantially all made the list of terms denoting likelihood is unclear. For example, in IFRS 16 | ASC 842 it is used in the context of risks and rewards, fair value or capacity, while ASC 715-20-55-17 uses it to refer to employees. Nowhere could we find a use case where it denoted probability. Nevertheless, as it did make the list, it may be interesting to examine how practitioners concluded it should represent a likelihood 83.5% to 90.7%. The most likely reason, IAS 17.10.d and its them counterpart US GAAP counterpart ASC 840-10-25-1.d used substantially all and 90% as synonyms. |
||
| Highly probable |
N/A |
|
|
In the study, highly probable was ranked above reasonably certain (assured). However, this ranking is not consistent with how the term is used in the guidance. In the guidance, highly probable (IFRS) and probable (US GAAP) are synonyms. The conclusion they are synonyms comes from comparing IFRS 15 and ASC 606. In this converged guidance, whenever highly probable appears in the standard, it is mirrored by probable in the topic. That the two terms are synonyms is also acknowledged in the BC to IFRS 15. BC 44: The boards noted that the term ‘probable’ has different meanings under US GAAP and IFRS. Under US GAAP, the term was initially defined in Topic 450 Contingencies as ‘likely to occur’ whereas under IFRS, probable is defined as ‘more likely than not’... BC 210: Consequently, the boards decided that the most appropriate level of confidence would be ‘highly probable’ for IFRS and ‘probable’ for US GAAP as a result of the usage of those terms in existing requirements... The likely reason highly probable ranks as high as it does is that it was introduced in 2014, only one year before the study was published. Further, IFRS 15 only become mandatory four years later and many practitioners fully incorporate new guidance into their thought process only after the standards have become mandatory. In contrast, US GAAP has used probable for many years and its quantification is generally agreed upon by the professional community. Consequently, if the study were conducted today, now that IFRS 15 has been in use for some time, its results would almost certainly bring the likelihood associated with highly probable to the same level as probable in US GAAP. |
||
|
Probable |
||
|
The IFRS master glossary defines probable as more likely than not, making the two terms synonyms. In the past, the IASB used to organize staff days for teachers where academics from all over the world had the opportunity to interact with the IASB staff. During one such meeting in 2012, the topic of probability came up. As the information provided was not particularly satisfying, the attendees asked the staff member in charge "so what are we supposed to tell out students?"
Source: local link. Unfortunately, Google is no longer able to find a link to this file on the IASB's web site. The staff member replied: "the conceptual framework does not quantify the term probable and it is not used consistently throughout the standards." Being academics, most of us had been able to read this for ourselves, so we requested a real answer. Eventually, the staff member relented and said "The conceptual farmwork does not quantify the term probable nor is it quantified the standard level. Nevertheless, in practice most practitioners interpret it as 50% or more, auditors interpret it as 50% or more, regulators interpret its 50% or more, at court it is generally interpreted as 50% or more and, when board members discusses probability, they also generally take it to mean 50% or more. However, and this is very important, you may never, ever tell your students that the IFRS conceptual framework, nor any standard, explicitly states that probable should be quantified as 50% or more. Satisfied with this answer, the discussion moved on to other issues. The ASC master glossary defines probable: "The future event or events are likely to occur." While not a quantification, this definition it is generally understood to mean 75% to 80% likely (link / local link). |
||
|
Likely |
||
|
While the IFRS glossary does not define the term likely, it does use it to define both probable and highly probable, suggesting that, while probable and likely may not be exact synonyms, they are very close to it. The ASC master glossary defines probable: "The future event or events are likely to occur" making the two terms synonyms. |
||
|
More likely than not |
||
|
IFRS defines probably as more likely than not, but does not define more likely than not. Nevertheless, in common usage, it is generally understood to mean over 50%. Note: the study simply refers to "more likely" leaving out "than not." It does not explain why. While US GAAP does not quantify likely or probable, it does quantify "more likely than not" (at the standard level). Specifically, ASC 740-10-30-5.e states: ... it is more likely than not (a likelihood of more than 50 percent) ... |
||
|
Used throughout the guidance, the term expected often expresses probability. However the exact probability it is supposed to express is difficult to pin down. This likely explains why it was excluded from the study. While not a defined term, expected is used throughout IFRS. Unfortunately, the way it is used makes the exact likelihood it is supposed to express difficult to pin down. For example, in the IFRS master glossary, an asset is a resource from which future economic benefits are expected to flow. In IAS 37.10 (definition of liability), expected denotes the probability with which a settlement will result in an outflow. In IFRS 15.95.c, expected denotes the probability that the costs to fulfill a contract will be recovered. Perhaps the best description appears in IFRS 15.53.a where: the expected value is the sum of probabilityweighted amounts in a range of possible consideration amounts. This implies that expected can express any probability, from virtually certain to remote, depending on the probability weights assigned to the various amounts in the range. Note: when a fractional outcome is not possible (i.e. rolling a 6-sided die), most likely amount (IFRS 15.53.b) is applicable instead. While not a defined term, expected is used throughout US GAAP. Unfortunately, the way it is used makes the exact likelihood it is supposed to express difficult to pin down. For example, in CON 8.Ch4.E8 an estimate of uncollectible amounts reduces a receivable to the amount expected to be collected. In ASC 326-20-30-1, an allowance for credit losses is the amount not expected to be collected. In ASC 340-40-25-5, expected denotes the probability that the costs to fulfill a contract will be recovered. Perhaps the best description appears in ASC 606-10-32-8 where: the expected value is the sum of probabilityweighted amounts in a range of possible consideration amounts. This implies that expected can express any probability, from virtually certain to remote, depending on the probability weights assigned to the various amounts in the range. Note: when a fractional outcome is not possible (i.e. rolling a 6-sided die), most likely amount (ASC 606-10-32-8.b) is applicable instead. While used extensively throughout the guidance, While not included in the study, the term expected often express probability. For example, in IAS 37.10 (definition of liability) it expresses the probability with which a settlement will result in an outflow while in IFRS 15.95 IFRS 15.95.c | ASC 340-40-25-5, it denotes the probability that the costs to fulfill a contract will be recovered. It can also combine value and probability as in IFRS 15.53.a | ASC 606-10-32-8.a where it denotes a sum of probability weighted amounts. Note: when a fractional outcome is not possible (i.e. expected value of rolling a 6-sided die), the term "most likely amount" (IFRS 15.53.b | ASC 606-10-32-8.b) is generally used instead. |
Reasonably possible |
|
|
While the term reasonably possible does appear in IFRS, it is generally buried deep in the bowels of the guidance (for example IFRS 13 93.h.ii, IFRS 15.C6.b or IAS 36 134.f) and is nowhere as prominent as in US GAAP, where it even given its own spot in the master glossary. This likely explains why reasonably possible is given a rank of 54.6% to 64.7% while plain old possible only 39.8% to 45%. Is it reasonable to assume adding reasonableness to a possibility will make that possibility more probable than probable? Likely not. The ASC master glossary defines reasonably possible: The chance of the future event or events occurring is more than remote but less than likely. This is a very large range, somewhere between less than 10% to 50%. While no study on how US GAAP practitioners view probability is known to have been conducted, it seems reasonable to assume they would interpret this term as less than 50%, but not by more than half. Note: ASC 450-20-25-1 outlines just three levels of probability: probable, reasonably possible and remote implying that reasonably possible ranges from over 10% to under 75%. However, as discussed in the contingent liabilities of this page, ASC 450-20-25-1 is an old standard well past its best by date, so cannot be interpreted literally. |
||
|
Possible |
||
|
While the study assigned it a relatively high probability (39.6% to 45%), the way the term possible is used in the IFRS glossary suggests that, in reality, it is impossible to assign it any particular probability at all. For example, expected cash flows are defined as the probability-weighted average (ie mean of the distribution) of possible future cash flows while lifetime expected credit losses are defined as the expected credit losses that result from all possible default events ... In both cases, the probability of one possible event could be anywhere from more than 0% to less than 100%. For example, it could be 0.1% likely that all loans in a portfolio will be collectible (one possible outcome), 99.9% likely that at least one will become uncollectible (another possible outcome), 5% likely that 90% of all loans will become uncollectible (another possible outcome), 95% likely that at least 10% will become uncollectible (another possible outcome), etc. Theoretically, the number of possible outcomes can be practically endless. So, and simply for the sake of argument, this schedule assigns it a probability half way between more likely than not and impossible. The way the term possible is used in the guidance suggests that it is impossible to assign it any particular probability. For example CON 8.Ch 3.QC 26 states: A range of possible amounts and the related probabilities also can be verified. For example, it could be 0.1% likely that all loans in a portfolio will be collectible (one possible outcome), 99.9% likely that at least one will become uncollectible (another possible outcome), 5% likely that 90% of all loans will become uncollectible (another possible outcome), 95% likely that at least 10% will become uncollectible (another possible outcome), etc. Theoretically, the number of possible outcomes can be practically endless. |
||
|
Unlikely |
||
|
While used throughout IFRS, the guidance does not outline how the term should be quantified. A good example of how it is used appears in IFRS 15.IE17 (edited): the entity learns that the customer has lost access to credit and its major customers and thus the customer’s ability to pay significantly deteriorates. The entity therefore concludes that it is unlikely that the customer will be able to make any further royalty payments for ongoing usage of the entity’s patent. So, for the sake of argument, this schedule assigns it a probability of around 20%, the ballpark suggested in the study. While used throughout US GAAP, the guidance does not outline how the term should be quantified. A good example of how it is used appears in ASC 606-10-55-109 (edited): the entity learns that the customer has lost access to credit and its major customers and thus the customer’s ability to pay significantly deteriorates. The entity therefore concludes that it is unlikely that the customer will be able to make any further royalty payments for ongoing usage of the entity’s patent. So, for the sake of argument, this schedule assigns it a probability of around 20%, the ballpark suggested in the study. |
||
|
Highly unlikely |
||
|
This term only appears in two standards (IAS 41 and IFRS 9) and neither suggest how it should be quantified. So, simply for the sake of argument, this schedule assigns it a probability of around 15%, the ballpark suggested the study. The term only appears (parenthetically) in ASC 815-20-55-200, which does not suggest how it should be quantified. So, simply for the sake of argument, this schedule assigns it a probability of around 15%, the ballpark suggested the study. |
||
|
Extremely unlikely |
||
|
Like highly unlikely, extremely unlikely also only appears fleetingly in 17.B18 and the BC to IFRS 2. Neither suggest how it should be quantified. So, simply for the sake of argument, this schedule assigns it a probability of around 10%, the ballpark suggested the study. Like highly unlikely, extremely unlikely also only appears in one paragraph (ASC 860-10-55-23), which does not suggest how it should be quantified. So, simply for the sake of argument, this schedule assigns it a probability of around 10% the ballpark suggested the study. |
Remote |
|
|
While used throughout IFRS (for example, IAS 37 uses it 4 times), IFRS does not (unlike US GAAP) define remote nor suggest it represents the lowest possible level of certainty (besides zero). Not that its definition is particularly helpful. Nevertheless, as the study shows, this is how remote is interpreted in practice. The master glossary defines remote: The chance of the future event or events occurring is slight. Somewhat more helpfully, ASC 450-20-25-1 states: ... The Contingencies Topic uses the terms probable, reasonably possible, and remote to identify three areas within that range. This guidance makes it clear remote denotes lowest rung on the probability ladder, just above zero. |
||
|
Not possible |
||
|
The term not possible shows up relatively frequently in the guidance, just not in the context of probability. A good example of how it is used appears in the BC to IAS 38 where the board states: it is not possible to assess reliably the amount that can be recovered from an internally generated intangible asset, unless its fair value can be determined by reference to an active market. When applied to probability, it is comparable to certain in that only the past transactions, events or conditions can be deemed not possible. For example, the collectability of a debt could be deemed remote if the debtor was about to be declared bankrupt. Nevertheless, it would not be impossible until the bankruptcy process was completed. Note (presumably for this reason): local GAAPs / tax laws generally require the bankruptcy process to be completed before a debt can be definitely written off. For this reason, the study omitted this expression of probability, which is only included here for the sake or completeness. The term not possible shows up relatively frequently in the guidance, just not in the context of probability. A good example of how it is used appears in the BC to ASU 2016-10 where the Board observed: it is not possible to appropriately recognize revenue when (or as) the entity satisfies its performance obligation if the entity does not first understand whether its promise in granting the license requires continued performance by the entity. When applied to probability, it is comparable to certain in that only past transactions or events can be deemed not possible. For example, the collectability of a debt could be deemed remote if the debtor was about to be declared bankrupt. Nevertheless, it would not be impossible until the bankruptcy process was completed. Note (presumably for this reason): local GAAPs / tax laws generally require the bankruptcy process to be completed before a debt can be definitely written off. Nevertheless, it is included here for the sake or completeness. |
||
