CPA Canada’s IFRS Discussion Group recently discussed the following fact pattern:

• Entity A establishes a long-term incentive plan at the beginning of 20X0 for senior management. Senior management of Entity A will be awarded with a bonus compensation of 10 per cent of their base salary, payable immediately upon achievement of the target EPS growth. The target is a one-year growth in EPS of 5 per cent, calculated based on the first day of 20X0 to the end of 20X1.
• For each 1 per cent of EPS growth achieved above 5 per cent, an additional 1 per cent will be added to senior management’s 10 per cent base salary as bonus compensation.
• Entity A has a historical EPS trend of negative 2 per cent, positive 3 per cent and positive 8 per cent for the preceding three years, prior to 20X0.

Maybe this doesn’t sound particularly “long-term,” but that’s how it’s classified under the relevant standard, IAS 19 Employee Benefits, which makes the distinction based on whether the benefits are expected to be settled wholly before twelve months after the end of the annual reporting period in which the employees render the related services. It requires that long-term employee benefits be accounted for in the same way as defined benefit pension arrangements. In discussing this fact pattern, the group first agreed that (as stated in IAS 19.72) “the probability that the specified event will occur affects the measurement of the obligation but does not determine whether the obligation exists.” In this case the period during which the services rendered in connection with the long-term employee benefit commenced in 20X0 and so, in preparing financial statements for that year, Entity A has a present obligation arising from past events that will be settled through the payment of bonus compensation.

However, IAS 19 doesn’t address how to take such contingencies based on the occurrence of an uncertain future into account in measuring a liability. As a reference point, the group considered the measurement guidance in IFRIC 23 Uncertainty over Income Tax Treatments on reflecting the effect of a tax uncertainty in determining taxable profit or loss and other matters. This directs using either of the following methods, depending on which method the entity expects to better predict the resolution of the uncertainty:

• (a) The expected value method – the sum of the probability-weighted amounts in a range of possible amounts.
• (b) The most likely amount method – the single most likely amount in a range of possible amounts.

Applying those notions by analogy, the liability might best be calculated by ascribing a weighted probability to each of the outcomes, including non-achievement of the target EPS growth, achievement of 5 per cent EPS growth and each incremental percentage in excess of 5 per cent. However, there may be one outcome that is most likely based on historical performance and future predictions of performance – if so, then that outcome might best be used in the calculation. As one group member noted, if one of the outcomes appears much more likely than the others, then the use of that amount probably becomes more appropriate. The most likely amount might, no doubt, be zero – even if so, however, the entity will still need to perform ongoing assessments to determine if the measurement of the amount has changed. And of course, if the item is material, then the entity should likely disclose its estimation process and the related uncertainties.

What’s interesting, perhaps, is that this doesn’t necessarily amount to measuring the liability at its fair value, because that’s not the measurement model imposed by IAS 19. So while zero might be an acceptable measurement under IAS 19 based on its concept of an entity’s best estimates of the variables that will determine the ultimate cost of providing post-employment benefits, it wouldn’t likely align with the IFRS 13 concept of fair value as the price that would be paid to transfer the liability (i.e. presumably a counterparty would demand at least some compensation for assuming it). One might understand why we don’t measure defined benefit liabilities at their fair value, given for example the highly extended settlement periods, but in this not-particularly-long-term fact pattern those considerations don’t really apply. So the question is really a bit of an oddity, arising from one of those spaces where the lines between standards may seem a bit arbitrary. The group threw in a mention along the way of IFRS 2’s share-based payment measurement model, with its thicket of a decision-tree (market conditions, performance conditions, vesting conditions), which might have mainly served to remind the reader that however much the current topic might make your head hurt, it could always be worse.

Anyway, the group closed by noting that this discussion was “to raise awareness about the factors to consider in recognizing and measuring long-term bonus plans that are contingent on future events or performance” and that no further standard-setting action is recommended. In other words, you already have all the information you need, so go for it!

The opinions expressed are solely those of the author