by John Hughes
I remember reading an article a few years ago criticizing a New York Times book reviewer for writing about a disproportionate number of short books. With that in mind, and wary of being accused of compiling a blog that muses on too many comparatively easy topics (not that it’s happened yet, happy to say), I’m writing again today on embedded derivatives – and you know that’s hard! Here’s a disclosure from Rona Inc.:
IAS 39 and CICA 3855 both provide examples of situations where the economic characteristics and risks of an embedded derivative are closely related to those of its host contract, and therefore the embedded derivative isn’t accounted for separately. In both instances, these examples include an embedded derivative in a contract to purchase or sell a non-financial item for which the price is denominated in a foreign currency, provided the contract isn’t leveraged, doesn’t contain an option feature, and requires payments denominated either in the functional currency of any substantial party to the contract, or the currency in which the price of what’s being acquired or delivered is routinely denominated in commercial transactions around the world. Under IFRS, but not under Canadian GAAP, the payments can also be denominated in a currency commonly used in contracts to purchase or sell non-financial items in the economic environment in which the transaction takes place.
The main purpose of the requirements for embedded derivatives is to shut down the possibility that an entity’s interests in derivatives, because they’ve been inserted into or aggregated with other instruments, might escape recognition and measurement; in turn, the point of the various exemptions for situations where the derivative and the host contract are “closely related” is that those derivatives are more likely to be inherent to the purpose and substance of the contract, and therefore not analogous to stand-alone instruments. The exemption described above acknowledges the inherent complexities of contracting in a global environment, causing entities to take on currency exposure as an inevitable aspect of doing business, not with any other speculative or financial engineering-related motive. The embedded derivative may not actually be ‘closely related’ to the host contract, except by enforced marriage - in a neutral environment, ensuring a steady supply of widgets would have nothing to do with taking on unwanted exposure to the Euro. But in the real world, things intertwine, and so applying the separation concepts rigorously would be “burdensome.” IAS 39 points out for instance that ”entities domiciled in small countries may find it convenient to denominate business contracts with entities from other small countries in an internationally liquid currency…rather than the local currency of any of the parties to the transaction.”
The difference arises in that third situation under IFRS, where the contract requires payments denominated in a currency commonly used in contracts to purchase or sell non-financial items in the economic environment in which the transaction takes place. As Rona explains, CICA 3855 allows (but doesn’t require) treating such contracts as a single instrument. The basis for conclusions to CICA 3855 explains it like this: “US GAAP does not permit this. However, the AcSB decided to permit this treatment because non-US entities more frequently use currencies other than the local currency of one of the parties to the contract when undertaking transactions. A Canadian entity wishing to conform to US GAAP as well as Canadian GAAP need not adopt this policy.”
In other words, I suppose, the US view of the world doesn’t really envisage itself ever being compelled to transact in a particular foreign currency, unless it’s inherent to the specific counterparty or market – the additional IFRS criterion just doesn’t have much resonance there. But Canadian entities have to take their opportunities as they find them, rendering a greater likelihood that they’ll be forced into embedded currency derivatives without having gone looking for them. I don’t know whether Rona chose to separate them out under Canadian GAAP because of a desire to conform with US GAAP – as far as I can see, the company isn’t listed in the US and wasn’t preparing a formal GAAP reconciliation…maybe they just thought it was the right thing to do.
The requirements relating to embedded derivatives have always seemed to many people, I think, as mostly an encumbrance, adding complexity without bringing accompanying enlightenment. The accounting standards in this area are very hard to read (and write about!), in that it’s even easier than usual to be overwhelmed by the criteria and the examples and the counter-examples and to lose track of what it’s all for. Unfortunately, the various accommodations based on good intentions – while no doubt justified from a practical standpoint – only add to the sense of this being a largely arbitrary cookbook: the measurement volatility attached to a currency derivative that wasn’t separated because it would have been “burdensome” might be much greater than that attached to those other embedded derivatives for which no escape route was offered. In particular, making any aspect of this into a choice, looked at rationally, seems bizarre, although of course it’s only a subset of the weird contortions into which financial instrument accounting has twisted itself over the years. IFRS at least removes this one specific oddity, but doesn’t do much – even after the introduction of IFRS 9, if that ever happens - to change the existing playing field for separating embedded derivatives from host contracts that aren’t assets. No doubt that reflects an “if it ain’t broke don’t fix it” philosophy, and I really don’t want to suggest this aspect of accounting is broken exactly…it’s just that if hardly anyone can follow the tune, you wonder about the effectiveness of the piano.
The opinions expressed are solely those of the author.