Non-IFRS measures – the Australian approach

Posted on February 22, 2012 by admin

by John Hughes

A recent interesting article from Australia talks about the Australian Securities and Investment Commission’s (ASIC’s) view on “non-IFRS profit information” in financial statements. Here’s the summary:

ASIC says that financial information which is prepared other than in accordance with the accounting standards should not be included in financial statements. It is prepared to make an exception for:

  • information required by law;
  • details of a breach of a lending covenant that is determined by reference to a non-IFRS financial ratio;
  • an explanation of director and executive remuneration that is determined by reference to something other than IFRS profit figures.

Non-IFRS financials can also be included in the notes to the financial statements, but only in what ASIC describes as the “rare circumstances” where the non-IFRS information would be necessary to give a true and fair view of the company’s financial position.

In addition, non-IFRS profit figures may not be included as line items or subtotals in the income statement, or presented in additional columns of financial statements.

I’ve written several times (here for instance) about the Canadian approach to what we call “non-GAAP financial measures,” which has continued to evolve as IFRS takes greater hold. It sounds like this has been a problem area in Australia, with many companies supplementing their IFRS-compliant numbers by providing “underlying profit” numbers, calculated on a variety of bases, and serving in most cases of course to increase profits or to turn a loss into a profit. ASIC will still allow these outside the financial statements it seems, within a disclosure structure similar to that long-required by Canadian regulators.

Insofar as it relates to financial statements, the Australian prohibition seems broader than that in Canada. Our CSA Staff Notice 52-306 defines a non-GAAP financial measure as “a numerical measure of an issuer’s historical or future financial performance, financial position or cash flow, that does not meet one or more of the criteria of an issuer’s GAAP for presentation in financial statements, and that either (i) excludes amounts that are included in the most directly comparable measure calculated and presented in accordance with the issuer’s GAAP, or (ii) includes amounts that are excluded from the most directly comparable measure calculated and presented in accordance with the issuer’s GAAP.” This doesn’t seem to encompass, say, a number computed with reference to a covenant compliance calculation and presented as such in the notes, because that kind of number doesn’t seem to be a measure of financial performance, financial position or cash flow. But as cited above, this would fall within the broader Australian prohibition, absent the specific exemption they provide for it.

Would the Australian approach have any impact if introduced into Canada? For some companies, perhaps. Financial statement notes do occasionally include other numerical information which might trip over the ASIC definition (frankly, I imagine auditors would be happy to see most of this information go, thus removing any pressure on them to audit it). For instance, some companies reporting revenue on a net basis might also disclose a number for “gross” or “system-wide” sales, without actually representing that (or perhaps even knowing whether) this broader number is calculated in compliance with IFRS. In other ways, ASIC’s definition might just open up a bunch of interpretation issues. For example, IFRS 8 requires disclosing segment information on the basis of the measures reported internally, even if these don’t conform with IFRS; any such differences are disclosed and then dealt with in reconciling the total numbers. Presumably because IFRS in this case requires disclosing “non-IFRS” information, disclosing the non-IFRS information is consistent with IFRS. Likewise, an entity might define and monitor its “capital” in a way that reflects, say, economic value added concepts rather than IFRS concepts. Because IAS 1 requires disclosing summary quantitative data about what an entity manages as capital, disclosing such non-IFRS information would presumably again be consistent with IFRS.

ASIC’s third exception, for “an explanation of director and executive remuneration that is determined by reference to something other than IFRS profit figures,” seems to reflect their local statutory requirements rather than those of IFRS. IAS 24 requires disclosing summary information about compensation paid to key management personnel, but doesn’t require explaining them. That job falls into the realm of the information circular, and I imagine accountants and auditors generally try to stay as far away from it as possible.

For better or worse, it seems plain investors don’t always view IFRS profit or loss as the most relevant (or in the worst case, even as a relevant) measure of performance – they also focus heavily on cash flow and EBITDA and various adjusted numbers. It’s futile to argue they shouldn’t do this – I tend to put it in terms of the hopelessness of trying to save people from themselves – but there’s no question that an investor’s desire for greater simplicity can play right into an entity’s desire to distract him or her from various problem areas. You can’t blame regulators for flashing a big amber light over all this, even if it starts seeming a bit neurotic, and even if they know people mostly drive straight through regardless.

The opinions expressed are solely those of the author

Posted in Australia, IFRS, Non-GAAP financial measures | Leave a comment

Operating lease revenue – systematically representative?

Posted on February 16, 2012 by admin

by John Hughes

In August last year I wrote about two of the three separate differences reported by Easyhome Ltd. in the area of revenue recognition, and said I’d return to the other item another day. Then I forgot all about it until now. Here it is:

CICA 3065 required including rental income from an operating lease “in the determination of net income over the lease term on a straight-line basis unless another systematic and rational basis is more representative of the time pattern of the user’s benefit.” It defined the lease term as “the fixed non-cancellable period of the lease,” extended by (among other things) periods covered by bargain purchase options, or by extensions at the lessor’s option, or for which not renewing the lease would impose a sufficiently large penalty to make renewal seemingly assured. In a situation where the lease term was entirely cancellable, this would seem to push companies toward considering what other “systematic and rational basis” would make most sense for recognizing income. But it’s no surprise if the lease term would still have ended up in many cases as the main determinant of the recognition period. Using the lease term would keep things systematic at least; whether it would also be rational and most representative of the time pattern of the user’s benefit would depend on actual cancellation experience.

IAS 17 has a similar basic requirement - to recognize income from operating leases on a straight-line basis over the lease term, unless another systematic basis is more representative (IFRS doesn’t mention the criterion of being “rational” – I suppose that word ought to be redundant in the overall context). One difference: IAS 17 refers to “lease income from operating assets” whereas CICA 3465 referred to “rental revenue from an operating lease.” It’s at least possible that under Canadian GAAP, some might have regarded income characterized as “processing fees” as being different from the core rental revenue from the lease, and subject to a different accounting analysis (although I don’t know how often this would ultimately lead to a different conclusion about the appropriate recognition pattern): that is, the IFRS concept of “lease income” might seem broader and more likely to capture all such increments and add-ons (other than receipts for services provided such as insurance and maintenance, which IAS 17 specifically excludes from the definition). Easyhome’s disclosure doesn’t seem to be emphasizing that point though.

Anyway, IAS 17 defines the lease term as “the non-cancellable period for which the lessee has contracted to lease the asset together with any further terms for which the lessee has the option to continue to lease the asset, with or without further payment, when at the inception of the lease it is reasonably certain that the lessee will exercise the option.” This appears to capture a broader range of situations than Canadian GAAP did – for instance, a lessee might be “reasonably certain” of going beyond the lease term just because of the attractiveness of the asset, the lack of reasonable alternatives, and suchlike. Under IFRS this might in theory be sufficient to base the lease term on that longer period; under Canadian GAAP, in the absence of bargain purchases, lessor options or other such identifiable factors, it wouldn’t ever be.

But that difference isn’t relevant to Easyhome, because its leases don’t seem to have any non-cancellable period. As in all cases like this, I’m just using the published disclosure as a springboard to muse out loud: I don’t know of course the specific analysis that drove the company to make this change. But it wouldn’t be surprising if it mainly reflects a sense of IFRS as bringing greater analytical rigour to this kind of estimation process. Using the lease term might be systematic, but if it’s systematically inaccurate, the spirit of IFRS (so to speak) pushes you toward using the best available estimates and then prospectively amending them as necessary. To take a random example from a different standard, IAS 16 has a more active concept of the depreciation calculation than CICA 3061 did, placing a sharper focus on whether the method used reflects the consumption of the asset’s future economic benefits, and on changing the method when that pattern of consumption changes.

So that’s that. Now of course, the IASB’s ongoing leasing project would shake all of this up considerably, with a different approach to lessor accounting and a revised definition of the lease term as well. I’ll return to all of that in the future, no doubt more than once. But I suppose a lot of companies like Easyhome will be hoping the IASB retains its plan of allowing short-term leases an exemption from the proposed model for lessor accounting, so that they continue to recognize lease income over the lease term on a systematic basis. Well, actually, I think a lot of companies are wishing the IASB would just forget about the project altogether, but that’s probably too optimistic…

The opinions expressed are solely those of the author.

Posted in IFRS, Leasing, Revenue | Leave a comment

IFRS 10 – is it a subsidiary, or are they just friends?

Posted on February 10, 2012 by admin

by John Hughes

Over a year ago, I wrote about the role of potential voting rights (embodied within convertible debt, warrants and so forth) in assessing whether one entity controls another, noting the Exposure Draft of the proposed amendments to IAS 27, as it was at the time, would change things considerably. Well, that project’s been finalized now, as IFRS 10, and that’s how things turned out: it’ll certainly generate a different conclusion in some situations where this issue applies. But that’s just one piece of it, because IFRS 10 (effective for annual periods beginning on or after January 1, 2013) is a gift basket of greyness. That’s not a criticism – the core question of whether one entity controls another is no doubt often inherently grey. But it’s not hard to anticipate some practitioners, once they apply the new standard to their marginal situations, pining for some old-time bright lines and arbitrary decision trees.

I’ve already signalled some of the issues here, but I thought it was worth returning to the well at least one more time (you can also access some CICA resources on the new standard here). Here’s a quick overview. An investor controls an investee if and only if these three criteria apply: (a) it has power over the investee; (b) it has exposure, or rights, to variable returns from its involvement with the investee; and (c) it has the ability to use its power over the investee to affect the amount of those returns. The key criterion for most of the potential problems I think (leaving aside problems arising from specifically structuring something to achieve a specific outcome), will be the first – whether an investor has power over an investee. Re the second criterion: an investor may have power over an investee without being exposed to variable returns if, for example, it’s limited to a fixed fee, with the gains or losses flowing elsewhere, a situation presumably most commonly applying for fund managers or similar entities. Re the third criterion: an investor may have power over an investee without being able to use that power to affect its returns if, for example, the investee’s activities and returns are predetermined, changing only if particular circumstances arise; as far as I know, this could only apply to a special purpose entity of some kind.

So on that first criterion, an investor has power over an investee when it has existing rights, giving it the current ability to direct the so-called “relevant activities” that significantly affect the investee’s returns (it may take a while for all of us to get used to the new terminology of “power” and “rights” and “relevant activities”). It won’t often change the existing answer where an investor holds more than half the investee’s voting shares, unless some specific restriction applies or, perhaps, someone else has potential rights they could exercise to snatch the investor’s power away. However, IFRS 10 says an investor with less than a majority of the voting rights in another entity still has sufficient rights to give it power over that entity, when it has the practical ability to direct unilaterally the relevant activities. The more voting rights the investor holds compared to anyone else, and the greater the number of other vote-holders needing to act together to outvote it, the more likely it has that practical ability. This might be supported by looking at voting patterns at past shareholder meetings, or other applicable information.

IFRS 10 goes on: “In some circumstances it may be difficult to determine whether an investor’s rights are sufficient to give it power over an investee. In such cases, to enable the assessment of power to be made, the investor shall consider evidence of whether it has the practical ability to direct the relevant activities unilaterally.” It provides a list of factors to which “consideration is given,” which in combination with other rights and indicators may provide relevant evidence. This includes fairly obvious items such as holding veto power, or the ability to appoint key management. But the list also includes more ambiguous situations, for instance where the investee’s key management personnel, or a majority of members of its board or other governing body, are related parties of the investor. IFRS 10 then provides “indications that the investor has a special relationship with the investee, which suggests that the investor has more than a passive interest in the investee,” and which in turn might tip the conclusion. These indications include situations where “the investee depends on the investor for critical services, technology, supplies or raw materials” or where “a significant portion of the investee’s activities either involve or are conducted on behalf of the investor.” But of course, plenty of unrelated companies have relationships evidencing one or more of these indicators, just as a matter of arm’s length negotiation, commercial necessity, past history and so forth.

The standard does make it clear that in the absence of any other rights, the fact of an investee being economically dependent on an investor doesn’t by itself lead to the investor having power over the investee. But in situations where one entity has a non-controlling equity interest of any magnitude in another, and there’s a lot of operational interactivity between the two, IFRS 10 may require quite a taxing analysis of why things exist as they do. Is it a sign of control, or just business?

The opinions expressed are solely those of the author.

Posted in Consolidation, IFRS | Leave a comment

Quiz time!

Posted on February 4, 2012 by admin

by John Hughes

I was looking at a website “designed to make learning International GAAP a lot easier and cater towards your own learning style.” The author is anonymous, but provides the following background: “When I was studying for my accounting exams I found it quite difficult to study the different International Accounting Standards that are out there. Most notes were too detailed and didn’t lend well to quick glances. I also had nowhere to practice what I had learned once I had studied the information.” The site includes notes, “mind maps” and quizzes on various standards.

It’s a nice-looking website and obviously well-intentioned; I don’t mean in any way to criticize the execution of it. But I took one of the quizzes, and it tended to reinforce my sense that there’s a problem in the premise: IFRS can’t really be made “a lot easier” to learn, and in fact, I’d argue, shouldn’t be. It’s a specialized, nuanced field of practice, necssarily “hard” to learn, at least with any degree of depth, which is the only reason so many people can make good money from it. If that sounds like mere knee-jerk protectionism, I’ll illustrate further. The quiz I took was on IAS 8, Accounting Policies, Change in Accounting Estimates and Errors, consisting of sixteen multiple choice questions. By my assessment, at least six of the answers are…not wrong exactly, but let’s say incomplete. That is to say, the quiz tests an ability to memorize the exact words of the standard rather than to grasp what they actually mean, or ignores potential complexities that could make other answers equally or more valid. It might be helpful in “learning” IAS 8, but not in promoting a vibrant, practical sense of why it matters, or what to do about it in what we accountants amusingly think of as “real life.”

Here’s an example: An accounting policy can be changed only if:

  • a) It results in showing more profit for the entity
  • b) There is a change in the Top Management
  • c) The reporting period is twelve months
  • d) It is required by an IFRS
  • e) None of the above

The designated answer is (d). But this implies accounting policies can only be changed if required by an IFRS, and ignores the broader acceptability of changing an accounting policy if the change results in the financial statements providing reliable and more relevant information. A more technically correct answer would have been (e), indicating that even (d) doesn’t convey a complete sense of the issue.

Here’s another one:  When a change in accounting policy is applied retrospectively then the change shall be:

  • a)      Effected in the income statement of the current reporting period
  • b)      Adjusted in the equity
  • c)      Effected in the income statement of the previous reporting period
  • d)      Disclosed in the notes to the financial statements without making adjustments in the financial statements
  • e)      Any of the above

The designated answer is (b). But that doesn’t completely capture the issue; in fact, you could almost argue (e) would again be better. A retrospectively applied change in accounting policy would typically affect opening equity (b) and the comparative income statement (c). But it also entails that the current income statement differs from what it would have been under the previous policy (a), unless that is the change in policy – although sufficiently material to be disclosed (for instance because of its anticipated impact on future periods) – doesn’t materially affect any of the previously-reported numbers (d).

Another question posits: “a company provides for bad debts at the rate of 2% of the sales of that period. With effect from 2012, it has decided to change it to 3% of sales. The sales for 2011 – $100000 and sales for 2012 – $200000. In the financial statements of the year 2012, which of the following treatments is appropriate.” The designated answer is to apply the change prospectively from 2012, reflecting that a change in the estimate for bad debts is typically a change in accounting estimate. But it seems reductive to leave it there, without acknowledging that not all estimates are created equal. The 2% used in previous years could theoretically reflect an error – an estimate made in willful disregard of the circumstances – or the proposed 3% rate might be intended to create a cookie jar for future periods rather than responding to the economic facts.

Another answer directs that “if an error is noticed in a financial statement, then such error shall be….corrected prospectively from the financial statements authorized for issue after their discovery.” This seems to misapply the meaning of “prospectively,” but also misses an opportunity to flag how the IFRS requirements aren’t the whole story here – a practitioner would also need to consider regulatory requirements, audit implications and so forth. And yet another states that “Accounting policies across various reporting periods…shall not be consistent,” which on the face of it seems incorrect, or at least would require much more explanation of whatever subtle sense in which it might be correct.

Again, it’s not that the website is “bad,” relative to its intended purpose: I’m sure many will find it helpful in studying for their exams. But by its nature, it seems to embody a grim and bothersome gap between the data required to pass exams and the knowledge required to do the job even semi-competently…

The opinions expressed are solely those of the author.

Posted in Accounting policies, IFRS | Leave a comment

Cash-generating units – the start of the conversation

Posted on January 29, 2012 by admin

by John Hughes

Winpak Ltd. is just one of numerous Canadian companies reporting the following kind of adjustment within their transition disclosures:

In applying the IAS 36 concept of cash-generating units, I’ve found it’s easy to become swept up in the technicalities, and to forget the underlying purpose of the exercise. The definition seems plain enough: a cash-generating unit is “the smallest identifiable group of assets that generates cash inflows that are largely independent of the cash inflows from other assets or groups of assets.” The standard provides a few examples to flesh this out, such as a bus company providing services under a contract with a municipality, under which it’s compelled to maintain five designated routes. Although the assets relating to and cash flows derived from each of the routes can be identified separately, they nevertheless form a single CGU “because the entity does not have the option to curtail any one bus route.” So if one of the routes operates at a significant loss, the entity doesn’t recognize an impairment loss for the assets attaching to that route.

It’s worth mulling over this example a bit. What would be the harm, you might say, of recognizing a loss on that unprofitable route, if by not doing that, some of your assets are recorded at a knowingly irrecoverable amount? Well, in that individual case, it’s hard to say there’d be any harm to it actually. But the example’s clarity rather undermines its usefulness. A more challenging situation would be a factory, encompassing a building and an array of machines and equipment and other assets. Some of those assets plainly don’t generate a cent – the meeting room table for instance, or the window frames, or the roof. But if you took all those away, eliminating the necessary elements of a functioning work space, the assets that do generate the dollars couldn’t function. It’s the equivalent of needing a village to raise a child: if the village works as a whole, you don’t worry about the unproductive contribution of a few individual buildings.

It follows from the bus company example that if the entity did have the option to curtail the loss-making route, then it would recognize the impairment loss on that route. Of course, it’s easy to make up circumstances in which the entity would never exercise that curtailment option, even if it existed. For example, the loss-making route might serve a high-visibility part of town, and curtailing the route might destroy the bus company’s standing with the municipality when the contract came up for renewal. If the other four routes were sufficiently profitable, then maintaining the loss-making route might be a strategic investment, as valuable to the company’s overall direction as the assets specifically attached to the money-making routes. Some Canadian companies have certainly recognized additional impairment losses under IFRS on assets that they looked at in a similar way.

Of course, it would be hard for the standard to make room for such potentially subjective considerations. But there’s always materiality. One of the accounting firm texts says this about retailers: “For the majority of modern multi-site retailers, some level of aggregation of sites is normally appropriate. A larger grouping can be treated as a cash-generating unit or each site can be taken to be a cash-generating unit, though in the latter case a pragmatic view of aggregation may be taken on grounds of materiality…In some circumstances, it may be impractical (or at least costly) to prepare detailed cash flow forecasts for each individual site. Furthermore, forecasts may, to some extent, be based on macro-assumptions about factors that affect larger groupings in a similar way.” Although no doubt intended to be helpful, this seems to me to cloud the issue considerably, by suggesting a generous degree of pragmatism and practicality that preparers (or at least their auditors, including auditors from that same firm) might not be as willing to apply in practice.

It’s certainly an advance, I think, that IFRS reduces the ability to avoid recognizing the consequences of loss-making assets by applying some fuzzy claim about (say) management regarding those assets as belonging with other (profitable, but entirely unrelated) assets within a single operating division. But at the same time that IFRS brings such losses out into the open, it may also force out other write-downs that don’t represent losses at all, within management’s way of looking at things. Of course, many entities choose to focus the attention of their users on key performance indicators that exclude these impairment write-downs (among other things), and they can explain the surrounding circumstances in their MD&A. And the best way to do that is to talk in real-world terms about what the losses actually mean, and what they don’t mean, taking the concept of a CGU as just the starting point for the conversation.

The opinions expressed are solely those of the author.

Posted in IFRS, Impairment | Leave a comment

The SEC’s analysis of IFRS in practice: something for everyone

Posted on January 23, 2012 by admin

by John Hughes

I’ve already made some comments about the SEC’s staff paper, An Analysis of IFRS in Practice, generated after examining the most recent annual financial statements of 183 companies domiciled in 22 countries, and setting out an array of apparent deficiencies, ambiguities and opportunities for improvement. But I think it’s worth returning to the subject one more time. After all, regulators don’t tell us that often how they look at things, and the SEC paper could be used as a kind of supplementary checklist, nudging preparers and auditors to tread particularly carefully in the areas it mentions, if they want to avoid receiving a comment letter on those issues. Of course, Canadian regulators don’t look at things in exactly the same way as the SEC, but presumably there’s enough overlap that the paper is useful for our domestic purposes too (not least, of course, because Canadian regulators have no doubt gone through it in detail and extracted some lessons).

An immediate problem in what I just said though is that it’s the kind of attitude that prioritizes petty compliance over materiality: you might be confident a particular disclosure isn’t material, but because its absence could be questioned by regulators, you make the disclosure anyway, rather than take the chance of having to explain it later. I must admit I succumb to this caution myself all the time. It’s entirely misplaced – why worry if regulators raise a comment, as long as you’re comfortable you have the right response to it? – but I think many people tend to regard any kind of communication from the OSC or others as an implicit adverse judgment, and one they want at all costs to avoid, regardless that it means cluttering up the statements.

The SEC is clearly aware of this problem: “The Staff does not intend to suggest that disclosures in these instances (i.e. the instances where they had unresolved questions about the statements) were necessarily deficient or that the disclosures should have been prepared with the purpose of communicating to a regulator the manner in which a company complies with a set of accounting standards. The Staff recognizes that financial statements are intended to facilitate investor decision-making, and additional information that would have benefited the Staff in this analysis may be of less incremental value to an investor.” It’s an important point, but one potentially overshadowed by the many detailed issues listed. For example, on the subject of inventories,, the report observes: “Some companies did not disclose the amount of reversals of write downs during the period and the circumstances that led to the reversal, as required by IFRS.” This seems like an obvious area in which one ought to be able to assume that an absence of disclosure simply means there’s nothing to say. But since the SEC mentions the issue, it’ll just encourage more companies to ploddingly specify that the amount of reversals of write-downs recognized during period was nil (comparative period – nil). The paper has many things along these lines.

It also muddies the waters in various places by seeming to comment on how disclosures might improve on what IFRS requires. For instance, in the context of property, plant and equipment, it says: “Some companies indicated that costs incurred during a start-up period were capitalized but did not describe the nature of costs capitalized or the time span of the start-up period. Such disclosures could help facilitate investor comparisons of financial statements, as the details of cost capitalization may vary across companies.” Well, I suppose they could, although it’s hard to see how an investment decision could ever turn, even marginally, on such trivia. But regardless, IFRS doesn’t require providing such information, so it seems like an arbitrary observation at best. Even geekier is the complaint that only two companies “provided additional disclosure regarding their application of the initial recognition requirements of IAS 12 to finance leases and the provision for asset closure and restoration costs.” I can see how that would ruin someone’s day…

In a somewhat more relevant vein, the paper makes a number of observations about the use of sub-totals on the income statement (they counted up eighteen different kinds). Here again, these observations don’t all necessarily seem to be getting at deficiencies in how entities applied IFRS, but it’s plainly an area the SEC staff doesn’t like and won’t be able to resist asking questions about (not that we didn’t know this already). Canadian regulators have spent some time on this aspect of things too.

Many of the other issues are long-established weak spots under Canadian GAAP. Over the last decade, I think Canadian regulators may have commented just about annually on the weakness of accounting policy disclosures about revenue, but it never really gets much better. No surprise to see this in the SEC report as well. Anyway, despite any reservations, there’s probably enough material in there to jog every practitioner’s thought processes in some respect at least.

The opinions expressed are solely those of the author

Posted in IFRS, Securities Regulation, United States | Leave a comment

The revenue exposure draft – muddied principles?

Posted on January 17, 2012 by admin

by John Hughes

The IASB has issued a revised exposure draft on revenue recognition, open for comment until March 2012. The document illustrates, I think, the hopelessness of any claim they might make to an elevated relationship with “principles-based” accounting. I don’t really mean that as a criticism. There might be something elementally harmonious to the core concept of double-entry bookkeeping, every debit yielding a yin-to-the-yang credit, but once you start building on that, the harmony can hardly avoid becoming a cacophony. There’s nothing self-evident or inevitable about anything in accounting, any more than there is to the laws of democracy or language or anything else that governs us. It’s just compromise and consensus, sometimes reflecting the best of us, sometimes not.

In particular, I was looking at how the new draft expands on this key criterion: “An entity shall recognize revenue when (or as) the entity satisfies a performance obligation by transferring a promised good or service (ie an asset) to a customer. An asset is transferred when (or as) the customer obtains control of that asset.” The original exposure draft, issued in 2010, contained something very similar, and went on to describe various ways in which the fact of the customer’s obtaining control over the asset might be evidenced: an unconditional obligation to pay, legal title, physical possession, or the customer-specificity of the asset. The intention, in broad terms, was to limit the recognition of assets not meeting the characteristics of the conceptual framework, created by applying overly subjective notions of when revenue has been earned. The IASB knew, no doubt, that this original concept would stir things up a bit.

Well, it stirred things up more than they expected, to the tune of over 1,000 often unenthusiastic comment letters. For example, as reported in the basis for conclusions document: “many respondents in the construction industry were concerned that they would be required to change their revenue recognition policy from using a percentage of completion method to a completed contract method (on the basis that the transfer of assets occurs only upon transfer of legal title or physical possession of the finished asset, which typically occurs upon contract completion).” So the new exposure draft now contains an alternate notion (which I’m condensing here) of when an entity can view itself as having transferred control of a good or service over time (and therefore as having satisfied a performance obligation, allowing it to recognize revenue):

  • “the entity’s performance does not create an asset with an alternative use to the entity…and at least one of the following criteria is met: (i) the customer simultaneously receives and consumes the benefits of the entity’s performance as the entity performs;(ii) another entity would not need to substantially re-perform the work the entity has completed to date if that other entity were to fulfil the remaining obligation to the customer… (iii) the entity has a right to payment for performance completed to date and it expects to fulfil the contract as promised.”

In other words, as it strikes me, in situations where one can’t possibly apply even a vaguely principles-based view of whether control over a good or service has been transferred to the customer, the IASB executes a rather glorious little leap to the other side of the stage and says: “Let’s look at this whole thing another way. We don’t necessarily mean what we say about transferring control. We just mean, is there enough of a thing there, enough completed work that you feel pretty good about getting paid for, that it’s not going to create big issues in hindsight if you recognize the revenue at this point.” Which doesn’t sound much more rigorous to me than the standards we have now. Of course the extra words in the proposed new standard, compared with the current IAS 18, will no doubt squeeze out the more aggressive existing applications. But the cleansing power of extra words has nothing to do with the power of principles.

None of the IASB members dissented from the exposure draft, but one of the FASB members did dissent from the parallel US GAAP project, saying among other things “he believes that the proposed model has introduced exceptions that permit revenue to be recognised in a manner that is inconsistent with the core principle on which the entire standard is purportedly based” – in his example: “it permits an architectural design firm to recognize revenue before the completion of its design drawings and the delivery of its unique work product to a specific customer as long as the design firm has the right to payments for design activities undertaken to date.” Which sounds good for the design firm, if not necessarily for the users of the statements (at least not, again, if you think the “transfer of control” criterion has anything going for it).

It would have been satisfying if this project looked likely to introduce some meaningfully greater objectivity to matters of accounting for revenue. As a key performance measure – often emphasized (rightly or wrongly) more than the bottom line – it needs no less than that. But as of now, and admittedly this is something of an initial reaction, I’m not so sure it’ll happen…

The opinions expressed are solely those of the author

Posted in IFRS, Revenue | Leave a comment

Related party transactions – a fair exchange?

Posted on January 11, 2012 by admin

by John Hughes

Here’s a disclosure from Acasti Pharma Inc.:

I wrote over two years ago about this broad area of difference between IFRS and Canadian GAAP. To recap briefly, the core principle of CICA 3840 was to measure related party transactions at the carrying amount, with two exceptions. One of these covered transactions in the normal course of operations, which were measured at the exchange amount (the agreed-to consideration) if they had commercial substance – this covered legal fees provided by directors and suchlike. Assuming these exchange amounts represent normal commercial billing practices, it’d be hard to argue why you’d rationally measure them any differently under IFRS. The second exception covered transactions not in the normal course of operations (such as a transfer of intellectual property). These were also measured at exchange amount if the change in the ownership interests of the item transferred or benefit of the service provided was substantive (usually meaning at least a 20% change in the ownership interests) and the exchange amount was supported by “independent evidence”. This independent evidence might come from independent appraisals, comparable recently quoted market prices, comparable independent bids on the same transaction, or comparable amounts of similar transactions actually undertaken with unrelated parties. If those two conditions weren’t met (and it was somewhat in the company’s power, perhaps, to determine whether it had adequate “independent evidence” or not) then these non-normal course transactions were measured at their previous carrying amount in the financial statements of the entity they came from. But since IFRS doesn’t address such non-normal course exchanges, most people conclude you should measure them using the same principles you’d apply to any other transactions. And since IFRS 1 doesn’t allow any exemption for this area, this would have to be implemented retrospectively on adopting IFRS.

I think I expected this issue to come up more often in the IFRS transition disclosures than it actually did. I thought a significant number of Canadian entities might have assets within their financial statements that were originally measured at a carrying amount, because of some past related party intrigue. But then, in many cases, those carrying value elements may have been written down by now as impaired, or amortized down to nothing over time, so the issue wouldn’t have presented itself as being material to the transition exercise. Acasti highlights the possible problem with this though – if those past transactions hadn’t been measured at (say) an artifically low carrying value in the first place, maybe there would still be something on the books. The different IFRS approach to impairment in IAS 36 might have consequences too – for instance, if the assumptions underlying a previous impairment loss subsequently changed, then that loss might be wholly or partly reversed at the IFRS transition date. This might not suggest itself as a big issue if the assets in question were only valued at a nominal carrying amount in the first place, but again, might be more consequential if assessed on the basis of a higher amount. But anyway, it’s not hard to understand why some companies might have overlooked that possibility, or else used every rationalization they could to make it go away.

Acasti very precisely says that it recorded the transaction under IFRS at cost, being the fair value of the instruments issued in the transaction. This isn’t necessarily the same as saying it recorded the overall transaction at fair value, because who knows what the consideration would have been if Neptune had implemented a competitive bidding process involving arm’s length parties. In this sense, although Acasti says the adjustment flows from the absence of special recognition requirements for related party transactions under IFRS, it’s still applying a sort of special recognition requirement, in that the exchange amount used to measure the transaction might still be skewed by the “special” relationship between the transacting parties. Hypothetically, the way to counter this might be to measure the transaction retrospectively at an estimate of fair value, and then treat the difference between that and the exchange amount as an equity adjustment, illustrating the transfer of net assets from one group company to another as a result of not transacting at fair value (IFRIC discussed this possibility in 2002 without reaching a conclusion). This might be quite an effective way of illustrating the economics of the transaction, but of course, for a first-time adopter, determining those long-ago fair values would be a subjective exercise to say the least.

Still, it wouldn’t be a surprise going forward if IFRS occasionally caused preparers or auditors to think more rigorously about the appropriateness of the “exchange amount” than they did under Canadian GAAP. The IASB inched towards this territory with a project to consider the accounting for common control transactions, but the work’s been “paused” since 2009; the Board has yet to decide when to pick it up again, if it ever does. Anyway, the project would seemingly have confined itself to considering combinations between entities under common control, rather than the broader transactional issues flagged here. The agenda proposal prepared by IASB staff in 2007 put it like this: “staff agrees that there would be conceptual merit in analyzing all transactions between entities under common control. However, we are concerned that such a comprehensive review might be time consuming and significantly delay the issuance of guidance on combinations between entities or businesses under common control.” Well, if they were concerned about the delay back then, imagine how they should feel now…

The opinions expressed are solely those of the author.

Posted in IFRS, Related Party Disclosures | Leave a comment

IFRS in practice – the broken “promise of comparability”?

Posted on January 5, 2012 by admin

by John Hughes

In connection with its ongoing deliberations about adopting IFRS, the SEC has issued a staff paper, An Analysis of IFRS in Practice, based on examining the most recent annual financial statements of 183 companies domiciled in 22 countries, some of them SEC registrants, others not (given the timing of the exercise, there aren’t any Canadian companies in the sample – 80% are from the European Union). The headline outcome is this: “The Staff found that company financial statements generally appeared to comply with IFRS requirements.” However, the report goes on to emphasize two recurring themes:

“First, across topical areas, the transparency and clarity of the financial statements in the sample could be enhanced. For example, some companies did not provide accounting policy disclosures in certain areas that appeared to be relevant to them…(and overall) certain disclosures presented challenges to understanding the nature of a company’s transactions and how those transactions were reflected in the financial statements. In some cases, the disclosures (or lack thereof) also raised questions as to whether the company’s accounting complied with IFRS…

“Second, diversity in the application of IFRS presented challenges to the comparability of financial statements across countries and industries…in some cases, diversity appeared to be driven by the standards themselves, either due to explicit options permitted by IFRS or the absence of IFRS guidance in certain areas. In other cases, diversity resulted from what appeared to be noncompliance with IFRS. The diversity arising from the standards themselves was, at times, mitigated by guidance from local standard setters or regulatory bodies (and) by a tendency by some companies to carry over their previous home country practices in their IFRS financial statements. While country guidance and carryover tendencies may promote comparability within a country, they may diminish comparability on a global level.”

Tom Selling in his Accounting Onion blog pounces on the report as follows: “The number of apparent departures from IFRS noted and the various ways that IFRS has been interpreted by the largest companies should explode the myth that worldwide adoption of IFRS will result in more comparable financial statements. The unavoidable conclusion is that even if a single set of global accounting standards were possible, the paper leaves no reason to believe that effective global enforcement would occur. A single set of standards without a single, consistent and rigorous enforcement mechanism will fail to deliver on the promise of comparability.”

Well, it’s hard to argue with that. The question though – and I know I’ve covered this territory before, so I apologize, but it’s just unavoidable – is why anyone would even claim that converting to IFRS comes with a “promise of comparability.” Certainly, the SEC report does set out some rank oddities, for example, in the area of share-based payments: “one company disclosed that it estimated the value of stock options once every three years and used that calculation as the basis to calculate share-based payments made during intervening years.” The report drily adds: “How these policies complied with IFRS was unclear.” But for the most part it documents the kind of oversights, corner-cutting and ambiguities that – if you’re realistic about it – just come with the territory for something as complex and multi-faceted as financial reporting. Selling knows of course that there’s not going to be any “single, consistent and rigorous enforcement mechanism” in our time – by setting that up as a criterion, he renders IFRS in the US an impossibility, forever.

But the truth is, investors can easily deal with a certain amount of incomparability, especially if it doesn’t affect key performance measures (which of course are frequently cash-driven – and although the SEC found a mishmash of issues relating to the cash flow statement, they don’t seem to suggest the core picture about the flow of cash is very often obscured). After all, investing may involve some measure of comparing and contrasting different possibilities, but it’s not as if material amounts of capital are going to be sent the wrong way based on quirks in calculating stock option expense. The more important question is whether a particular company’s financial communications make sense on their own terms, whether they adequately convey performance and financial condition and prospects and risks. Mature investors can deal with variations and untidiness around the edges (and as for immature investors, well, there’s not much you can do for them anyway).

To repeat again, I’m not really an advocate for the US adopting IFRS – I doubt it can effectively take such a leap in the face of so much opposition. Maybe a form of voluntary adoption would be more plausible, although of course it’s easy to set out arguments against that as well. But the case against IFRS shouldn’t turn on its failure to deliver some mythical ideal of uniformity. The questions for me – recognizing that like anything else, the payoffs will always be incremental rather than absolute, would be more along these lines: (1) in what ways does (or will) the existence of US GAAP hinder rather than help the optimum flow of capital and (2) would replacing it with IFRS improve that long-term assessment enough to justify the cost and effort of adoption? I don’t know the answers, but I also know they won’t be obtained by burying one’s head in a bunch of financial statements.

The opinions expressed are solely those of the author

Posted in IFRS, United States | Leave a comment

It’s not compensation – I swear!

Posted on December 30, 2011 by admin

by John Hughes

When the comprehensive history of pragmatism gets written (or is that the history of caving in?), there ought to be a chapter on accounting standard-setting, specifically on the urge to clutter things up with exemptions, exceptions, carve-outs, and the like. IFRS is less susceptible to it than old Canadian GAAP was – it pretty much has to be, given the famous emphasis on principles. Still, you find such compromises here and there, especially I suppose in the financial instruments area (I discussed one of them here).

What’s often amusing is the attempt to paint these deviations as being consistent with the basic principles they’re deviating from, in the same kind of way that politicians will try to argue that outrageous spending excesses are essentially consistent with their message of fiscal restraint. Sherritt International Corporation points to one example that IFRS does away with:

CICA 3870 included transactions with employees as being fully within its scope, “unless the transfer clearly is for a purpose other than compensation – (for example) a transfer to settle an obligation of the principal shareholder unrelated to employment by the reporting enterprise.” Sounds fair enough. The section then went on to list various criteria that would cause an employee stock purchase plan to be “not compensatory.” In a nutshell, these included a limit on the time allowed for employees to enroll in the plan and on the discount from the market price (a discount of five per cent or less from the market price was defined as being acceptable in all cases), and a stipulation that substantially all full-time employees meeting limited employment qualifications be allowed to participate in the plan on an equitable basis. If all of these criteria were met, then an entity simply recorded the stock issued under the plan on the basis of the consideration received, without recognizing any element of expense. But is it really true to say a plan set up under these criteria wouldn’t have the primary motive of compensating the employees? Of course not – the whole point of the plan would be to incentivize them, to provide them another kind of stake in the company’s success, and thereby in theory to increase the value they provide in return. Labeling the plan as “non-compensatory” seems merely to have been code for “circumstances where we don’t really care if you don’t apply the standard.”

The IASB considered this issue when they developed IFRS 2, noting the opinion of some commentators that broad-based plans “should be exempt from an accounting standard on share-based payment. The reason usually given was that these plans are different from other types of employee share plans and, in particular, are not a part of remuneration for employee services. Some argued that requiring the recognition of an expense in respect of these types of plans was perceived to be contrary to government policy to encourage employee share ownership.”  But the Board wasn’t going to fall for that one:  “(it) noted that the fact that these schemes are available only to employees is in itself sufficient to conclude that the benefits provided represent employee remuneration. Moreover, the term ‘remuneration’ is not limited to remuneration provided as part of an individual employee’s contract: it encompasses all benefits provided to employees. Furthermore, that governments in some countries have a policy of encouraging employee share ownership is not a valid reason for according these types of plans a different accounting treatment, because it is not the role of financial reporting to give favourable accounting treatment to particular transactions to encourage entities to enter into them.”

As for the related question of whether to exempt certain plans from the standard when the discount is small, however defined: “If the rights given to the employees do not have a significant value, this suggests that the amounts involved are immaterial. Because it is not necessary to include immaterial information in the financial statements, there is no need for a specific exclusion in an accounting standard.” In other words, you want to exclude your employee plan from IFRS 2? – fine, so write a good memo and get the auditors to agree to it!

And that’s an obviously superior approach to the whole thing. Preparers and auditors are likely to engage much more substantially with the issues if they’re required to step back and make a reasoned judgment on whether a particular issue is material to the statements, as opposed to trying to tick the boxes on whether the issue fits into a convoluted structure that’s been pre-ordained as not material. The less standard-setters mess with the core concepts, the more likely practitioners are to actually grasp those concepts and apply them diligently. I wonder how many Canadian GAAP students forced themselves to learn the “non-compensatory” criteria without ever really understanding how these conditions could magically cause an employee incentive to be regarded as something else. I always thought it was just my problem. But now I realize it wasn’t!

The opinions expressed are solely those of the author

Posted in IFRS, Share-based payments | Leave a comment