Do you remember the Big Four mantra that IFRS adoption in the U.S. was “inevitable”? Today, thankfully, the operative word is “dead.” Even former SEC chair, Christopher Cox, whose brainchild was the IFRS adoption Roadmap, now comes to “bury IFRS.”
But as Yogi Berra famously said, “It ain’t over ’til its over.”
Mary Jo White now reigns at the SEC, and hasn’t said much about the future of the Roadmap. Methinks, that like her predecessor, Mary Schapiro, Chair White is in no hurry to stick her neck out on an accounting policy question. Consequently, the SEC’s official position hasn’t budged since way back in February 2010:
“…a single set of high-quality globally accepted accounting standards will benefit U.S. investors and that this goal is consistent with our mission…” [italics added]
To the above, Chair White added only the following last May when speaking at a Financial Accounting Foundation function:
“I have made it a priority for the Commission to position itself to make a further statement on this very important subject … and I hope to be able to say more in the relatively near future.” [emphasis added]
Considering the glacial pace of accounting standard setting, “relatively” could be years! And, I wouldn’t be too surprised if it is. But, let’s optimistically assume that Chair White will gird herself to say something about the future status of IFRS soon after the mid-term elections. That being the case, any assessment of the efforts to incorporate IFRS into U.S. standards has to begin, and could very possibly end, with the revenue recognition project.
Forgive me for stating the obvious, but revenues are the most scrutinized of all of the balances in a set of financial statements. And, after 12 years of shifting positions and joint deliberations, the resulting new standard (ASU No. 2014-09, and IFRS 15 are almost identical) is practically a fresh start on the topic. As a case study, no other project/standard would be a better place to start an evaluation of whether the collaboration with the IASB on major projects has been “consistent with [the SEC’s] mission.”
What is Taking So Long?
See Berra, supra. The joint standard took 12 years to issue in its (near) final form, and won’t become effective until 2017. Despite the curious fact that the IASB is permitting early adoption, the long time lag between issuance and effectiveness is intended to give issuers time to submit the many many questions of application that remain unresolved (after 12 years!) to a Joint Transition Resource Group.
Taken together, 15 years from project initiation to effective date must be some sort of record. I know that “glacial” is often used to jab at the (overly) deliberative pace of accounting standard setting “due process,” but I fear that we may be insulting glaciers; they move ever faster, while “due process” gets more and more bogged down. (I wonder how much global warming is due to the hot air produced by FASB “due process” deliberations? Okay, enough already!)
One major cause for 12 years from inception to (near) final draft is that the Boards kept changing their minds about why, other than simply for the sake of convergence, a new standard became a top priority. Most prominently from my perspective, is that the FASB started out by aiming to improve US GAAP through consistent application of the balance sheet view to revenue measurement, as set forth in its conceptual framework (in contrast to “matching” revenues and expenses), much as the balance sheet view is more consistently applied to expenses. Since the criteria for recognition and measurement of assets and liabilities would, of necessity, have been broad-based, a revenue recognition standard would be capable of being applied to all transactions with an entity’s customers — regardless of industry or the type of goods/services provided.
The (near) final standard in ASU 2014-09 is nothing like a balance sheet view, and predictably, the FASB would like to pretend in subsequent “due process” documentation that it even tried to adhere to its own principles. There is nothing in the (near) final document’s basis for conclusions to indicate that consistency with expense measurement was ever a goal; which avoids having to answer why it was ultimately abandoned.
Impolitic me did have an opportunity to ask a Board member why the balance sheet view was abandoned, and this is what he said:*
“The standard that was looked at, or what was being considered over the first five years of the project, while it might have been easier to write, would not have been easier to apply. I guarantee you that.” [emphasis added]
I remain skeptical, despite the “guarantee.” As I observed this project unfold (or dare I say “devolve”), the choice to abandon the balance sheet approach was followed with death by a thousand comment letters. First came the scope exceptions for insurers and lessors, essentially because they didn’t like the new patterns of revenue recognition. Then came the construction companies who feared that their ability to smooth earnings under the percentage-of-completion method would be upended. Then came everyone else who were riled up because an objective definition of a liability wouldn’t provide sufficient wiggle room to smooth out cash collections from licenses, upfront initiation fees, etc.
So, the new thinking on why we need every company in the U.S., both public and private, that wishes to comply with U.S. GAAP to change the way it recognizes revenue goes like this, according to a Powerpoint presentation by the same FASB member (I’m paraphrasing):
Increase comparability across industries and capital markets;
Address an area in which there has been a high frequency of restatements;
Provide a uniform (notwithstanding the numerous scope exceptions) and robust framework for addressing revenue recognition issues as they arise;
Require better disclosure.
Except for the disclosure improvements, which I heartily support, let’s take these one at a time.
Comparability—It is inarguable that the application of the revenue recognition standard will call for issuers to make many more judgments than under extant U.S. GAAP. Here is a very quick and dirty list:
- Just to get into the scope of the new standard, an issuer will have to conclude that collection of the amounts specified in the contract are “probable.” Of course, probable is not defined, and it has already been conceded that U.S. GAAP issuers will be held to a different (albeit unspecified) version of “probable” than IFRS issuers. (Examples of where this could be a sensitive issue are providers of emergency room services, installment payment arrangements to consumers with low credit scores, and residential real estate development.)
- Multiple contracts with a customer might have to be accounted for as if they were one single contract. While this is clearly an anti-abuse provision, and criteria for combining contracts are spelled out to some degree, any realist knows that games will be played; and auditors will be put between a rock and a hard place by those who wish to game this aspect of the standard.
- What used to be called “multiple element arrangements” under U.S. GAAP is now referred to as the identification of separate “performance obligations” (POs). The objective of the standard is that goods/services that are “distinct” should be treated as separate POs.
- A PO can be either an explicit promise, or an implicit promise that creates a “valid expectation” that a good/service will be provided in the future. Here are just a couple of examples of problems: How will intent affect whether a customer has a valid expectation of free or discounted access to future software upgrades? If a broadcaster purchases a license from a producer to air re-runs of the first season of a popular series, how much of the revenue should be deferred and for how long? Because the broadcaster expects that new episodes will enhance the value of the license to re-broadcast the first season, is that an implicit promise that justifies deferral of revenue? If so, what is the pattern of revenue recognition over time?
- When determining the transaction price (i.e., the amount of consideration to be allocated among the POs, “variable consideration” has to be estimated. The first eye-opener is that the Board is now effectively requiring issuers to estimate and recognize contingent gains before they are realized. (Goodbye FAS 5 prohibiting contingent gains.) Second, no two issuers are likely to come up with the same estimate given identical contracts. (Example: an estimate of cash to be received by a healthcare provider from the patients insurance company.)
- Variable consideration is also constrained to the extent it is “probable” that a significant reversal in the amount of cumulative revenue recognized will not occur when the uncertainty is ultimately resolved. “Probable” is not defined … whoops, I already said that.
- The transaction price is to be allocated to each PO on a “relative standalone selling price basis.” But, of course, many good/services do not have observable selling prices; hence the standard allows for considerable latitude in estimating those prices and the use of unspecified residual estimate techniques for making the allocations. There is also the problem of allocating discounts and contingent amounts to the POs.
Enough already. I could easily add ten more consequential items to my list, but the above should be more than enough to make my point that expecting comparability when so much judgment is involved — and before any field testing has taken place — is a huge leap of faith on the part of the FASB.
Surely, Chair White is acutely aware that comparability also depends on a lot of external factors — like whether accounting standards will be audited, enforced and applied in the same way throughout the world. To believe that financial statements will become more comparable just because everyone is reading from the same vague set of instructions with its own operational problems is just not realistic.
Frequency of Restatements—I also asked the FASB member why we should expect the new standard to decrease the incidence of restatements. This was his response:
“I’m not sure I actually said that I think it’s going to decrease restatements. I said one of the reasons why we took on the project was because of the incidence of restatements in revenue recognition. Those are subtle differences. [But,] I think at the end of the day we are hoping that with education and through the implementation group [i.e., without any field testing after 15 years] that we’ll have people making similar judgments when they are faced with similar situations.”
Actually, I do agree that the new standard will result in fewer restatements, but for two different reasons. Both of them don’t provide much reason to believe that financial reporting quality is improved by the new standard.
The first reason is that the companies that try to manage their revenue recognition by adjusting the arrangements with their customers (by often giving concessions to the customers solely for the sake of financial statement window dressing) risk tripping over a new requirement that they didn’t know about until someone brought the mistake to their attention after the financial statements were issued. One thing I will say about ASU 2014-09 is that its requirements are not intricate. Thus, (value-destroying) earnings management can proceed with less fear of falling into a restatement.
The second reason is that more efforts at earnings management will be based on management’s “judgments” and stated intent. Even the most conscientious of auditors and regulators will have difficulty reigning that in. (And, to add insult to injury, I fully expect that the auditors and regulators who are the cheerleaders for the Accounting Establishment will use the reduced incidence of restatements to trumpet the news that its “principles-based” revenue recognition standard had the intended effect.)
Uniform standard for addressing issues as they arise—If you have followed the discussion at the first meeting of the Transition Resource Group, perhaps you got the impression, as I did, that the participants wasted a few hours chasing their tails over one of the more mundane topics addressed by the new standard: gross versus net presentation of revenue.
IMHO, after the balance sheet approach was abandoned, the boards spent much of the ensuing seven years engaged in ad hoc reverse engineering—i.e., trying out new rules that looked good on paper and would more or less produce the same numbers as the rules that are currently in effect. It’s going to be interesting to see whether further TRG deliberations will be just more of the same trial and error.
* * * * * *
Here’s a thought experiment. Pretend you don’t actually know that the final standard is the result of a collaborative effort between the FASB and IASB; and that you have to guess which board produced it. (Sort of a Turing test.) I would be willing to bet that most accountants, if they were being honest with themselves, would say it looks like the IASB called the tune.
That alone should give Chair White some pause, but it is even more important for her to recognize, before it’s too late, that the new standard will cost every public company serious money to implement, and for no solid reason that issuers or investors should appreciate.
The idea for a new revenue recognition standard may have been a good one 12 years ago, but today, it’s a turkey.
*Each of the quotes from the FASB member in question were extracted from the webcast of the June 25th meeting of the PCAOB’s Standing Advisory Group, of which I am a member. The views expressed herein are my own and do not necessarily reflect the views of the PCAOB, its members or staff.