Two of our prior columns critiqued the Financial Accounting Standards Board's recent preliminary views document on its new Conceptual Framework project, which, like many other FASB efforts, is a joint undertaking with the International Accounting Standards Board. Those columns examined the boards' take on financial reporting objectives and relevance, both of which are basically enhanced versions of FASB's first framework from the 1980s. While we see some improvement, we wish that the boards had pushed the envelope further and set the stage for more reformation in practice.
Like what you see? Click here to sign up for Accounting Today's daily newsletter to get the latest news and behind the scenes commentary you won't find anywhere else.
This column focuses on faithful representation, which occupies the place held by reliability in the original framework. The new basic definition of faithful representation is the "correspondence or agreement between the accounting measures or descriptions in financial reports and the economic phenomena they purport to represent." (Par. BC2.28)
That is, a measure is a faithful representation if it is a dependable depiction of what is being described in the financial statements. The original framework broke reliability down into the sub-qualities of representational faithfulness, verifiability and neutrality, where representational faithfulness included both completeness and freedom from bias. This time, FASB and the IASB propose that "a faithful representation of ... economic phenomena ... must be verifiable, neutral and complete." (Par. QC16)
While the name of the concept has changed, the underlying nature of the whole package shows that the other changes are slight. We found the situation to be like a newspaper puzzle that challenges you to find subtle differences between two very similar drawings. Because it's hard to find the differences, one might wonder why the boards changed the name. We think it was more than semantics.
WHAT'S IN A NAME?
Specifically, the boards wanted to avoid measurement arguments that used to hang upon the meaning of "reliability." The boards describe their rationale as follows: "In considering the issues related to ... reliability, as well as standard-setters' experience with assessing reliability, the boards observed the existence of a variety of notions of what the concept means. For example, some constituents focus on verifiability to the virtual exclusion of the faithful representation aspect of reliability." (Par. BC2.26)
There you have it - they tossed reliability out simply because too many people refused to understand its precise meaning in the framework.
Instead, many firmly clung to the familiar but invalid idea that reliability only means "subject to verification." These people think that a financial depiction is reliable if it can be verified by another person. Despite attempts over the years to convince them otherwise, the boards concluded that it's better to start over by renaming the overarching concept and then help others see that verification is only one facet of the package.
The boards' document defends the renaming decision by saying that "further efforts to explain what reliability means did not seem likely to be productive." (Par. BC2.27)
VERIFICATION ISN'T ENOUGH
There are several important points about faithful representation and its relationship to relevance, the other highest-level qualitative characteristic.
To begin, the boards insist that faithfulness is much more than simple verification. For example, a doublecheck of a depreciation schedule may verify that its calculations are correct, but does nothing to verify the more important proposition that the resulting book values actually provide faithful representations of the assets' future cash flows.
In other words, it does no good to validate numbers that do not describe the reporting company's cash-flow potential. Now that the proposed framework clearly states that relevance is based on helping users evaluate future cash flows, many traditional measures, especially any that involve systematically allocated historical costs, have no real possibility of being faithful representations of that potential. If a measure does not faithfully represent the magnitude of the assets' and liabilities' potential cash flows, there can be no usefulness in it, no matter how many people verified the supporting calculations.
WHO'S ON FIRST?
The preliminary views document wisely stays away from the unwinnable game of arguing whether relevance or faithful representation is more important. Not even Solomon could resolve that issue, because both qualities are inextricably linked and necessary for information to have decision-usefulness.
Despite this interrelationship, the document clearly states that relevance should be addressed first: "If information about a particular real-world economic phenomenon is not pertinent to investment or credit decisions, none of the other qualitative characteristics matter. Accordingly, it would be inefficient to consider faithful representation, comparability or understandability for irrelevant items." (Par. BC2.62)
Because the set of relevant measures is much smaller than the set of faithful representations, the most efficient screening process begins by identifying relevant numbers and then assessing their faithfulness.
Once relevance is settled and it's determined that a proposed measure reflects the relevant attribute's magnitude, then the next test is for its verifiability. The document explains more about verification by identifying two different ways of accomplishing it. Specifically, information can be verified directly by actually repeating the original process, or it can be verified indirectly by determining that the original process is appropriate. Either approach works to confirm the result's validity.
While the boards' discussion makes it clear that verification is not enough, it's disappointing that the other components of faithful representation don't include a key point. As mentioned earlier, the proposal identifies three sub-qualities of faithfulness: verification, neutrality and completeness. We have no problem with these three, but we regret the omission of the idea of congruence of the verified measure with the relevant attribute's real magnitude.
For example, if an asset's market value were $25 million, then a reported number would be faithful only if it's close to that. (Other terms that could be used for this quality might be validity - or accuracy.) Leaving out congruence as a specific component of faithful representation leaves the door wide open for someone to claim that it really isn't that important, despite what is said elsewhere in the framework.
We think including a separate component for congruence is crucial to making it clear that verifiable measurement methods are worthless if their results don't correspond to reality.
We hope that the new framework will elevate understanding of faithful representation, so that people fully comprehend that little is achieved by simply checking off others' work. We are especially pleased that the new framework will clarify that relevance and faithful representation are both essential, thereby discrediting those who cling to verifiability as justification for historical cost-based measures. The new terminology will make it difficult for opponents of market value reporting to find traction with their old arguments.
Our optimism is promoted by several recent FASB decisions that have achieved real progress in giving users the kind of information they sought, but that was adamantly opposed by others in the past. Specifically, users can now easily find facts about options expense and pension obligations right there in the statements; in addition, the board has given its blessing to expanded market value accounting in many situations.
The board's resolute moves despite strong opposition tell us that the proposed changes in the framework are likely to become official. And that's good news for everyone, although not everyone will see it that way.
Paul B. W. Miller is a professor at the University of Colorado at Colorado Springs and Paul R. Bahnson is a professor at Boise State University. The authors' views are not necessarily those of their institutions. Reach them at firstname.lastname@example.org.