Can We Go Back to the Good Old Days?

By Dennis R. Beresford

E-mail Story Print Story

I was still shaking my head later that same day when I was clicking mindlessly through the 150 or so channels that my local cable TV service makes available to me. I happened to land on The Andy Griffith Show, and the few minutes I spent with Andy, Barney, Opie, and Aunt Bea got me thinking about the Good Old Days. Wouldn’t it be nice, I thought, to go back to the Good Old Days of the profession in the early 1960s when I graduated from college?

Back then, accounting was really simple. The Accounting Principles Board hadn’t issued any standards yet, and FASB didn’t exist. So we didn’t have 880 pages listing all of the current rules and guidance on derivative financial instruments, for example. The totality of authoritative GAAP at that time fit in one softbound booklet about one-third the size of the new derivatives guidance.

In those Good Old Days, the SEC had been around for quite a while but it rarely got excited about accounting matters. Neither mandatory quarterly reporting nor management’s discussion and analysis (MD&A) had yet come into being, for example. And annual report footnotes could actually be read in an hour or so.

The country had eight major accounting firms, and becoming a partner in one was a truly big deal. Lawsuits against accounting firms were rare, and almost none of them resulted in substantial damages against the accountants.

In short, accounting seemed more like a true profession, with good judgment and experience key requirements for success.

Of course, however much we might like to return to simpler times, it’s easier said than done. And most of us would never give up the many benefits of progress, such as photocopiers, personal computers, e-mail, the Internet, and cellphones. But I think that accounting rules may have become more complicated than necessary.

Let me start with a mea culpa. You may remember the famous line from the comic strip Pogo: “We have met the enemy, and he is us!” Well, you may be tempted to rephrase that quote to “We have met the enemy, and he is … Beresford!”

I plead guilty to having led the development of 40 or so new accounting standards over my time at FASB. A number of them had pervasive effects on financial statements, and some have been costly to apply. I always tried to be as practical as possible, however, although probably few would say that I was 100% successful in meeting that objective.

In any event, more-recent accounting standards and proposals seem to be getting increasingly complicated and harder to apply. Even the best-intentioned accountants have difficulty keeping up with all of the changes from FASB, the AICPA, the SEC, the EITF, and the IASB. And some individual standards, such as those on derivatives and variable-interest entities, are almost impossible for professionals, let alone laypeople, to decipher.

Furthermore, these days, companies are subject to what I’ll call quadruple jeopardy. They have to apply GAAP as best they can, but they are then subject to as many as four levels of possible second-guessing of their judgments.

• First, the external auditors must weigh in.
• Second, the SEC will now be reviewing all public companies’ reports at least once every three years.
• Third, the PCAOB will be looking at a sample of accounting firms’ audits, and that could include any given company’s reports.
• Finally, the plaintiff’s bar is always looking for opportunities to challenge accounting judgments and extort settlements.

Broad Principles Versus Detailed Rules

I suspect that all this second-guessing is what leads many companies and auditors to ask for more-detailed accounting rules. But we may have reached the point of diminishing returns. In response to the complexity and sheer volume of many current standards, some have suggested that accounting standards should be broad principles rather than detailed rules. FASB and the SEC have expressed support for the general notion of a principles-based approach to accounting standards. (It’s kind of like apple pie and motherhood: Who can object to broad principles?) Of course, implementing such an approach is problematic.

In 2002, FASB issued a proposal on this matter. And last year the SEC reported to Congress on the same topic. Specific things that FASB suggested could happen include the following:

• Standards should always state very clear objectives.
• Standards should have a clearly defined scope and there should be few, if any, exceptions (e.g., for certain industries).
• Standards should contain fewer alternative accounting treatments (e.g., unrealized gains and losses on marketable securities could all be run through income rather than the various approaches used at present).

FASB also said that a principles-based approach probably would include less in the way of detailed interpretive and implementation guidance. Thus, companies and auditors would be expected to rely more on professional judgment in applying the standards.

The SEC prefers to call this approach “objectives-based” rather than “principles-based.” SEC Chief Accountant Donald Nicolaisen recently repeated the SEC’s support for such an approach, agreeing with the notion of clearly identifying and articulating the objective for each standard. Although he also suggested that objectives-based standards should avoid bright-line tests such as lease capitalization rules, he called for “sufficiently detailed” implementation guidance, including real-world examples.

Although FASB and the SEC may have reached a meeting of the minds on the overall notion of more general principles, they may disagree on the key point of how much implementation guidance to provide. FASB thinks that a principles-based approach should include less implementation guidance and rely more on judgment, while the SEC thinks that “sufficiently detailed” guidance is needed, and I suspect that would make it difficult to significantly reduce complexity in some cases.

In any event, FASB recently said that it may take “several years or more” for preparers and auditors to adjust to a change to less detail. Meantime, little has changed with respect to individual standards, which if anything are becoming even harder to understand and apply.

I’ve heard FASB board members say that FASB Interpretation (FIN) 46, on variable-interest entities (VIE), is an example of a principles-based standard. I assume they say this because FIN 46 states an objective of requiring consolidation when control over a VIE exists. But the definition of a VIE and the rules for determining when control exists are extremely difficult to understand.

FASB recently described what it meant by the operationality of an accounting standard. The first condition was that standards have to be comprehensible to readers with a reasonable level of knowledge and sophistication. This doesn’t seem to be the case for FIN 46. Many auditors and financial executives have told me that only a few individuals in the country truly know how to apply FIN 46. And those few individuals often disagree among themselves!

Such complications make it difficult to get decisions on many accounting matters from an audit engagement team. Decisions on VIEs, derivatives, and securitization transactions, to name a few, must routinely be cleared by an accounting firm’s national experts. And with section 404 of the Sarbanes-Oxley Act (SOA) and new concerns about auditor independence, getting answers is now even harder. For example, in the past, companies would commonly consult with their auditors on difficult accounting matters. But now the PCAOB may view this as a control weakness, under the assumption that the company lacks adequate internal expertise. And if auditors get too involved in technical decisions before a complex transaction is completed, the SEC or the PCAOB might decide that the auditors aren’t independent, because they’re auditing their own decisions.

When things become this complicated, I wonder whether it’s time for a new approach. Maybe we do need to go back to the Good Old Days.

Internal Controls

Today, financial executives are probably more concerned about internal controls than new accounting requirements. For the first time, all public companies must report on the adequacy of their internal controls over financial reporting, and outside auditors must express their opinion on the company’s controls. Many people have questioned whether this incredibly expensive activity is worth the presumed benefit to investors. While one might argue that the section 404 rules are a regulatory overreaction, shareholders should expect good internal controls. And audit committees, as shareholders’ representatives, must demand those good controls. So this has been by far the most time-consuming topic at all audit committee meetings I’ve attended in the past couple of years.

Companies and auditors are spending huge sums this year to ensure that transactions are properly processed and controlled. Yet the most perfect system of internal controls and the best audit of them might not catch an incorrect interpretation of GAAP. A good example of this was contained in the PCAOB’s August 2004 report on its initial reviews of the Big Four’s audit practices. The report noted that all four firms had missed the fact that some clients had misapplied EITF Issue 95-22. As the New York Times (August 27, 2004) noted, “The fact that all of the top firms had been misapplying it raised issues of just how well they know the sometimes complicated rules.”

Responding to a different criticism in that same PCAOB report, KPMG noted, “Three knowledgeable informed bodies—the firm, the PCAOB, and the SEC—had reached three different conclusions on proper accounting, illustrating the complex accounting issues registrants, auditors and regulators all face.”

Fair Value Accounting

Even those who are very confident about their understanding of the current accounting rules shouldn’t get complacent: Fair value accounting is right around the corner, making things even harder. In fact, it is already required in several recent standards.

To be clear, I’m not opposed in general to fair value accounting. It makes sense for marketable securities, derivatives, and probably many other financial instruments. But expanding the fair value concept to many other assets and liabilities is a challenge.

Consider this sentence from FASB’s recent exposure draft on fair value measurements: “The Board agreed that, conceptually, the fair value measurement objective and the approach for applying that objective should be the same for all assets and liabilities.” In that same document, FASB said, “Users of financial statements generally have agreed that fair value information is relevant.”

So the overall objective of moving toward a fair value accounting model seems clear. Of course, that doesn’t necessarily mean that we will get there soon. In fact, in the same exposure draft the board said that it would continue to use a project-by-project approach to decide on fair value or some other measure. But in reality the board has been adopting a fair value approach in most recent decisions:

• SFAS 142, on goodwill, requires that impairment losses for certain intangible assets be recognized based upon a decline in the fair value of the asset.
• SFAS 143, on asset retirement obligations, requires that these liabilities be recorded initially at fair value rather than what the company expects to incur.
• SFAS 146, on exit or disposal activities, calls for the fair value of exit liabilities to be recorded, not the amount actually expected to be paid.
• FIN 45, on guarantees, says that a fair value must be recorded even when the company doesn’t expect to have to make good on a guarantee.

A fair value approach is also integral to other pending projects, including the conditional asset retirement obligation exposure draft. Under such a standard, a company might have to record a fair value liability even when it doesn’t expect to incur an obligation. Fair value is also key to projects on business combination purchase procedures; differentiating between liabilities and equity; share-based payments (stock options); and the tremendously important revenue recognition project.

I have three major concerns about such pervasive use of fair value accounting. First, in many cases determining fair value in any kind of objective way will be difficult if not impossible. Second, the resulting accounting will produce answers that won’t benefit users of financial statements. Third, those answers will be very difficult to explain to business managers, with the result that accounting will be further discredited in their minds.

The approach that FASB is using for what I would call operating liabilities is particularly troubling. Take, for example, a company that owns and operates a facility that has some asbestos contamination. The facility is safe and can be operated indefinitely, but if the company wanted to sell the property it would have to remediate that contamination. The company has no plans to sell the property. But FASB’s exposure draft on conditional asset retirement obligations calls for the company to estimate and record a fair value liability. This would be based on what someone else would charge now to assume the obligation to clean up the problem at some unspecified future date. The board admits that it might be difficult to determine what the fair value would be in this case, and companies could omit the liability if they simply couldn’t make a reasonable estimate.

Although FASB and the SEC expect most companies to be able to make a reasonable estimate, in reality I think that will be possible only rarely. Even more important, does it really make sense to record a liability when the company might believe that there is only a 5% chance that it will have to be paid? Consider how this line of reasoning might apply to litigation. Presently, liabilities are recorded only when it’s probable that a loss has been incurred and that a reasonable estimate of the loss can be made. So if a company were sued for $1 billion but there were only a 1% chance that it would lose, nothing would be recorded. The fair value approach would seem to call for a liability of $10 million in this case, based on 1% of $1 billion.

One might think this kind of accounting will apply only in the distant future, but FASB is due to release its proposal on purchase accounting procedures in the next few months, and I understand that the proposal will require exactly this kind of accounting.

In addition to the very questionable relevance of this, I don’t know how anyone would ever be able to reasonably determine the 1% likelihood I assumed. How would an auditor attest to the reliability of financial statements whose results depend significantly on such assumptions? And where would an auditor go to obtain objective audit evidence against which to evaluate such assumptions?

Fair value definitely makes sense in certain instances, but FASB seems intent on extending the notion beyond the boundaries of common sense. FASB also seems to have an exaggerated notion of what companies and auditors are actually capable of doing. Perhaps we should consider FASB’s faith in the profession to be a compliment. Rather than feeling complimented, however, I think that this just makes many of us long for the Good Old Days.

Fair Value Accounting and Revenue Recognition

Currently, asset retirement obligations and exit costs apply to only a few companies, and even guarantees are not an everyday issue. All companies, however, have revenues—or at least they hope to have them. And for the past year or so, FASB has been engaged in a complete rethinking of revenue recognition. This, of course, was precipitated by the numerous SEC enforcement cases on improper revenue recognition. Most cases, however, involved failure to follow existing standards, and most cases also resulted in premature recognition of revenue.

Now there’s no doubt that the current revenue accounting rules are overly complicated, with many specific rules depending on the type of product or service being sold. But FASB’s current thinking would replace these rules with an asset and liability–oriented approach based on fair value accounting. This may well make revenue accounting even more complicated than the detailed rules that we are at least used to working with.

For example, assume product A is being sold to a customer. It costs $50 to produce product A and the customer has agreed to pay a nonrefundable $100 in exchange for the company’s promise to deliver this hot product next month. What should the company record at month-end?

Most accountants would probably think first of the traditional approach and conclude that the earnings process had not been completed. Because product A hasn’t been completed and shipped to the customer, the $100 credit is unearned income. Some aggressive accountants would probably say that the company should record the sale now because the $100 is nonrefundable. In that case the company would probably also record a liability for the $50 cost that will be incurred next month.

FASB has a surprise for both. The board is presently thinking about whether revenue for what it calls the “selling activity”—the difference between the $100 received and the assumed fair value of the obligation to deliver the product—should be recorded now. This assumed fair value would be the estimated amount that other companies would charge to produce product A. In other words, it’s the hypothetical amount a company would have to pay someone else to assume the obligation to produce the product. The company would have to make this assumption even though it is 100% sure that it will make the product itself rather than have someone else make it.

If one could ever determine what other companies would charge, I suspect that the amount would be higher than the $50 expected cost, because another company probably would require a risk premium to produce a product that it isn’t familiar with. It would want to earn a profit as well. Let’s assume in this case that the fair value could be determined as $80. If so, the company would record now $20 of revenue and profit for what FASB calls the selling activity. Next month it would record the $80 remaining amount of revenue, along with the $50 cost actually incurred. It’s unclear when the company would record sales commissions, delivery costs, and similar expenses, but I assume these would have to be allocated somehow.

Given that this project was added to FASB’s agenda in large part because of premature recognition of revenue in some SEC cases—Enron recognized income based on the supposed fair value of energy contracts extending 30 years into the future—it is ironic that the project may well mandate recognition earlier than most accountants would consider appropriate. That kind of premature revenue recognition is now generally prohibited, but other examples could follow, depending on the outcome of this FASB project.

Although the revenue recognition project is still in an early stage and both my understanding and the board’s positions could change, FASB seems determined to use some sort of fair value approach to revenue recognition in many cases. If this happens, we will all be wishing for the Good Old Days to return.

Is All That EITF Guidance Really Necessary?

In early 2004, FASB’s board members began reviewing all EITF consensus positions. A majority of board members now have to “not disagree” with the EITF before those positions become final and binding on companies. This gives FASB more control over the EITF process, and it should prevent the task force from developing positions that the board sees as inconsistent with existing GAAP.

Although I think the task force has done a great deal of good over its 20-year existence (I was a charter member), I think it’s time to challenge whether everything that the EITF does is necessary or even consistent with its original purpose. Too many of the task force’s topics in recent years can’t really be called “emerging issues.” Rather, the task force often takes up long-standing issues where it thinks that some limitations need to be placed on professional judgment.

For example, a couple of years ago the SEC became concerned about the accounting for certain investments in other companies. For years we’ve had standards that call for recognition of losses when market value declines are “other than temporary.” The EITF discussed this matter at eight meetings over two years and also relied on a separate working group of accounting experts. Earlier this year, a final consensus position was issued. It includes a lengthy abstract that tells companies what factors to consider, including the following matters:

• Evidence to support the ability and intent to continue to hold the investment;
• The severity of the decline in value;
• How long the decline has lasted; and
• The evidence supporting a market price recovery.

So now we have a “detailed rule” on this matter. Will this result in more consistency in practice? Will investors and other users of financial statements receive better information as a result? Is the result worth the additional effort?

Moreover, after two years of effort on this project, FASB had to reconsider the whole thing because no one had considered the effect on debt securities held as available for sale by financial institutions. So now the board is developing even more specifics to deal with the unintended consequences of the rule.

Again, I support the EITF, and I believe it has generally done a great job. The members try to develop practical ways to deal with current problems. Nonetheless, both the task force and FASB may need to more carefully challenge whether all of the EITF’s projects are really needed. If FASB actually issued relatively broad standards, there probably would be a need for the EITF to provide supplemental guidance on some issues. But we now seem to have the worst of all worlds, with quite detailed accounting standards being accompanied by even more detailed EITF guidance.

A Multitude of Challenges

I don’t intend to seem overly critical of FASB and others who are working to improve financial reporting. It’s a tough job, and the brickbats always outnumber the bouquets. If I didn’t strongly
support accounting standards setting I wouldn’t have spent 10 Qs years on the inside of the process. Still, those years at FASB, as well as my time before and after, have caused me to develop strong views on these issues. And I truly do believe that standards have gotten just too complicated.

The announced move to broader principles is one I fully support. That job won’t be easy, but it has to be tried or the sea of detail will become even deeper in the near future. FASB needs to actually start doing this and not allow its actions to speak otherwise. And companies, auditors, and regulators need to support such a move and resist the temptation to seek answers to every imaginable question. Furthermore, companies and auditors may have to become more principled before a principles-based approach will work.

Part of this process could be for the EITF to be more judicious in what it takes on. Also, I urge FASB to reevaluate its attitude toward fair value accounting. I believe FASB is moving much faster in this area than preparers, auditors, and users of financial statements can accommodate. Furthermore, the SEC and other regulators may not yet be on board with this new thinking.

In the final analysis, we won’t be able to return to my so-called Good Old Days. But we have to make sure that what accounting and accountants can do is meaningful and operational. We never want to look back and ask, “Remember the Good Old Days, when accounting was important?”