Speech by SEC Staff:
Remarks before the 2005 AICPA National Conference on Current SEC and PCAOB Developments

by

Scott A. Taub

Acting Chief Accountant
U.S. Securities and Exchange Commission

Washington, D.C.
December 5, 2005

Good morning. It's a pleasure to be here today. This is the 7th year in a row that I've spoken at this conference, and I want to congratulate the AICPA on what once again looks like it'll be a very successful gathering. I also want to commend the leadership of the AICPA for continuing to maintain the vitality and relevance of the organization in the past few turbulent years.

As you've just heard, I was recently named the Acting Chief Accountant at the SEC after Don Nicolaisen left. I think I can speak for everybody in my office when I say that we'll all miss working with and for Don, and that we're thankful for the leadership he provided while he was here. As you may also have heard, we are, at the end of this week, losing a Deputy Chief Accountant, Andy Bailey, who has led our efforts related to the PCAOB and other auditing matters. Again, I think I speak for everybody in the office when I say Andy will be missed, and that his work has contributed significantly to the progress made in the past couple of years.

While I was confident that I could speak for the whole office about Don and Andy, I do need to tell you that from now on, my remarks are solely my own and do not necessarily represent the views of the Commission or anybody else on the Commission's staff. And I should add that a similar disclaimer applies to everybody else from the SEC that you'll hear today.

A Status Report

Taking on the role of Acting Chief Accountant prompted me to stand back and think about the status of financial reporting. I joined the SEC as a Deputy Chief Accountant just over three years ago. At that point in time, the Sarbanes-Oxley Act was brand new, and we were just recovering from a spate of reporting failures that rocked the markets and the accounting profession.

The changes that arose out of Sarbanes-Oxley have been far-reaching and affected nearly everybody in the financial reporting process. Even now, three years later, I am somewhat awed by the scope of those changes, which include the following:

Back in 2003, I followed up a similar recitation of these changes with the question. "So, what does this all mean?"1 I suspect the answer to that question would be very different depending upon who was asked, but I believe, as I have said before, that what it means is that Congress was telling us, in case we forgot, that financial reporting works best when we act in the public interest.

So, today's question is, how are we doing?

Again, I suspect the answer would depend significantly on who was asked. But I believe we've come a long way in the right direction.

For example, members of senior management spend more time on financial reporting than they did before. Moreover, their approach has changed. A few years ago, I remarked before this audience that I was all too familiar with the question from public company registrants that went "Where does it say we can't do this?" Now, however, I also hear, perhaps with more frequency, the question that says "Are we sure that this treatment is all right?"

Rather than treating financial reporting departments as a place to cut costs and headcount, companies are adding resources. And we are seeing signs that audit committee members are recognizing the importance of their enhanced role. The benefits of all of this, while intangible in some respects, should be pervasive on both internal and external reporting.

Auditors, too, are making some positive changes. I see audit firms refocusing on auditing and on technical expertise, as opposed to selling work. And I've seen plenty of instances of firms making tough choices by refusing to be associated with inappropriate accounting. The work of the PCAOB should help to ensure these changes take root, as opposed to fading away.

Standard-setters have also made changes. Both the FASB and IASB are looking to improve their processes, are taking up projects in areas that have long needed improvement, and are striving to write standards that emphasize objectives and principles. And, as Chairman Cox has already mentioned, the FASB is taking other important steps towards making accounting standards more accessible to practitioners.

In addition, reporting on internal controls has brought new information to the markets, and renewed recognition of the importance that internal controls play in the reporting and governance of public companies. While the costs have been greater than anticipated, the benefits to the market have already been significant. We have all seen a number of reports of preparers that have streamlined operations, uncovered frauds and errors, and otherwise improved their position because of their work on evaluating internal controls. I have been encouraged by the way the markets have digested the information on internal controls, by focusing on the nature of the weakness, and not merely on the fact that one exists.

But the thing that most makes me feel like we're headed in the right direction is that I have heard so many participants in the process acknowledge the importance of acting in the public interest. Auditors, both in the US and overseas, have been particularly clear in acknowledging and addressing their public interest role. I have heard the importance of the public interest and interest of investors discussed by people in all kinds of corporate roles, as well. And students considering the accounting profession seem to be attracted to it, in part, by the chance to be on the side of the public interest.

In summary, I have to say that I am encouraged by what I have seen recently. We have come a long way, and I'm glad to have been a part, however small, of the improvements that have occurred. Of course, saying that I'm encouraged isn't at all the same as saying that I'm satisfied, or that further improvements aren't needed. After all, I don't think many people in this audience would suggest that the financial reporting system is operating at peak effectiveness and efficiency.

Today you'll hear from many members of the Office of the Chief Accountant and the Division of Corporation Finance about ideas for how financial reporting can be further improved. Every person you'll hear from today is committed to working towards improving financial reporting in the interests of investors and the capital markets as a whole. I inherit a great staff in OCA, and we and our colleagues at the SEC look forward to continuing to be in the forefront of efforts to further improve financial reporting. We will, however, need a lot of help, and also, leadership, from you in the audience and your colleagues. Real and lasting improvement will take efforts from everybody. You've heard that a lot of times before, but I hope that in the balance of this speech, I can perhaps make the case a little bit, and get a few more people interested in actively working towards these improvements.

Reducing Complexity

Let me start by giving you a bit more on a topic that Chairman Cox already commented on, the complexity of financial reporting. Various people from SEC, the FASB, and other organizations have been talking about this for quite some time, and you'll hear others discuss it later in this conference. Complexity is an undercurrent to pretty much the entire staff report on off-balance sheet activities. The responses to FASB's annual survey of constituents about its agenda contain repeated requests to simply accounting standards. And I'll bet if I asked this audience who is in favor of less complexity, everybody who is still paying attention would raise their hands.

In an article published in the November 2005 issue of "Financial Executive", Colleen Cunningham, the President of FEI, did a great job of concisely explaining the complexity issue. She wrote that "When regulators and standard-setters think about "complexity", they understand it to mean the inability to succinctly determine the principle of an accounting standard…" 2 An example: try walking through all of the literature that applies (or might apply) to the issuance of convertible debt. Amongst the pieces of accounting literature that might apply are APB 14 3, EITF 00-19 4, EITF 00-27 5, SFAS 133 6, SFAS 150 7, EITF 05-4 8, ASR 2689 …. And it isn't exactly obvious how these pieces of literature interact with one another in all circumstances. That's complex, and similar examples abound all over the accounting literature.

But why is that? It certainly didn't get that way on purpose — nobody consciously decided a complicated set of standards is preferable to a straightforward one. That being said, I do think it got that way because we asked for it, even if we didn't realize we were asking for it.

Many of the areas of GAAP that are considered overly complicated are areas that include detailed bright lines, exceptions to principles, or optional treatments. And these facets of the accounting literature inevitably are there because constituents asked for them. Consider a couple of examples -- the short-cut method of hedge accounting, and the existence of Qualifying Special Purpose Entities. The short-cut method was added to make hedge accounting easier for certain interest-rate hedges, while Q's were created to make sale accounting easier in certain securitization transactions. However, their existence has resulted in a significant number of restatements, because they have been used incorrectly. So we wind up with restatements caused by mistakes in applying exceptions that were supposed to make things easier.

Other areas of complexity exist due to a desire to reduce volatility in the income statement. The accounting for defined-benefit pension plans10 is a prime example. A simple principle might be to say the difference between the best estimate of the pension obligation and the best estimate of the value of the plan assets should appear on the balance sheet. I recognize that estimating the obligation would still necessitate some difficult choices, but the principle for determining the balance sheet number would be clear and understandable. Instead, we have a complex series of optional smoothing mechanisms that renders the balance sheet number meaningless. And pension accounting is even more interesting because every company that uses the smoothing mechanisms has consciously chosen additional complexity, and additional chances to make errors, for things like choosing an estimated asset return and an amortization life for deferred gains and losses — things that wouldn't be needed with a simpler accounting principle.

In other instances, complexity seems to exist in response to concerns about being second guessed — by auditors, by the plaintiffs' bar, by the SEC. Later today, you'll hear a couple of OCA staff people discuss certain aspects of the accounting for share-based payments, the subject of FASB Statement 123R11 , and Staff Accounting Bulletin No. 10712 . After FAS 123R came out, the SEC staff was asked about various estimation methods that might be used in relation to key assumptions, and we provided those who asked the questions our thoughts on those matters. However, we were told by some of those whose questions we had answered that they would not feel comfortable deviating from methodologies that were felt to be "safe" unless our views that other methods were acceptable eventually appeared in writing. Following along the same topic, I've heard many people complain about the FASB moving to issue an FSP13 on how to determine when a grant date has occurred if the Board of Directors approves the issuance of options before communication to employees is made. But the FASB staff felt this needed to be issued, because some constituents had concluded that in the absence of written guidance, they would only be comfortable with the most conservative of interpretations.

It's fair at this point for me to acknowledge that the SEC staff has played a role in increasing complexity. Because of actions we have taken in the past, we have been accused of being unwilling to accept judgments that differ from our own, of relying on bright-line rules, including some that don't exist in the literature, and of being unwilling to accept reasonable diversity in practice. I could tell you, over and over again, that we don't as matter of policy, do any of those things. But the fact of the matter is that the belief that the SEC staff does these things has certainly helped to encourage requests for more guidance, for exceptions that specifically allow certain treatments, and for specific delineation of what is acceptable and what is not.

So I've talked about some of the causes of complexity, but what of the consequences? Well, we can start with difficulty for preparers in knowing how transactions should be accounted for, difficulty for auditors in figuring out what evidence can support management's conclusions, difficulty for users in understanding what financial statements mean. And complexity certainly leads, at least indirectly, to more errors and restatements.

But there's more. Last year at this conference, I spoke about the problems caused by a compliance focus to financial reporting, including a willingness — perhaps even a desire — to structure transactions around the accounting literature. I believe that the complexity of accounting standards is amongst the things that foster transaction structuring and lead to such a compliance mindset.

Fair Value

Sometimes when I talk about reducing complexity, I am asked whether I believe the use of fair value contributes to the problem. While I'll admit that fair value may at times be more difficult to apply than other measurement attributes, I believe that greater use of fair value can actually reduce complexity of standards, particularly with regard to financial instrument accounting. I've used as an example before the following simple question — "How does an investor in a debt security account for that investment?" Unfortunately, the answer to the question is that which of the four potential treatments will be applied depends on various facts, circumstances and choices available to management. It seems to me that asking users to understand all of this and factor it into their analysis is asking a lot.

The use of fair value has the potential to reduce the problems of complexity faced by preparers as well. FAS 13314 is often cited as one of the more difficult and complex accounting standards. But how much of that complexity could be reduced if all financial instruments were recorded at fair value? Consider — there would be no need to look for embedded derivatives if the whole instrument was already accounted for at fair value. And fair value hedge accounting, with its plethora of rules, documentation requirements, and other opportunities for error, would no longer be necessary. Indeed, a lot of restatements related to hedge accounting would never have happened if all financial instruments were recorded at fair value.

The SEC staff's Report on Off-balance Sheet15 activities expresses the view that accounting for all financial instruments at fair value is an appropriate goal. The report acknowledges, though, that many of the concerns about fair value have merit, noting that work should focus on addressing those concerns.

When we talk about concerns about fair value, we must return again to income statement volatility. Some believe that the income statement will become less meaningful, understandable or useful if more items that flow through it come from value fluctuations rather than completed transactions. And they could be right, if we don't clearly communicate which amounts come from which sources. The good news is that FASB's project on Reporting Financial Performance should help to relieve some of the pressure. The project team is considering various ways in which what we now think of as the income statement could be formatted, and various captions and subtotals that could allow for much more meaningful reporting of components of profit and loss. I hope that those who are concerned about the volatility that results from the use of fair value will provide substantive suggestions to the FASB on how to report performance in a way that meets their desires to separate changes in the value of investments from income elements relating to the sale of goods and services and other operating activities.

Beyond volatility, the other major category of concern with fair value is the reliability of the measurements, and/or the ability to actually do the measurement. And it is true that many items can be difficult to measure at fair value, and that a fair amount of judgment is needed to do it well. But a lot of work has been done in regards to determining the value of even the most complex financial instruments, and even the most difficult to value intangible assets. A couple of weeks ago, I spoke at a conference of the AICPA and American Society of Appraisers' business valuation specialists. These are the people who provide expert valuation services for intangible assets, businesses, and various equity and debt instruments. In my career, I have seen many things that make me skeptical of valuations that are used to support financial statement items. However, at that conference and elsewhere, I have recently encountered a growing number of valuation professionals who understand the role they can play, and who also understand the public interest mindset with which they must approach their work if they want to do it right. There are certainly a lot of things that can be improved about the infrastructure of the valuation profession, and accountants and valuation professionals need to work together to foster common understanding of how valuations can be used, but I'm optimistic that further work in these areas can make us all a lot more comfortable with fair value measurements. And, I might add that the disclosures that are planned as part of the FASB's final standard on fair value measurement should help a lot in communicating to users about the differing levels of reliability of various fair value measurements.

Interactive Data

Let me turn now to the promise that technology holds for improving financial reporting. The internet has fueled major changes in the way people receive and interact with data. When applied to financial reporting, it is easy to see the possibility for flexible and faster access to more reliable information. But the path to realize that possibility is somewhat unclear. Until recently, I didn't believe that the path that interactive data efforts in financial reporting, including XBRL, were taking held a lot of promise. But I'm now convinced that, whether through XBRL or some other format, interactive data promises to improve the process of financial reporting in the context of better, faster and cheaper access to data.

For those who don't know a lot about this technology, I'd ask you to consider the following simple example. Imagine that you wanted to compare statistics across a number of companies for, let's say, revenue growth. Right now, you'd need to pull up each company's filing, find the numbers and then key them into a spreadsheet to run the calculations. In the future, the use of tagged data can make this unnecessary. With tagged data, the software itself would go out and gather the revenue figures you need, by finding the appropriate electronic tags. What has previously been a painstakingly long manual process could become a quick, easy, automatic one.

For preparers, data tagging could help streamline the process of collecting, collating and confirming the accuracy of information from a variety of sources and systems. Additionally, it would enable efficient distribution of their data directly to investors and other stakeholders. For investors, tagged data offers a much better way to consume and analyze data. First, as in the simple example I gave, the data within reports becomes instantly usable without manual intervening steps. Second, tagged data offers the opportunity for users to easily personalize the data, focusing on information based on their preferences in a format they desire.

The Commission established an XBRL Voluntary Program in February of this year. The program was designed for flexible participation and as a mechanism to further evaluate the capabilities of interactive data in general, and XBRL in particular. I'd like to thank the participants we've had to date and encourage your wider support and participation, so that we can continue our assessment of the ability for this kind of technology to increase the usefulness of reported information. The accounting profession has demonstrated significant leadership in the development of the XBRL standard and providing us feedback to help structure the Voluntary Program on EDGAR. As we move forward we welcome your continued feedback and I'd specifically encourage you to contact the SEC staff listed in the final rule establishing the XBRL program with your comments.

Using Professional Judgment

Before I leave the podium to the rest of today's speakers, I want to bring up one more topic — the importance of professional judgment. This won't be news to anybody, but accounting and auditing are not exact sciences — not even close. To do either of them well requires the exercise of judgment. Later today, you'll hear many of the SEC staff members that speak discuss the importance of judgment with regard to a variety of accounting and auditing matters.

While accounting and auditors are generally quite good at making these kinds of judgments, in my 15 years in this profession, I've seen a couple of odd swings in how judgment is applied. During the internet bubble of the late 1990's and into 2000, it seemed that professional judgment was too often being applied in attempts to justify overly aggressive interpretations of the accounting literature, less audit work, and reduced efforts towards good external communication. We know where that led us.

And while we do still see aggressive interpretations of the literature, at times now it seems the pendulum may have swung in the other direction. Accountants and, particularly, auditors seem to sometimes be unwilling to use judgment to support anything other than what is perceived as the "safest" answer. With the scrutiny that has been placed on financial reporting, and on auditors in particular, it is not all that surprising that responses tend to lean towards doing more work rather then less, and to conservative interpretations of accounting literature. But while the current reactions may be predictable, they nonetheless make it quite difficult to move to less complex accounting standards that rely more on principles and less on rules. And the importance of professional judgment in regards to scoping the work to be done for internal control evaluations was a major subject of the guidance the staff issued in May of this year.

I hope that pendulum is ready to settle back towards the middle, where the intelligence, training, and analytical skills of practitioners will be applied neither in attempts to justify aggressive interpretations, nor to just look for answers that seem safe, but instead to apply the accounting literature in the way that will best communicate with investors. As I've said before, I believe that great improvements in transparency are achievable with no new standards, no new regulations, no new scandals to react to, if the focus of everybody involved in financial reporting could be on communication, on transparency, on plainly telling people what's going on and why.

I'll now step away from the podium and let others provide you with their thoughts. I hope that you'll be as attentive an audience for them as you have been for me. I look forward to answering your questions during the Q&A session later today.


Endnotes