O n a quiet Friday in New York—less than a month after that chaotic Friday when Bear Stearns’ stock went into a free fall—a panel of financial reporting experts met with about 200 corporate directors and other stakeholders to discuss the role of accounting in the market turmoil that began in August with the collapse of two Bear Stearns funds. The April 11 meeting was organized by the Directors Roundtable and moderated by Michael R. Young of the law firm Willkie Farr & Gallagher LLP. In addition to Young, the speakers included FASB Chairman Robert Herz; Michael Hall, a partner with KPMG LLP; Gordon Goodman, trading control officer for Occidental Petroleum Corp.; and Antonio Yanez Jr., a partner with Willkie Farr & Gallagher. All speakers acknowledged that their views are not necessarily the views of their respective organizations. The following is an edited transcript of the discussion.
I n securitization accounting, there’s been in place, as part of the rules, a device called a qualified special-purpose entity (QSPE). It basically was a notion that if assets were placed into a trust, a vehicle, and then interests were issued out of that vehicle to various forms of security holders, what are called beneficial interest holders; basically the form of that vehicle, that trust, was to collect the proceeds on the assets and then remit them to various security holders. They were fairly passive, and the rules talked about how the powers would be very limited—entirely specified up front—and I think that worked for a fair amount of time.
But I think what we’ve learned in the last three to five years is in residential mortgages (also to a certain extent in commercial mortgage space and some other assets) that these assets are not passive in nature. Certainly, the subprime assets that were put into these vehicles called “Q’s,” with a lot of hindsight, because they took a lot of management when they went bad in terms of the servicing or having to restructure the loans, modify them, do all sorts of workouts. That clearly was not intended. I think the lesson learned here is that they were not actually “Q-able.”
But that’s with a lot of hindsight. What are we doing? We’re going to kill Q’s. It doesn’t work, because people just did not use them for things that were just passive assets in the dynamic marketplace.
There are very few passive assets except Treasury bills, and some people have their doubts about those nowadays (just a joke!). So that’s one thing that we’re looking at, and that will probably result in a lot of the current securitization transactions not qualifying for sale accounting and remaining on the balance sheet.
If you do away with the Q, you’re left with just the regular SPE that doesn’t get this kind of hallway pass, and the rules there are called FASB Interpretation no. 46(R). These rules were put in place after Enron to deal with the problems revealed there. And these rules basically say “look to whether there’s a party that has a majority of risk and/or reward in the entity” and it provides a statistical approach to doing that based upon the variability of potential returns of the assets in the vehicle. And those kind of rules apply to what are now dubbed “bank conduits” and “structured investment vehicles.” And I think they’re an OK set of rules. With the over–optimism of the subprime era, the calculations that were done on alot of these were in some cases overoptimistic.
A lot of people thought things were OK until very recently. And so we’re going to have to re-examine those as well and probably tighten them up, probably say “just don’t rely solely on these mathematical techniques.” Also you’re going to need to have a very good qualitative look at exactly what’s going on and who the parties are. It’s been clear in some of these that the bank sponsor would provide a liquidity backup. Those were not considered to be particularly risk-prone until recently when the music stopped, and they basically became the only bank in town to finance the assets in these vehicles.
We are taking a very hard look at all of that and probably will, in addition to the mathematical techniques, really tell companies and their auditors to look much more carefully and comprehensively at all the arrangements and the potential arrangements that exist around these vehicles.
If a company has debt securities that are neither trading nor held to maturity—held to maturity securities are held at cost—but are available for sale; they would carry them on their balance sheet at fair value, but the change in the fair value from period to period does not go directly through earnings; it goes through something else called other comprehensive income, which is outside of earnings. All of this has been in place for eons, as have been the rules relating to down markets. Now herein lies the rub in a down market—in down markets, obviously things that you’re carrying on a markto-market basis on your balance sheet go immediately through earnings.
If it’s carried as available for sale or at cost, but the value has gone down severely for a prolonged period, then there’s what’s called an other-than-temporary impairment, and it has to be marked down to value at that point through earnings. In other words, if there’s been an impairment, you’ve got to get to the new level. And of course in a down market that’s what you get. Again, these rules have been in place for eons and have gone through many down markets.
First, there is a much more rigorous disclosure framework under FASB Statement no. 157, Fair Value Measurements . FAS 157 did not require any new items to be carried to fair value. What it said is, if the existing rules either require or permit fair value, this is what you do and this is what you disclose and the approaches to it. And it’s a much more robust disclosure framework. So these kinds of things were going on already, but they weren’t always disclosed.
Second, there are more assets now being carried at fair value. That’s because over the last few years we’ve, largely at the request of financial institutions, provided what’s called a fair value election for financial assets and liabilities . That is, at the inception of when you buy a financial asset or enter into one, you can elect to carry that at fair value and fair value it through earnings. In fact, that came into place in the beginning of 2007, and many major financial institutions voluntarily decided to elect fair value for a number of the classes of their assets and, in some cases, financial liabilities as well. So there are more of these on the books of large banks and even some smaller financial institutions that are now carried at fair value due to their elections that began in 2007 before the credit crisis.
Now the third thing that has happened is, of course, that a number of classes of assets that start with the subprime assetsdirectly and those have been securitized and then go up the ladders through the securities that are backed by these things through derivatives that are based on the instruments that are backed by these things.
I wouldn’t say they all fell immediately to Level 3. I think some of them are in Level 2 and some of them are at Level 3, or both. The hierarchy is not a strict hierarchy once you get out of Level 1. If you can find “an active market” for an identical item (Level 1), that is clearly the best view. If you can’t, then the idea is to come up with the appropriate techniques and data points to get to the objective, which is what would be the sale price right now in an orderly transaction.
That’s another myth about fair value. People say this is a liquidation price. Well, the objective is not a liquidation price in a fire sale. It’s the price you would get in a reasonable exchange between knowledgeable parties. Now, that’s often difficult to discern now. We have a distressed market for a lot of asset classes and discerning where something is liquidation or forced liquidation versus a regular somebody liking the price and buying it. Finding the right data points is hard and so is doing the underlying economic analysis because, essentially, what we’ve got is a bunch of extremely problematic assets supporting a whole edifice of complex securities.
It’s a difficult exercise, but it needs to be done. And the reason it needs to be done is because investors have, most investors have, quite clearly told us this is the information they want. Not exclusively, but if financial assets are going to be valued, that is the basis on which they want them valued.
There are a lot of subjective estimates involved. There are a lot of estimates involved in other accounting areas. This is not the first time or first set of assets—even financial assets—for which modeling techniques and getting economic underlying data projecting things have had to be used. For many, many years people have been doing it. Again, what is new is that people were caught unaware while the music was still playing. They had prices from a market—now they’ve got to do a much more difficult and intensive analysis. In cases where that is involved, disclosure becomes extremely important to the investors as to how those values were derived if there were ranges around them. And if the investors think that they’re going to hold to them and they think that they’re going to recover and they’re going to get more cash flows than would be indicated by the current value, they ought to talk about this as well. To that point, the SEC a couple weeks ago put out a letter, an advisory kind of letter, to people saying here’s the kind of things we would like you to disclose in MD&A in these kinds of circumstances, which I think is very useful.
W hat I’ve seen over the years has been each time there’s a market event, it’s usually a call for a change in the accounting. If you go back to the savings and loan crisis, what did it change? It changed a lot of accounting around allowance for loan losses as banks were wrestling with the underlying reserving for the loans. It also necessitated changes in internal controls as clients processed information to get their arms around what they have. It also started changes with respect to investment accounting.
This was the beginning of fair value making its way onto financial statements, where previously most investments were accounted for on a cost basis. Then you had a concept of some investments at fair value marked through P&L, and maybe those that you’re not sure whether you should have it at cost or fair value should be shown as mark to market through equity.
It was one of the first times where we began investment accounting by intent. You intended to hold it at cost basis if that’s how you managed your business. Market events after Enron necessitated more change in accounting for investments in entities to determine whether you control those entities or not, and we had FASB Interpretation no. 46(R), which tried to help companies understand and come up with a determination of whether they control these special-purpose entities. So we had new standards coming out. And then today we have market events, and then we have calls for changes in accounting.
When I look at FAS 157 dealing with fair value, it hasn’t caused a new requirement for fair value accounting. It really encompasses the whole balance sheet, not just financial assets. So what we’ll see over the next year is more concepts of FAS 157 exit-value pricing being analyzed and presented on the balance sheet in areas such as intangible assets, goodwill and trademarks. Anything that requires a fair value determination would be under this exit-price concept. So we’re seeing it in the financial instruments area in a distressed market now, and then we’re going to see it again in other areas of the balance sheet over the next year as companies adopt FAS 157
We will see a lot of changes in securitization accounting in the next year. We’ll probably have a new FASB standard, and then more changes in securitization accounting as we move towards international standards.
The IASB [International Accounting Standards Board] is also looking at the financial asset accounting model. So we have a period of change in front of us in addition to the market aspects relating to securitizations. Also in context of consolidation, we’ll also see changes to that model where today sometimes the consolidation analysis of structured investment vehicles and commercial paper conduits have been really based on a very detailed, analytical, quantitative judgment. Some of these judgments were made by a large Monte Carlo system calculation as to what types of risks a company had. If consolidation is only based on the risks and whether you have the majority of the risks, it depends on how you define what risks are relevant to the transaction. In some cases, you weren’t always measuring every risk that was involved. There were other elements besides the mechanical calculation, and, if we see changes in FIN 46(R), we may see a lot of assets coming back onto the books of sponsors of certain types of vehicles.
Now that won’t be particularly different from when we look across to Europe and international accounting standards. They have a different model, and a lot of the financial institutions in Europe have certain of their vehicles and commercial paper conduits consolidated onto their financial statements. So our system may not become that different from where Europeans are today.
I s there a real difference in the quality of earnings that are reported by different types of companies? If you looked at the income statements of Enron in the late 1990s and into 2000, most of what they were reporting—half of what they were reporting—was unrealized. It was related to mark-to-market or fair value changes. It didn’t mean that those are not necessarily interesting pieces of information, but they’re quite different pieces of information. And what this really points to is the distinction between things that are fixed and things that are floating.
But this has to do with both pricing—there are fixed prices, and there are floating prices. It also has to do with temporal issues: things that have happened in the past and things that have not yet happened in the future. And when you look at the question of the difference in the quality of earnings, I think there’s a significant difference between the things that have happened that are realized and are fixed and the things that have not yet happened that are unrealized and that are floating.
When you look at our current financial statements, if you pick up an income statement today, you’re looking at a combination—a hybrid—of both realized events and numbers and unrealized events and numbers. And at a glance it’s very, very difficult for the average investor or creditor to understand what percentage of the income that you’re looking at even today is realized and what percentage is unrealized. The same is true on the balance sheet. If you looked at the Enron balance sheet, a significant portion of the values that you saw reflected in the Enron balance sheet were unrealized. They were historical or previous changes at fair value, were prior periods but had not yet been realized. It may relate to contracts that stand 20 years into the future in some cases.
So in thinking about these questions, what I realized was that it would be very simple to pull the two apart and present it in a very clear fashion, transparent fashion, to investors and creditors. From the perspective of hard-asset companies, a company from Main Street in the United States, we can represent to the average investor and creditor most of what we’re doing, who we are, with an income statement showing only our realized values. We don’t need to incorporate the unrealized values to show you what’s going on with our company. That’s not to say that the unrealized events and the unrealized values aren’t of interest to us. They are. But we think it would be much clearer to investors and creditors to be able to see the realized income in a simple realized income statement and not have to pull it apart. The same is true on the balance sheet. We can pretty much tell you who we are in terms of our balance sheet by showing you realized values. And we think that also should be presented in a realized balance sheet format.
Now if you did these things, and obviously if you pulled all your unrealized values, your fair value type relations out of the income statement, out of the balance sheet, well, as an investor I would still want to know what’s happening on the fair value side. And so the suggestion is that all those calculations be presented, but separately, in a fair value statement, showing all the projected income, rises in changes from fair value from the current period, all the changes in fair value from prior periods reflected in basically a fair value balance sheet, but all that shown separately in a fair value statement.
Well, since I made this suggestion a few years ago, there have been some changes at FASB. And if you look at the proposed comprehensive income statement, some of these ideas are now reflected in it. And if you look at some of the disclosure requirements that are now being proposed around fair value measurements, you can get to the structure that I’ve talked about; but you have to still pull things apart to be able to see it the way I described to you. What I think really has happened in our financial reporting system is that there’s temporal confusion.
There’s literally confusion about the timing of events of when things have happened and when things will happen. In order to get to a better place in terms of financial reporting, we need to clarify that so that when you pick up a financial statement you have to literally know whether you are looking at past events, which should be, in my mind, on the balance sheet.
Are you looking at present or recent events, which should be on the income statement? Or are you looking at future events, which should be on the fair value statement? This is a modest proposal. So having talked about the question of dividing things up between the past, the present, and the future, let’s talk a little bit about fair value.
If you were to present things separately in a fair value statement even inside of those events, those unrealized future events, there’s a significant difference between things that are fixed and things that are floating—things that are hard and things that are soft. And that difference is discussed a little bit in the idea of the fair value hierarchy 1, 2, 3. Things that are exactly the same, identical, Level 1. Things that are similar or analogous, Level 2. Things where we don’t have observable market prices, Level 3.
I would agree that in terms of measuring financial activity, fair value is best. In fact, it’s the only useful way to think about that. But for companies that are involved in manufacturing or extraction of minerals, some of the older ideas around accounting are really quite useful and, perhaps, would be a better way to think about how those companies are doing. I would also note that, if we had made this differentiation between realized and unrealized values, hard earnings versus fair value statement presentations, a lot of the changes that we’re currently seeing for the financial companies would be relatively minor events on their income statement, their balance sheets if those were just realized statements, and the changes you’d be seeing would be over in what I call the fair value statement, which are in the future. Those would be estimates, some of which are very good, some of which are very hard, some of which are relatively fixed, but they haven’t yet occurred.
And I think it would tend to pull some anxiety down from the kind of changes we’re seeing in the current marketplace where those changes flow immediately through the income statement, flow immediately over to the balance sheet. I think it would be a more useful way to think about these as a series of estimates, which is what fair value calculations really are. Some of which are very good estimates, but they’re not yet fixed. Even for something that we know the value of, like a stock or a bond or a commodity, until you pull the trigger, until you actually sell it and close out the transaction, the value is going to change on a daily basis. Whereas things that are fixed won’t change any time in the future.
F rom a litigation perspective, I see three key implications flowing from fair value accounting under FAS 157 and the related standards. The first is that we are likely to see an increase in accounting-related litigation. The second is that the judgment of financial statement preparers and auditors is going to be front and center in that litigation. And the third is that fair value accounting and other trends could bring to a head a conflict that’s been brewing for some time between the evolution of financial reporting on the one hand and our litigation system on the other.
The first implication is the increase in accounting-related litigation. And let me start by taking a step back.
Over the past few years we’ve seen a steady downward trend in accounting lawsuits (securities class actions that center on accounting issues). Different explanations are offered for that trend. Some people say that Sarbanes-Oxley has wrung manipulation out of the system. Others focus on improvements by the accounting profession. But there’s one factor that just about everyone has cited in common, and that is the relative absence of volatility in stock prices over recent years.
Increased volatility is the byproduct of fair value accounting that is likely to lead to more litigation. Quarter after quarter, investors will be getting more information about changes in the value of certain assets, and they will be in a position to react to those changes. And as they do, stock prices will go up and they will go down. And to the extent stock prices do go down, there will be those that assert that the stock price drop was caused by the market having earlier been misinformed, and they will have an economic incentive to make those assertions in litigation.
By the way, fair value accounting is not the only development in accounting and financial reporting that could lead to more volatility. Proposals to provide investors with more real-time information, with more and better forward-looking information, and other proposals could lead to increased volatility and increased litigation as well.
But sticking for the moment with fair value accounting, the fact of the matter is that we’re already seeing an increase in litigation owing to the subprime situation. I saw a report earlier this week that there were 163 securities class actions filed in 2007. That represents an almost 50% increase from the number filed in 2006. And expectations are that the earlier downward trend has now been reversed.
Now, I’d like to spend just a moment on the types of allegations being made in the subprime cases that are going to resonate and that we’re going to hear in other lawsuits related to fair value accounting going forward. These include that write-downs should have been made earlier than they were, that there was a failure to disclose exposure to whatever it was that was written down, that there should have been warnings about the write-down before the write-down was taken, and that there was a failure to maintain systems that would have limited exposure to what was written down.
I would also add that, depending on the circumstances, it might not only be write-downs that are challenged in litigation; write-ups could be challenged as well. One can imagine a situation where there is a write-down that is followed sometime later by a write-up as asset values increase. And then allegations could be made that the initial write-down was too big; that it was a big bath charge or part of some other manipulation; that the company was taking the write-down in order to stock away some value that it might need at some point down the road. So it’s all aspects of volatility—upward volatility and downward volatility—that may contribute to an increase in litigation.
Now, I want to emphasize that, in saying that litigation is going to increase, I am not saying defendants are going to lose more. There are going to be defenses, but the main point is that an increase in litigation is likely.
The second litigation implication of fair value accounting is that the judgment exercised by financial statement preparers and by auditors is going to be front and center in litigation. And it’s going to be front and center from the very early stages of the lawsuit.
The reason there’s going to be a focus on judgment has to do both with FAS 157 and with more general trends in accounting and financial reporting. As to FAS 157, there’s obviously not a lot of judgment in a Level 1 valuation. A market price for an identical asset is what it is. But there is judgment in Level 2 and Level 3 valuations. For example, there is judgment in determining whether markets are sufficiently analogous in making a Level 2 valuation. And there’s obviously judgment involved in building valuation models in Level 3 situations.
These types of judgment calls, like all judgment calls, can be second-guessed. And they will be second-guessed aggressively. The basic allegation could be that a write–down should have been made earlier than it was. And to support that allegation, plaintiffs might assert, for example, in a Level 2 case that markets considered to be analogous for valuation purposes in fact weren’t analogous enough to yield a fair valuation. In a Level 3 case, plaintiffs could argue that models were based on improper assumptions or that they had improper inputs or that they were poorly constructed and so on.
And this focus on judgment is only going to be heightened given more general trends. As accounting and financial reporting move more towards principles—which require the use of judgment—so, too, will accounting and financial reporting litigation focus increasingly on judgment.
Now, as to why the focus on judgment will begin from the early stages of the litigation, that has to do with recent Supreme Court precedent. In a securities case decided just last year—the lawyers in the room will recognize that I’m talking about the Tellabs case—the Supreme Court basically held that, before a securities case can get to discovery, a court must weigh the allegations in the complaint and come to a view as to whether they suggest fraud or the absence of fraud. Where the suggestion of fraud is at least as strong as the absence of fraud, the case goes forward. Otherwise it doesn’t.
What that means in fair value, and for that matter all other accounting-related securities lawsuits, is that courts at the outset will be called upon to consider the accounting judgments that were made in connection with whatever accounting is being challenged; and courts will have to come to a view early on in the lawsuit whether it appears from the facts as presented in the complaint that the accounting judgments were tainted by fraud or whether they appear to have been made in good faith—that is whether the facts suggest that absence of fraud.
Now here, too, I want to emphasize that this focus on judgment isn’t necessarily a bad thing for defendants. Pointing to good judgment exercised in good faith can be very effective with juries. It can be a very good defense, but that judgment is going to be questioned.
That, finally, brings me to the third implication of fair value accounting, and that is that fair value accounting—as well as the trend toward principles and other trends—really bring to a head the conflict that I mentioned earlier between the evolution of financial reporting on the one hand and our system of litigation on the other.
Fair value accounting is one aspect of an evolution in financial reporting which seeks to give financial information users more timely and more useful information. It is intended as an improvement. At the same time, our system of litigation will permit plaintiffs to second-guess the judgments that are made in developing that information, and you’ve heard me say that will result in more litigation overall.
Now, you’ve also heard me say that more litigation doesn’t necessarily mean more defense losses and that the focus on judgment isn’t necessarily a bad thing from a defense perspective. But the fact remains hat the process of litigating is painful. It is expensive, and it is disruptive even if you end up winning at the end of the day.
So there’s the conflict. On the one hand, we’re seeing an evolution in financial reporting that’s intended to benefit the users of financial information. But on the other, given our system of litigation, that evolution will likely result in more litigation and subjecting those that prepare financial statements to the pain and expense and disruption of that litigation even, again, if they end up winning at the end of the day. And the question is, is anything going to be done about that? There have obviously been calls for reform for some time, but I think that fair value accounting will bring the need for reform into focus like never before.
I also want to mention, just before I conclude, a key player in all of this that I’ve not mentioned to this point and that is the SEC and, in particular, the Commission’s Enforcement Division. How Enforcement is going to factor fair value accounting into its program is anybody’s guess. For that matter, what role the SEC is going to have going forward more generally is, at least based on the newspapers, subject to some debate at this point.
But I will say this. Enforcement has tremendous power here. We obviously need robust enforcement where judgments are made in bad faith or without sufficient inquiry into the underlying facts and circumstances or otherwise improperly. But if people are penalized for good faith judgment based on diligent inquiry—even if somebody can come back sometime later and say that those judgments were wrong—the evolution of financial reporting that I’ve been talking about will be stopped dead in its tracks. No responsible accountant or auditor is going to make difficult judgment calls when doing so could be a career-terminating event.
Question: (To Herz) Fair value, generally, in the broadest sense, where do you see it going in the future?
Herz: Some nuances might change. I don’t see it being spread much further beyond where it is now. We have a conceptual framework project jointly with the IASB that includes a key part on measurement attributes in accounting. The issue goes into various cost models, and cost is not always objective either. Fair value is one version of a current value model, but it’s not the only one. If the IASB makes some tweaks to FAS 157 that make sense, FASB will make the same kind of tweaks. Most users have urged us, at least in the financial instruments space, to move to mandatory full fair value; but I don’t think we’re going to do that next year.
Question: (To Herz) First, you mentioned in your talk that the solution of variable issues in FASB Interpretation no. 46(R) was an improvement on the standard that was used at the time of Enron, where Enron gamed the system. It would appear that FIN 46(R) for many companies, particularly financial institutions, was used to game the system as well. How did FASB allow this to happen?
Second, we have a tendency in this country to game the system using accountants, using the financial experts, using the consultants and work all the way around these rules. Now we’re moving toward an international system, where we have to use principles. If we’re used to gaming the system on the one hand and now have to move to principles on the other, do we have the right mind-set for this?
Herz: I think 46(R) was a definite improvement over the 3% rule that existed before. The new standard we’re working on is still going to employ the backstop of calculations, but it is going to front-end it with some more characteristics of looking at exactly what’s going on and to whom. It is actually going to put more responsibility on the companies and the auditors, but probably less wiggle room using mathematics and the like.
We can see from studies performed after FIN 46(R) was enacted that hundreds of billions of dollars did go back on balance sheets and more hundreds of billions of dollars out of SPEs—in essence the SPEs were unwound.
In regard to moving toward more principles, I think the SEC will take up what’s called a professional judgment framework. It will have some rigor in it as to what is expected in exercising a judgment, including what information, where you got the information, did you consider the alternatives, did you discuss it, and did you make choices that were based on trying to give a good report rather than make the earnings per share for the quarter?
I’m not an eminent legal scholar, but there is a belief by some that if this is put into policy by both the SEC and adhered to by the PCAOB, including their enforcement people, that the courts might start to respect that kind of approach to things. I think one of the biggest impediments to a lot of progress in our country is litigation. This is because everybody is too scared to exercise reasonable judgment, therefore, they want a rule. The good people try to comply with the rules, but the bad people try to use it to structure transactions around the rules.
Question: (To Young) We just heard more litigation, more judgments, more potential stock market volatility. What we haven’t heard is what do you advise directors to be doing in this new environment?
Young: If you’re a director, there’s only so much you can do. There are two things in particular, though, that directors can do. One is to set the right tone. Work hard to have the organization in general, the senior executives and the accounting staff in particular, to embrace transparency. The way you get into trouble is not by making judgment calls. True, there may be more litigation, but the way you get into trouble is by distorting judgment calls and not being transparent about the basis for the judgment call, which is one reason that FAS 157 is such a contribution—because you not only put forth the value, but you put forth the basis for the value. Then, second, you can encourage your company to put in place the infrastructure so that the transparency that should be part of the culture can be achieved as a matter of logistics.