Michael R. Jordan
August, 2009
The Fair Value, or so called “Mark to Market,” debate continues to rage, even though application of Statement of Financial Accounting Standards No. 157 (SFAS 157), Fair Value Measurements was required back in 2008. Financial guru Steve Forbes, former General Electric Chairman Jack Welch and now the United States Congress are all on record criticizing “mark to market” accounting and implicating it as a culprit in the collapse of the financial sector. Think of it, a change in accounting guidance has brought the world to its knees. We accountants must be a pretty powerful group! Would that we could end armed conflicts by using similar means. Of course, we all know there is plenty of blame to go around for the current economic state of affairs.
What is the present status of fair value accounting?
The truth is, “Fair Value” is nothing new. Accounting Principles Board Opinion No. 18 (APB 18) was issued in 1971 and it refers to fair value. SFAS 157 did not require any new fair value measurements, nor change any previous guidance that require or permit fair value measurement. The real change was defining fair value as an “exit price.” When an asset is acquired or a liability assumed, that transaction price represents an “entry price.” When an asset or liability is disposed of or transferred, that transaction is referred to as an “exit price.” If you think about it, entry prices and exit prices can be very different, just as bid and ask prices can be. This one change in concept has caused about as much tumult in the accounting world as the issuance of SFAS 133, Accounting for Derivative Instruments and Hedging Activities did back in 1998.
There are numerous arguments against fair value accounting. One such argument says that market prices do not always reflect the economic substance of a transaction, especially for assets held to maturity. This is actually a very common situation. For example, assume I have a U.S. Treasury security that pays a coupon interest rate of 1%. For at least the past year, market rates on these securities have been in the 5% to 7% range and, in management’s opinion, these interest rate levels will continue for the foreseeable future. Based on the SEC Staff Accounting Bulletin No. 59 (SAB 59), this security could very well be “other-than-temporarily impaired” and should be written down. I am sure that anyone holding a Treasury security fully expects to receive their total principal and all interest payments at the agreed upon rate of return in a timely manner. Has the economic substance of the security changed just because of a change in market interest rates? You be the judge.
A related viewpoint is that fair value is simply a form of liquidation accounting. If the company is marked to market, this is equivalent to the amount that would be received today to sell off all assets and settle all debts (we won’t get into fair value determination for liabilities, that’s an issue the Financial Accounting Standards Board (FASB) is still struggling with). To carry this full circle, if Company A is marked to liquidation value, then Company B transacting with Company A will eventually be forced to mark their Company A asset down to liquidation value. Then Company C transacting with Company B will have to take a markdown, and so on. You can almost visualize the dominoes falling if you think about the financial sector over the past year or so.
In countering these arguments, you’ve probably heard or read the statement that fair value is the most relevant measure for financial instruments. Measures, other than fair value, are not typically indicative of the effect current economic conditions have on an entity’s financial position. By providing new fair value guidance, financial reporting was to be more transparent. Other than the “exit price” notion, SFAS 157 was very much about developing and standardizing disclosures. As the hierarchy level goes up, the disclosure requirements increase. That does not necessarily mean a Level 3 valuation is any less valid than a Level 1 or 2. In fact, more work and thought and cost probably went into generating those (hopefully reasonable) valuations than the other two levels combined. Level 1 valuations are easy, right? Additionally, the attendant disclosures for each level would tend to maximize transparency, especially for those hard to value instruments.
Assumptions About Assumptions
But, believe it or not, fair value was never intended to be an automatic “mark to market” valuation, though it seems many interpreted it that way. How do I know this? Well, here are a few tidbits that are dead giveaways.
Right out of the gate, SFAS 157 muddies the water with “The transaction to sell the asset or transfer the liability is a hypothetical transaction at the measurement date, considered from the perspective of a market participant that holds the asset or owes the liability.” Any hypothetical transaction will necessarily involve a hypothetical price. There is nothing to base a price on but estimates, since the holder did not actually transact. Let’s face it, on many exchanges, even closing prices for actively traded instruments are derived from numerous transactions and are themselves, estimates.
Second, SFAS 157 states that: “A quoted price in an active market provides the most reliable evidence of fair value and shall be used to measure fair value whenever available.” This seems to be a reasonable premise given the market based approach to asset and liability valuation. However, what exactly is an active market? Generally, there are plenty of transactions such that reasonable market prices can be found through available trade information. Liquidity premiums are normally quite low and default premiums are consistent with the credit quality of the borrower. These attributes don’t always hold true in inactive markets.
Third, fair value is defined as “the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date.” The Free Dictionary (www.thefreedictionary.com) defines order as: “A condition of methodical or prescribed arrangement among component parts such that proper functioning or appearance is achieved.” This definition seems to describe many of the markets we are most familiar with, in normal times. However, we all know there has been a great deal of market disruption and the term “orderly” is not one I would use to describe a number of markets lately.
Fourth, SFAS 157 provides for three valuation techniques to determine a fair value. They are the cost approach, the income approach and the market approach. If fair value was intended to be solely “mark to market,” the FASB would not have included the cost and income approaches.
Lastly, my favorite quote from this guidance is contained in paragraph 30 which states: “unobservable inputs shall reflect the reporting entity’s own assumptions about the assumptions that market participants would use in pricing the asset or liability.” You read that right, assumptions about assumptions. Instead of a verifiable historical transaction, we may now use, not just assumptions, but assumptions about assumptions to value balance sheet items. You can’t get much further from a true market price than that.
An Example of the Dilemma
So, let’s look at an example of something many have faced over the past year (Table 1). Suppose I have an investment grade mortgage security with a face value of $100 that pays a coupon rate equal to LIBOR which is currently 5%, plus a 2% premium, or 7%. For simplicity sake, assume that a 1% change in interest, or discount, rate results in a $1 change in the security’s price. That will make it easier to relate rate of return to price. Research has shown that a normal liquidity premium in this market is around 2%, hence the premium being received above. I have recently determined, using the income approach and a beautiful mathematical model, that I will probably collect only 75% ($75) of my investment amount because of the deteriorating credit quality, or default risk, of the underlying mortgages. If the market has similar information, I should be able to sell that security for around $75 or a 32% discount relative to the total contractual cash flows (5% LIBOR + 2% Liquidity premium + 25% Default premium). I have been able to find two trades in very similar securities during the last quarter and they were both in the $25 range. That translates into an 82% discount rate relative to the total contractual cash flows (5% + 2% + 75%). If the market has become so inactive that the liquidity premium doubled to 4%, and that is a quite severe premium, what is the rationale for the additional 50% in discount? (82% - 25% - 5% - 2% = 50%) It appears the current market has become not only inactive, but completely uncoupled from the underlying economic and financial risk and return metrics of 5% for LIBOR, 4% for liquidity and 25% for default risk, or 34%. This is one of those instances where you might hear the term “market dislocation.”
Table 1.
|
| Purchase Price and Contractual Rates | Price Adjusted for Additional Default Risk | Price Adjusted for Additional Liquidity Risk | Price Seen for Market Transactions |
| Price | $100 | $75 | $73 | $25 |
|
|
|
|
|
|
| LIBOR | 5% | 5% | 5% | 5% |
| Liquidity Premium | 2% | 2% | 4% | 2% |
| Default Premium | 0% | 25% | 25% | 25% |
| Panic Premium |
|
|
| 50% |
| Total Discount Rate | 7% | 32% | 34% | 82% |
What is the additional 50% attributable to?
I don’t have an answer to that, a “panic premium,” maybe. But this is the kind of illogical pricing information investment holders had to deal with. Some markets became quite illiquid as demand dried up, but an additional 50% liquidity premium is not sensible. If you remember, the dot com boom resulted in market values in the opposite direction. Models using then current risk and return metrics could not approximate the very high prices seen in the market. Could it be that markets truly are not rational?
Audit Fears Come to Fruition
It’s very difficult to prove a point using a “lack of information” as your support. Have you ever tried to adequately document an inactive market? It’s nearly impossible. Proving that transactions in that market are distressed is equally challenging. One reason the original exposure draft of FASB Staff Position (FSP) FAS 157-4 presumed that transactions in inactive markets were distressed, unless they could be proven otherwise, was to help address that quandary. Because of this difficulty, some accounting firms were uncomfortable with clients designating markets as inactive even though the press, the government and the markets themselves continuously bombarded media channels with the fact that markets were inactive, distressed and dislocated. This effectively precluded the use of Level 3 valuations and in some cases Level 2.
We accountants know enough about finance to be dangerous. (Of course, some might say we know enough about accounting to be more than dangerous.) Many accounting firms utilized financial experts to assist in the implementation of the new fair value guidance. A number of these experts hailed from Wall Street. Firms leveraged this real world expertise to evaluate the pricing techniques clients were using. The problem here was that many of these financial minds perceived the valuation measures through purely market colored glasses. They didn’t fully understand the nuances of the accounting guidance and some auditors didn’t really understand all of the complex finance. Clients were also struggling to interpret and ultimately explain the bizarre risk and return metrics implied by the markets. This triad of communication channels, between auditor, client and expert, was filled with so much static it could not all be filtered out. The consequence was a number of severe impairment write downs of Held to Maturity and Available for Sale investments, some because of this miscommunication and resulting misapplication of the fair value guidance. In the end, conservatism, not neutrality, ruled the day.
FASB to the Rescue
To address all of the misunderstandings, FASB was forced to issue two additional clarifications, FSP FAS 157-3 on October 10, 2008 and 157-4 on April 9, 2009. These two FSPs provided examples and checklists for determining if a market is inactive and if transactions in that market are orderly. (How is that for getting away from rules based accounting? Maybe we are not ready for the principles based approach of International Financial Reporting Standards (IFRS)). Audit firms and their clients now have specific guidance to follow in their fair value determinations and impairment evaluations.
Will this reduce the impairment write downs we’ve been seeing?
Originally, SFAS 115, Accounting for Certain Investments in Debt and Equity Securities, required an other-than-temporary impairment be recognized as a loss in the income statement if the loss was probable. The loss equaled the difference between fair value and carrying value. The amendments to SFAS 115, which accompanied FSP FAS 157-4, require that only the credit, or default, portion of an other-than-temporary impairment be recognized net as a loss in the income statement, if the holder does not intend to sell. The loss, however, no longer has to be probable. Holders must now develop their best estimates of credit, or default, losses on all securities that meet their criteria for other-than-temporary impairment, which is no small task. This is in addition to the fair value measurements already required. We may end up seeing fewer large impairment charges and many more, but smaller, credit loss charges. What the total dollar value of these write downs will be remains to be seen.
Along with these clarifications, FASB provided additional guidance in the form of FSP FAS 107-1 and APB 28-1, Interim Disclosures about Fair Value of Financial Instruments, which was also issued in April. While a modest five pages long, it may significantly expand disclosures for some companies. In essence, fair value disclosures for financial instruments within the scope of SFAS 107 must now be included in the interim financial reports, rather than just annual reporting, of all public companies. Here again, the FASB felt that increasing the frequency of these disclosures would increase transparency by providing more timely and robust information to financial statement users.
Onward to Year Two
After some of the fits and starts we looked at, it appears this fair value thing may just survive the onslaught of critics. And after reading some of the comment letters on the recent FSP exposure drafts condemning FASB to the nether regions, that is probably saying something. It may be that this was the perfect economic environment to test it out, to shake out many of the bugs. Also, it doesn’t look like we’ll end up with anything like that 600 page “Green Book” of Derivatives Implementation Group (DIG) Issues, as we did with SFAS 133.
Already in the works are additional tweaks to the guidance, particularly in relation to the valuation of liabilities, and discussions on expanding the use of fair value in financial statements. But a number of industry groups still feel the principle is flawed. All in all, I believe we accountants have become a little more comfortable with this new conception of fair value, as have the users of the financial statements we produce. But I also think we will probably need some additional education in finance to further our understanding of how financial instruments can be valued. The good thing about this is we will be supporting colleges and universities that need it, providing jobs for teachers and contributing to the next economic upturn and resulting market rally!
To wind this up, 2009 brings with it the addition of certain nonfinancial assets and liabilities to the SFAS 157 domain. Generally, these will include assets and liabilities in a business combination, those tested for impairment under SFAS 142, Long-Lived assets, Asset Retirement Obligations and Exit or Disposal Activities. Their inclusion will add some new complications to the current valuation processes being employed. Some of them will have observable trading markets, though they may not be active. I suspect the predominance of valuations will involve discounted cash flow models, the income approach, and appraisals or replacement cost estimates, the cost approach. Given the experience we now have with financial instruments, hopefully this next round of fair value measurement implementation will go more smoothly.
Posts
