What SFAS 157 does, and does not, Accomplish

Authored by: Alfred M. King

The Routledge Companion to Fair Value and Financial Reporting

Print publication date:  May  2007
Online publication date:  August  2012

Print ISBN: 9780415423564
eBook ISBN: 9780203815151
Adobe ISBN: 9781136713101




There are very few finance professionals who are without a strongly held viewpoint on the topic of fair value accounting. Some (mostly financial statement preparers and financial analysts) dread the day that IASB and FASB formally adopt a fair value basis of accounting. Others (primarily academicians and theoreticians) are counting the days until we get rid of historical cost accounting and move to a total fair value (‘FV’) basis. Just from the way this paragraph is written, it is not hard to deduce that the author is predicting that accounting regulators, sooner or later, are going to shift to a FV accounting and reporting paradigm.

 Add to shortlist  Cite

What SFAS 157 does, and does not, Accomplish

There are very few finance professionals who are without a strongly held viewpoint on the topic of fair value accounting. Some (mostly financial statement preparers and financial analysts) dread the day that IASB and FASB formally adopt a fair value basis of accounting. Others (primarily academicians and theoreticians) are counting the days until we get rid of historical cost accounting and move to a total fair value (‘FV’) basis. Just from the way this paragraph is written, it is not hard to deduce that the author is predicting that accounting regulators, sooner or later, are going to shift to a FV accounting and reporting paradigm.

‘Beware the unintended consequences’ is this author’s advice to all participants in this ongoing debate. Full adoption of FV accounting, as and when it is adopted, will fundamentally change both accounting and auditing as we know it today. Security analysis will be significantly affected, as will be individual and institutional investors, albeit perhaps to a lesser extent.

Neither side of this debate – and it is a real debate within the accounting profession – is going to be happy with the FASB’s latest pronouncement, SFAS 157, Fair Value Measurements, issued in September 2006. Those looking for quicker adoption of FV accounting will be disappointed that the Statement merely tells one how to determine fair value, not when FV can or should be used in financial statements. Those opposed to FV accounting will be equally disappointed because the Statement appears to be one in a series of steps that will inevitably result in full FV accounting.

This chapter summarizes the Standard and its multiple changes in determining FV. It then comments on the proposed changes from the author’s perspective as a valuation specialist with 38 years of experience who has valued over $100 billion of assets during that time. Finally, it goes out on a limb and makes a prediction about how imminent fair value accounting really is.

What does SFAS 157 Actually Say?

SFAS 157 is explicit that it does not require fair values in any new area of financial reporting where FV is not now required. All it does is provide a common definition, with accompanying rules for how to determine FV. So people wanting FV accounting as quickly as possible will be disappointed that the Board did not move more quickly. On the other hand, those who fear FV accounting were not assuaged because it is clear that with an approved methodology the FASB will now be able to move forward and implement FV more quickly.

Now let us look at some of the specifics of the Standard. Discussed below is the fact that the Standard introduces a brand new definition of fair value, one that has far-reaching consequences.

The Statement ‘emphasizes that fair value is a market-based measurement, not an entity-specific measurement’. This too is discussed below in looking at the Board’s new definition of fair value.

The Statement establishes a ‘hierarchy of values’ with three levels. One should always use values determined from the highest possible level. Level 1 represents prices of items that trade in an active market (e.g. stocks quoted in the Wall Street Journal or Financial Times). Of course very few individuals or firms look to an appraisal specialist to determine the value of 100 shares of Microsoft or AT&T. For practical purposes, this ‘Level 1’ is a non-issue.

Level 2 represents the value of assets for which there are no directly quoted prices, but where adjustments can readily be made from quoted prices or other information. An example would be the restricted stock of a publicly traded company. The price without the restriction is clearly known and the required adjustment for the lack of marketability (the restriction) can be determined from comparable market transactions. Level 2 requires some professional judgement, but essentially is limited to the valuation of financial instruments, a subject outside the scope of this chapter.

That leaves Level 3, which represents the lowest level in the hierarchy, but one where the vast majority of valuations take place. For example, in a purchase price allocation the total purchase price is known, but the specific values of the software, trade name or customer relationships are not available from quoted sources. So an appraiser has to determine the values based on the best available information. The ‘market-based measurement’ rather than the ‘entity-specific measurement’ is to be used whenever possible. It is this requirement that represents the big change for financial statement preparers and appraisers and is discussed in the next section.

The Statement requires valuation of liabilities, including recognition of the risk or credit-worthiness of the firm with the liability. In practice, this means that if a firm has a liability, say, for future environmental costs, and its credit deteriorates, then the value of the liability will be reduced. Many observers believe this is counter-intuitive but it is a mechanical result of looking to the marketplace rather than to the entity itself.

The Statement explicitly precludes so-called ‘blockage discounts’, whereby a large block of stock is valued at a discount because trying to sell it all at once would itself depress the price. This requirement appears hard to reconcile with the emphasis on marketplace participants who in practice would suffer a price decline from trying to sell a large block of stock. The fact that this requirement is illogical with respect to the rest of the Standard is just one of the quirks of having accountants (FASB) tell appraisers how to do their job.

Finally, the Statement greatly expands the requirements for supplemental disclosures in the footnotes regarding the values of assets and liabilities at fair value. Again, it is outside the scope of this chapter to discuss the specific disclosure requirements and readers are referred to the FASB document itself. Suffice it to say that the length of total footnote disclosures will certainly increase as a result of SFAS 157.

The New Definition of Fair Value

For over 110 years, appraisers, lawyers, judges and businessmen thought they knew the definition of fair market value (FMV). While disputes could rage over the actual valuedetermination, virtually all parties agreed on a single definition. While taking many different wordings, all the definitions essentially read as follows:

Fair market value is defined as the price for which property would exchange between a willing buyer and a willing seller, each having reasonable knowledge of all relevant facts, neither under compulsion to buy or sell, and with equity to both.

The essential elements of this definition are:

  • willing buyer and willing seller agree on a transaction price;
  • both buyers and sellers are knowledgeable about the item being valued;
  • the transaction is voluntary for both parties – not a forced sale or liquidation;
  • the transaction regarding the asset is perceived as fair to both parties.

In most cases appraisers are determining ‘value’ in the absence of a transaction for the specific asset/assets. Therefore, in a buy-and-sell agreement the parties look to an appraiser to determine what the shares of stock would sell for if they were put on the market. The appraiser determines who the most likely buyer might be and then evaluates the asset in terms of its utility to that ‘buyer’. In effect, a specific transaction is contemplated or assumed and the appraiser provides his professional judgement as to the price at which both parties would be willing to buy or sell the asset.

In an allocation of purchase price we know the total payment made for the bundle of assets but then have to determine the specific values of all the individual assets which made up the business that had been bought. Again, there is a specific transaction willingly entered into between the parties, and the only differences of opinion under current rules prior to SFAS 157 will be whether the property, plant and equipment (PP&E) is well maintained and worth more, or poorly designed and maintained and worth less. Or, in the case of intangible assets, the appraiser has to determine the values of trade names and customer relationships for the buyer. We take into consideration the buyer’s plans for those assets (e.g. whether or not acquired software will continue to be used or discarded because the buyer’s software is better). We determine how much value is assigned to the software based on our understanding of the transaction.

Now let us look at the FASB definition of FV in SFAS 157:

Fair value is the price that would be received for an asset, or paid to transfer a liability, in a current transaction between marketplace participants in the reference market for the asset or liability.

On the surface this definition is similar to the standard definition of FMV quoted above, but there is one major difference that at first may be quite subtle. Instead of dealing with the proverbial ‘willing buyer and willing seller’, the FASB now uses the term ‘marketplace participants’. Is there a difference between willing buyers and willing sellers and marketplace participants?

The answer turns out to be ‘Yes’, there is a difference because of one more change inserted by the FASB. Now their definition of FV is supposed to be applied to the price that the seller would receive, not the price that the buyer would have to pay. Take a very simple example and this difference becomes apparent. A used milling machine has two separate values, one to the buyer and one to the seller. If you want to buy a milling machine you have to pay a used equipment dealer the price asked. On the other hand, if you have a surplus machine that you want to sell, the used equipment dealer will offer you much less. The reason is simple. The dealer has to make a profit and cover the costs of transportation, storage and cost of money. In the used equipment market there will often be a 100 per cent difference (or 50 per cent, depending on which you use as the base for calculation) between the price at which dealers will buy an asset and the price for which they will sell the identical asset.

Further, the emphasis on marketplace participants has a dramatic effect on certain asset valuations in a purchase price allocation. Take a manufacturer of original equipment (OEM) auto parts. There are only a dozen or so major auto companies in the entire world and most large OEMs sell to all of these buyers. Now if Company A buys another OEM, Company B, the issue is simple: Do the customer relationships of Company B have any value to Company A?

Most corporate officials, and almost all appraisers, would argue that since Company A is both larger and already selling to the 12 potential customers, Company B’s relationships with the purchasing staffs of those auto companies have little value. Company B is bought because of its production facilities and its product line, not because the Sales VP is a close friend of Honda’s Purchasing VP. After all, Company A already knows everyone it needs to do business with at Honda.

Thus, from the perspective of the buyer, the entity to use the FASB’s term, Company B’s customer relationships have little value and certainly were not the reason to enter into the transaction. But look at this customer relationship asset from the perspective of the ‘marketplace participant’. If Company B were acquired by an equity buyout firm, one with no expertise in auto parts manufacturing, the customer relationships established by Company B would be crucial. A firm such as Texas Pacific, Bain Capital or KKR would buy Company B as a financial investment and would rely on the existing management to continue selling to Honda, and so forth.

Consequently, the value of Company B’s customer relationships is strictly a function of who you assume the ‘marketplace participant’ is. If Company B would only be sold to another OEM firm, then we can assign a low value to customer relationships. But if Company B could, or might have, been sold to a buyout firm then the customer relationship asset is quite valuable.

Now the FASB thinks it has resolved this issue by requiring appraisers to determine who the ‘market participant’ is. If the appraisal firm makes a judgement that the only realistic buyer for Company B is another OEM, how do we respond to an auditor or a regulator who says, ‘Well, in my judgement, Company B could have been bought by a buyout firm’? The fact that Company B actually had sold to another OEM is interesting, but not dispositive.

The impact on Company A’s financial statements is going to be material depending on where one comes out on this issue; the more that is allocated to the customer relationship intangible, the higher will be the amortization expense over the next few years. If less is allocated to customer relationships, then goodwill, the residual, will be higher. Since goodwill is not amortized (although tested annually for impairment) Company A’s reported profit-and-loss statement will directly reflect the judgement on who is actually the appropriate marketplace participant for valuing that customer relationship.

The problem is that in one case we have a real buyer and in the other we have only a ‘hypothetical’ buyer. Determining values ‘as if’ something happened which in practice did not happen seems to this author as if we are entering an ‘Alice-in-Wonderland’ never-never world. This writer learned in Accounting 101 that one looks to transactions as the most solid evidence of value. Well, now we stop looking at actual transaction and start determining the value of what might happen, even though it did not happen.

Implications for Valuation Specialists

In performing valuations that will be used in financial reporting, valuation specialists will now have to look beyond the actual transaction and ask a number of questions about hypothetical transactions in order to determine what the mythical marketplace participants might do. Just a single example here will suffice to point out the problem.

Suppose a Picasso painting sold at Sotheby’s for $30 million. The winning bidder thought the painting, at least to him, was worth $30 million. The next highest bidder, the loser, had dropped out at $29 million. Now the winner comes to an appraiser and asks what his new Picasso should be insured for. Until recently most appraisers would have said $30 million, or perhaps even more because in case of loss he might have to pay even more for a comparable painting. But under the marketplace participant approach the value has to be considered $29 million. This is because there is nobody else in the world who is willing to pay $30 million, but we do know of a real live buyer at $29 million. Obviously, the underbidder is a ‘marketplace participant’. So far so good. But what about the buyer’s financial statements? Does he have a $30 million asset (cost) or a $29 million asset (based on marketplace participants)? If it is the latter, how do you explain the immediate one-day loss in value of $1million? The fact is that the buyer is not going to turn around and sell the painting; he paid $30 million and a comparable work of art would probably cost him the same $30 million.

Valuation specialists no longer look to the economics of the specific transaction, i.e. expected synergies to the buyer. Now we have to look at what other prospective (but not real) buyers might do if they had bought the target company, which they did not.

Some members of the FASB appear to understand this conundrum and are suggesting that allocations of purchase price under SFAS 141 should perhaps be based on the buyer’s actual transaction and assumptions. But then the values so determined would not meet the new SFAS 157 definition of fair value. Result? Perhaps SFAS 141 should be changed to indicate that allocations are not made at fair value!

Meanwhile, until the FASB resolves this issue we as appraisers will have to become highly creative and use a lot of imagination in reading the minds of marketplace participants who in fact have not acquired the asset. How do we explain this to our clients?

Even more important, how do companies explain to investors and creditors that they are taking a big impairment charge because the asset they acquired last year is not being used and has lost all of its ‘value’? Look at just one real life example. When Oracle bought PeopleSoft they stated very publicly that they were buying the customer relationships because they were going to migrate users to Oracle’s own software. Oracle had no use for and was going to disregard both PeopleSoft’s name and software.

Now in valuing the acquisition of People-Soft for Oracle, should significant value be ascribed to the acquired software and trade name? If PeopleSoft had been bought by KKR or Bain Capital, then those buyers would have kept, and used, the name and software. So if KKR and Bain Capital are the marketplace participants, we as appraisers would have to place large values on the trade name and software.

But in the hands of Oracle those same highly valued name and software have no value and they are not going to spend a nickel maintaining the value of either asset. In fact they bought PeopleSoft in order to get rid of the name and software! Under SFAS 157, however, we would place large dollar amounts on the Oracle balance sheet for the trade name and software; then a year later when the time comes to do the annual impairment test, guess what? The assets have gone down in value and within two or three years will be worthless.

Under SFAS 144 Oracle would test the assets for impairment, they would have zero value and they would now be written off. How does Oracle tells its shareholders and creditors, ‘Well we took a loss of $x billion this year, but don’t worry, we never planned to use those assets in the first place’?

One of the objectives of financial reporting is to provide useful information to investors and creditors about management performance, and future cash flows. How does the development of hypothetical values on the one hand, and the almost immediate impairment charge of those same values a year later, help investors and creditors in any way whatsoever? It is submitted that it not only does not help, it positively hinders informed analysis of management performance.

Worst of all, it is not the FASB but the valuation specialist who will bear the burden of explaining the new paradigm of financial reporting: ‘never-never land accounting’.

Implications for Accountants in Industry

Because we will be entering a new world of financial reporting as soon as SFAS 157 becomes effective (late 2007) it behoves financial officers to study prospective acquisitions closely. The accounting implications of a purchase transaction are no longer going to be straightforward and susceptible to common sense. As just one example, take legal liabilities.

Under current accounting in the United States (SFAS 5) liabilities are not shown on a balance sheet unless they can be determined and have better than a 50 per cent probability of being incurred. Now under SFAS 157, all liabilities acquired are going to have to be valued and placed on the balance sheet.

This means that if a company has been served with a lawsuit asking for $100 million in damages, but the company’s attorney says the plaintiff has only a 1 per cent chance of prevailing, the buyer still has to reflect a $1 million liability (1 per cent of $100 million). Maybe that is good accounting and maybe it is not. The one thing we can be sure of is that the buyer will never pay out $1 million in damages. The real answer is binary: either zero or $100 million will be paid out. So, assuming the buyer wins when the case is dismissed, under the new accounting it will now have a $1 million gain, which flows directly to P&L.

It does not take an experienced CFO to realize that in this case companies are going to ask their attorneys and appraisers to be ultra-conservative. Who knows, in the example above, maybe the plaintiff’s chances are 10 per cent, not 1 per cent. In that case one would set up a $10 million reserve that would flow back into income at some future date.

Once again we will be using ‘Alice-in-Wonderland’ accounting. In fact, we might require all entering accounting students to read that book prior to commencing accountancy studies!

My recommendation for CFOs and corporate controllers is simple, albeit self-serving: Talk to a competent valuation specialist before you get too far along on the due diligence effort for a prospective M&A transaction. Actually try and work out what the future accounting is going to be and the impact on both the balance sheet and the P&L. Perhaps this advance look will prevent some unpleasant (or perhaps pleasant!) surprises.

Implications for Auditors

Auditors are definitely going to be affected by SFAS 157. How does an auditor review the judgements made by management, and the valuation specialist, as to what hypothetical marketplace participants might do? Once one gets away from a real transaction between an actual buyer and an actual seller and go to ‘what might be’, there are going to be a lot of problems (for a detailed review, see Chapter 20). True, auditors currently have to review management judgements on bad debt reserves or inventory valuations. But in those situations at least the auditor can look at past performance and judge the reasonableness of current assumptions.

When we get to one-off unique M&A transactions – and have to speculate about who the hypothetical marketplace participants are and how they might behave – auditors are going to be in a weak position to assert that they disagree with the valuation specialist’s assumptions.

The former Chief Auditor of the Public Company Accounting Oversight Board (PCAOB) in charge of all US auditing standards said in a speech that in his opinion it was impossible to audit value information. Because the valuation specialist, at the end of the day, is using personal judgement, it is hard to audit that judgement.

Can you ‘audit’ a physician who diagnoses you with a stomach ulcer instead of stomach cancer, or vice versa? No, but subsequent events will either bear out the diagnosis or invalidate it. Similarly, a valuation will ultimately be borne out by a transaction or other evidence.

But it is impossible for me, as an appraiser, to ‘prove’ that I am right. I may have an excellent track record, be respected by my peers and contemporaries, and make a persuasive case as to why I came up with my value indication, but I still cannot ‘prove’ the answer the way an accountant can prove the original cost of a milling machine by going to the paid invoice.

The FASB has set up a hierarchy of values, so it will be possible to demonstrate that a Level 1 value is better than anything else – except that there are very few Level 1 values in corporate financial statements. Similarly, Level 2 will be better than Level 3, but as required in the Standard, it will still require professional judgement. In my experience there are relatively few Level 2 values in financial statements. So that leaves virtually all valuations for financial statements in Level 3, which effectively gets us back to square one in supporting the work of the valuation specialist.

The only way to resolve this issue is to accept the fact that valuation is an art, not a science. Accept that a valuation specialist must use his or her professional judgements. Accept the fact that an auditor can only review the valuation specialist’s reasoning and assumptions, and the auditor can either disagree or agree with what has been presented.

As more and more companies’ financial statements are based on fair value, the work of the auditor as we have known it is going to change and change dramatically. Perhaps auditors are going to have to study valuation and in effect train themselves to become valuation specialists. Just as there are auditors who specialize in information technology, so some auditors are likely to specialize in fair value.

Unintended Consequences of Fair Value Accounting

I have not addressed all the implications of going to a full-bore fair value financial reporting model because SFAS 157 does not yet go that far. The FASB and IASB have publicly stated, however, that fair value accounting is their goal. Issuance of SFAS 157 then becomes a necessary, but not sufficient, step on the way to full fair value accounting. The Boards can now say, as they move inexorably to FV accounting, ‘Well, you know how to do it, now we are simply extending the use of FV to one more asset category’. Some people are anticipating this ‘salami slicing’ as being how the two Boards will gradually introduce full fair value accounting.

The subject of full fair value accounting is outside the scope of this chapter, which deals with SFAS 157. But if this writer were forced to make a prediction it would be that fair value accounting will be with us within 15 to 18 years, i.e. before 2025. Historical cost accounting has served us well for the past 75 years, but its days are numbered.

The problems with fair value accounting are very straightforward. Changes in values of assets are going to be orders of magnitude larger than the impact of a business’s regular operations. A small change in external interest rates (determined of course by market participants) will have a magnified impact on the valuation of most intangibles. Assume the trade name Coca-Cola® is worth $50 billion, as many observers have stated. A 5 per cent change in the value of that intangible – some $2.5 billion in a year – would then totally swamp Coca-Cola’s reported earnings from operations. It should be noted that a 5 per cent change, either plus or minus, is very common in valuing intangibles.

A second problem with fair value accounting is that developing and displaying the ‘fair value’ of assets that the business has no plans to sell or turn into cash provides a misleading picture. The proponents of fair value argue that if shareholders ‘know’ the real fair value of the assets owned by the company in which they hold shares, they can then determine how well management is performing. But take the case of Microsoft. What is the value of its Windows® software? At least $50 billion based on current and projected cash flow. But can Microsoft actually sell the software in total to a single financial or strategic buyer? No, but Microsoft will continue to generate cash flows every year from individual buyers. If sales go up, the value of the intangible asset will increase. If the sales and cash flows decrease in a year, the value of the intangible that year will drop. But in either case the current sales and operating profit will tell shareholders and analysts what is happening. Of what additional benefit is it to shareholders and analysts to have Microsoft’s own evaluation of the value of its major asset?

In fact, many security analysts have been against fair value disclosures for that very reason. They want to analyse and tell their clients what they, the analysts, think the company is worth. If companies go to full fair value accounting, the role of a security analyst will be to second-guess management’s determinations of value. As discussed above with respect to auditors, trying to second-guess someone else’s determination of value is for practical purposes an exercise in futility.

To end this chapter, I predict that FASB and IASB will in fact march inexorably towards fair value accounting. It is also predicted that there will be serious unintended consequences and that both company managements and shareholders, as well as analysts, will then demand that companies provide, as supplementary information, financial information without fair value!

(NB. The author would appreciate hearing from readers as to their views on this topic. Contact him at [email protected])

Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.