While I generally think highly of the The Economist, this article contains one of my pet peeves, which is a loose discussion of fair value. In particular, the article states that “In 2009, under its previous chairman, Bob Herz, FASB narrowly voted 3-2 that all assets should be booked at fair value on banks’ balance-sheets” (emphasis added).

Actually, this vote was part of the boards’ financial instruments project and so would not affect the accounting for any nonfinancial asset. Some might say, well, most bank assets are financial assets. That’s fair, but banks also hold nontrivial amounts of nonfinancial assets (e.g., Bank of America had over $100 billion of PPE, goodwill, and other intangible assets at the end of 2010).

The bigger point is that a casual reader might mistakenly infer from this quote that the FASB was on a mission to extend fair value accounting to all assets for all companies. That is certainly not the impression I have ever gotten when listening to the boards deliberate on their various projects. And the inclusion of one word could help avoid misunderstanding.

Oh well, I’ll forgive The Economist this time…

The answer to the latter is a clear “yes” because the new guidance requires substantially more disclosures for level 3 than for level 2.  This led to some very interesting discussion, which unfortunately was not archived.  In particularly, if you look at the first slide deck that Phil posted, slides 6 and 7 discuss the additional disclosures.  The reporting entity will group assets into categories, and for each category, the company must disclose the type of valuation technique used and the unobservable inputs.  Ben mentioned a few of FVM approaches currently used by companies:

non-binding brokers quotes:  The reporting entity gets an estimated price from a broker.  Since the broker is not offering to buy at the quoted price (hence the term “non-binding”), the broker might spend very little time preparing the FVM.  Another problem is that the reporting entity must ascertain details underlying the broker’s methodology in order to comply with the new disclosure requirements, but brokers are hesitant to provide information about their proprietary pricing models.

in house valuation experts:  For some reporting entities, executing the company’s business model requires frequent FVM’s.  These companies may hire valuation experts for other reasons, and as a result, will generate its own FVM for financial reporting.

binding broker quotes:  The broker promises to buy the asset for the quoted price (quote expires after a few days).  This almost sounds like level 1 FVM to me.

Some other FVM approaches were mentioned briefly, but I do not recall what they were.

Academic accounting researchers have shown great interest in companies’ accounting choices.  At first, the new FVM guidance that Ben discussed appears to mandate accounting policy, which makes it a poor venue for studying accounting choice.  But what seems unique is that firms can choose how to comply with the mandate (measure FVM using non-binding broker quotes, in house estimates, binding broker quotes, or multiple methods), and the firms are asked to make fairly transparent disclosures about these choices.

I asked Ben what factors are expected to drive the choice of FVM method.  One factor is whether in house valuation expertise is available, because if this expertise does not already exist, he doubted entities would develop it just for complying with the new guidance.  Second, he expected market pressure over time might influence the choice; the hypothesis (although he did not label it this way) went as follows – companies initially disclose a methodology that is pretty weak and subject to significant errors (e.g., non-binding broker quotes or FVM’s prepared by in house experts using questionable inputs), the market applies a larger cost of capital to those companies, the companies improve their FVM methods and disclosures, and the cost of capital falls.

Obviously an archival test of this hypothesis will have to wait until we have a long enough time-series of disclosures.  But I expect some academics will be looking closely at the new disclosures, either as preparation for future research or for interesting class discussions about accounting disclosure choices.

Our next FASRI roundtable is scheduled for 4pm-5pm ET Wednesday, November 17th.  Richard Frankel (Washington University) will be joining us to talk about his recent paper, Panacea, Pandora’s Box, or Placebo: Feedback in Bank Mortgage-based Security Holdings and Fair Value Accounting.  The paper continues several themes that FASRI roundtable participants have seen before.  Like Bertomeu and Magee’s paper in the last roundtable, the paper is from the most recent JAE conference, and also like that paper, this one addresses a process by which the choice of accounting methods can have real impacts on future value creation.

In Bertomeu and Magee, shocks to asset values create pressures to change accounting standards, which in turn affects value creation.  The present paper is based on work that Haresh Sapra discussed at a roundtable in Spring of 2009.   We don’t have an audio archive from Haresh’s session, but this will give you a sense of Haresh’s arguement.

Pedestrians on the bridge react to how the bridge is moving. When the bridge moves from under your feet, it is a natural reaction to adjust your stance to regain balance. But here is the catch. When the bridge moves, everyone adjusts his or her stance at the same time. This synchronized movement pushes the bridge that the people are standing on, and makes the bridge move even more. This, in turn, makes the people adjust their stance more drastically, and so on.

In other words, the wobble of the bridge feeds on itself. When the bridge wobbles, everyone adjusts his or her stance, which makes the wobble even worse. So, the wobble will continue and get stronger even though the initial shock (say, a small gust of wind) has long passed.

What does all this have to do with accounting standards and financial markets? Financial markets are the supreme example of an environment where individuals react to what’s happening around them, and where individuals’ actions affect the outcomes themselves. The pedestrians on the Millennium Bridge are rather like modern banks that react to price changes, and the movements in the bridge itself are rather like price changes in the market. So, under the right conditions, price changes will elicit reactions from the banks, which move prices, which elicit further reactions, and so on.

Rich will be discussing evidence that supports such feedback effects.  From the first and last paragraphs of his paper’s introduction:

We study how changes in commercial bank mortgage-backed-securities (hereafter MBS) holdings relate to changes in MBS prices and mark-to-market accounting during the Financial Crises of 2007. Our purpose is to understand how security holdings influence the role of commercial banks in providing liquidity during the crises. Theory suggests banks’ security holdings add systemic risk to the economy, because banks can be forced to sell securities when prices fall (Shleifer and Vishny, 2009, Allen and Carletti, 2008, Brunnermeier and Pedersen, 2009) and that mark-to-market accounting can accentuate this “feedback” effect (Plantin, Sapra, and Shin, 2008, Allen and Carletti, 2008). In these models, liquidity constraints cause prices to deviate from discounted-cash-flow values and precipitate sales. Therefore, we examine bank holdings in the face of declining prices to understand whether banks increased holdings of securities (as would be expected of liquidity providers) or whether they sold securities suggesting banks were overwhelmed by the financial avalanche.

Overall, our study provides evidence that banks are more likely to sell MBS when liquidity declines. Furthermore, we provide evidence consistent with the easing of mark-to-market accounting rules limiting banks’ feedback trading and thus enhancing shareholder wealth. This study lends credibility to the belief that financial reporting rules have real effects. The pressure placed on the FASB to alter the valuation of distressed assets and change the accounting treatment of unrealized losses is indicative of these effects. Our tests suggest a specific consequence for a bank’s trading activities.

Join us for what promises to be a very interesting discussion.

You can join us on the web (Windows only, no Macs) or phone in.  Details on how to participate are here.

UPDATE:  SLIDES FOR THE PRESENTATION ARE HERE.

UPDATE:  GRAPH FOR FEDERAL RESERVE BALANCE SHEET HERE.

UPDATE:  AUDIO FOR THE SESSION IS HERE.

Of course, with such a significant change, there are bound to be questions and concerns. Here are a few that have occurred to me.

1. One of the debates we had in the early days of this project was whether to define revenue, or just address how revenue should be recognized. As this exposure draft makes clear (see paragraphs 1-2), the boards ultimately decided not to define revenue, but instead only to address when revenue is recognized in contracts with customers. This made me wonder whether the boards will be happy with revenue being recognized in the absence of contracts with customers. Does revenue recognition require a customer? Apparently not. Revenue is recognized today for changes in the value of some mineral, biological or agricultural assets (IAS 41), even in the absence of a contract with customers. Why do we get a difference for revenue recognition principles across various industries or types of assets? The proposed new standard scopes out non-contractual situations (such as biological and agricultural assets) as well as contractual situations (such as leases, insurance, financial instruments, and guarantees). Conspicuously absent from the exposure draft is any rationale for scoping these areas out of the proposed standard. From experience working on this project, I’m guessing that’s because there was no general agreement about WHY these would be scoped out. There is simply a consensus that they should be scoped out. Wouldn’t it be great if the boards could provide some rationale for scoping out particular types of transactions?
2. The exposure draft states that an entity shall account for each promised good or service as a separate performance obligation only if that good or service is distinct (see paragraph 22). A good or service is distinct if either (a) an entity  sells an identical or similar good or service separately, or (b) the entity could sell the good or service separately because (i) it has a distinct function AND (ii) it has a distinct profit margin. My question–how do you know whether something has a distinct profit margin unless you (or someone else) actually sell that something separately? And if you already sell it separately, then there’s no need for condition (b). The exposure draft states that a distinct profit margin exists if that good or service is subject to distinct risks and the entity can separately identify the resources needed to provide the good or service. This is one area of the proposed new standard that I think people may have some difficulty with—determining whether a potentially separate good or service has distinct risks. If there’s anything our recent past has taught us, it’s that people are pretty bad at identifying and quantifying risks. Who knows, perhaps it won’t be so hard to do for revenue recognition purposes! (By the way, don’t misinterpret my tone here…I actually like the direction the boards have gone with this guidance, but I think it will be a little perplexing at first.)
3. Determining the transaction price in a contract receives lots of attention in the exposure draft. The boards decided that the transaction price should be adjusted based on the time value of money (if financing is a significant component of the arrangement), customer creditworthiness, collectibility, any non-cash consideration received from the customer, and any consideration payable to the customer (such as rebates). I don’t really have a question here, but I do want to highlight how this will be a significant departure from past practice in some situations. Previously, revenue would not be recognized in some situations if certain criteria were not met, for example in real estate sales. If the criteria were met, revenue for the full contract amount would be recognized. Instead of asking whether to recognize revenue, this standard moves more toward asking how much revenue to recognize. So, for instance, if you have sufficient history selling real estate to particular classes of customers, and your history suggests to you that 60% of the time customers follow through with their promises to pay over time, then either through discounting with a high interest rate or through an adjustment due to collectibility, you would recognize the sale of real estate at a drastically reduced amount. Any amount received beyond the original revenue recognized would be treated as a gain or loss. Personally, I think this is an improvement over the criteria approach, but it has its own problems as well, including the estimation process required to determine the amount of revenue to recognize in such a sale.

I’ll stop here for now. I still want to go through the application guidance with a finer tooth comb, but I’m pretty impressed with most of what I’ve read. As much as the project was originally conceived to be a general standard on revenue to replace all other revenue standards, I think the boards have appropriately scaled back their ambitions to focus strictly on settings where a contract with a customer exists. This represents the lion’s share of situations in which revenue is recognized today, and the proposed new guidance will not change most accounting out there. However, what it does change could be pretty significant—including construction contracts, real estate sales, and software revenue recognition. I look forward to reading the comment letters over the next few months!

The basic argument tying fair value accounting to amplified procyclicality is that auditors required banks to write down affected assets to unrealistically low values as reflected by ABX index prices. Because the Financial Crisis caused a drop in liquidity, ABX index prices allegedly reflected distressed prices rather than prices from an orderly market. Bank managers contended that ABX prices, being artificially low relative to the bank managers’ perceived asset values, caused unnecessarily large impairment charges. That is, impairment charges would have been lower had bank managers been permitted to use their personal assessments of value, and the economy would have suffered a less severe downturn.

Although, in principle, an “excessive” fair value-related impairment charge could have amplified procyclicality of bank asset prices, we believe this is unlikely for two reasons. First, this claim can only apply to those bank assets that were either measured at fair value or for which fair values apply when determining impairment. The proportion of bank assets for which this is the case is limited. Laux and Leuz (2010) reports that during the 2004 to 2006 period banks held approximately of 50% of their assets in loans and leases, which are not subject to fair value accounting and are not impaired to fair value. Although, for the 14 largest US commercial banks, Shaffer (2010) reports the decline in Tier 1 capital during the Financial Crisis arising from impairments of loans averaged 15.6%, those impairments were based on an incurred loss model and not on fair value. Therefore, as discussed in section 6, although impairments of loans, i.e., loan loss provisioning, during the Financial Crisis likely had procyclical effects, fair value accounting played no role relating to loans.

In other words, there was a downward spiral in loan values, but fair value accounting played no role in it because loan values are not calculated using fair value. They are calculated using the incurred loss model.

The authors go on to explain that fair value accounting may have played some small role in the devaluation of other assets, however, bank regulators apply a prudential filter to neutralize some fair value gains and losses when calculating Tier 1 capital requirements. In fact, “Prudential filters neutralized the effect on Tier 1 capital of some fair value losses and the larger effect on Tier 1 capital arose from loan losses that were not determined using fair value.”

The authors also discuss the role that asset securitization, derivatives, and loan loss provisioning played in the financial crisis. The paper is forthcoming in the European Accounting Review.

“June 30, 2011, is an arbitrary deadline and it’s not one that’s been put in place by the SEC or by our road map,” said Kroeker. Citing FIN 46(R) as an example of an accelerated project that later needed to be reworked, Kroeker said that what’s most important is to ensure through the exposure process that the final standards are a “long term, sustainable solution.”

I suspect the FASB is not all that surprised by Kroeker’s view, given how good the lines of communication typically are between the FASB and the SEC. However, I suspect the IASB and other supporters of a single, global accounting standard will be a little surprised and will interpret Kroeker’s comments as (further) evidence that the US will not be adopting IFRS any time in the near future. They may even increase the volume on their arguments that the IASB should not give so much preferential treatment to the FASB and SEC in its standard setting activities. Should make for some interesting articles over the next few weeks!

Stephen Ryan (NYU) joins FASRI to talk about his recent research on the fair value option in the banking industry.  Stephen is one of the most respected accounting academics studying banking and financial instruments these days, so I expect people will have plenty of questions beyond the research study that will form the heart of the discussion. From his web page, we learn that:

Stephen has been actively involved in financial accounting standards setting. He currently serves on the Financial Accounting Standards Board’s Liabilities and Equity Resource and Financial Institutions Advisory groups. Until 2003, he was a member of the Financial Accounting Standards Advisory Council, the advisory body for the Financial Accounting Standards Board. He has also previously chaired the American Accounting Association’s Financial Accounting Standards and Financial Reporting Issues Conference committees. Professor Ryan has served as the editor of the Review of Accounting Studies since 2006.

Definitely take a look at his SSRN author page.

Here is the abstract of the paper we’ll be discussing, with Chi-Chun Liu and Yao-Lin Chang:

We examine the determinants of the timing of and financial instruments involved in banks’ initial fair value option (“FVO”) elections upon their adoption of SFAS No. 159. We focus on regular (in 2008:Q1) adopters of the standard, and distinguish their FVO elections from those of early (in 2007:Q1) adopters. Song (2008), Henry (2009), and Guthrie, Irving, and Sokolowsky (2009) find that early adopters’ elections exploited SFAS No. 159’s transition guidance to manage their accounting numbers. These studies provide essentially no evidence that either early or regular adopters complied with the standard’s intent that FVO elections remedy accounting mismatches for economically offsetting positions. In contrast, we hypothesize and provide evidence that regular adopters complied with that intent, having learned from guidance the SEC and others provided about that intent and from the scrutiny early adopters’ FVO elections received. Specifically, we predict and find that variables related to accounting mismatches explain regular adopters’ FVO elections but not early adopters’ elections. We predict and find that variables related to the management of accounting and regulatory capital numbers explain early adopters’ FVO elections but not regular adopters’ elections.

We also examine banks’ initial FVO elections for the three most frequently elected types of financial instruments: AFS securities and debt for early adopters and loans held for sale for regular adopters. We provide four distinct economic and accounting reasons why banks’ FVO initial elections for AFS securities and debt were both amenable to exploitation of SFAS No. 159’s transition guidance and likely to create accounting mismatches, whereas banks’ initial FVO elections for loans held for sale were both not amenable to exploitation of the standard’s transition guidance and likely to remedy accounting mismatches. Based on these reasons, we predict and find that regular adopters’ FVO elections for loans held for sale remedied accounting mismatches and did not exploit SFAS No. 159’s transition guidance. We predict and find the opposite for early adopters’ FVO elections for AFS securities and debt.

Our findings are consistent with regular adopters’ FVO elections, particularly for loans held for sale, complying with SFAS No. 159’s intent. Our findings are broadly consistent with Henry’s (2009) evidence that some early adopters rescinded or revised their FVO elections because of informal mechanisms that arose to help firms interpret and implement SFAS No. 159.

I hope you will join us on Tuesday to hear Stephen’s thoughts, and remember that you can attend Round Table Discussions in Second Life (instructions here) or on the web at our LIVE page.