The following are excerpts from a discussion hosted on March 18, 2009, by Duff & Phelps.The panelists were Warren Hirschhorn, Managing Director in Duff & Phelps' New York Office and global head of the firm's Portfolio Valuation practice, and David Larsen, Managing Director in Duff & Phelps' San Francisco Office and member of the firm's Portfolio Valuation practice.Mr. Larsen serves on FASB's Valuation Resource Group and participated in the SEC's Mark-to-Market Accounting Roundtable in November 2008.
Sing Chan, VicePresident: Welcome. Id like tointroduce Warren Hirschhorn, who heads our Portfolio Valuation practice, andDavid Larsen, who is part of our Portfolio Valuation practice as well. As they sit in front of you, please feel freeto interject with questions at any time. Its going to be a very open forum but I think David and Warren willkick it off with some opening comments.
Hirschhorn: Iwould also like to add my welcome. It is an interesting time to be discussingfair value and mark-to-market issues. Most of you know that yesterday [March 17], FASB released two new proposedFSPs [FASB Staff Proposals] that people are diving into quickly and theres ashort comment period. I believe commentsare due April 1 and FASB is going to put the new regulations in place April 2,so this is moving very, very quickly. One of the reasons we are conducting this forum is that Duff &Phelps is in an interesting place in the mark-to-market world, having apractice that deals very heavily with valuing illiquid securities.
David and Iare going to do this interactively. Ivehad the privilege of working with David on this a number of times, but whatsmore important is that we get your questions as well. Theres an agenda, but we really want tofocus this on the questions that you have.
Larsen: Most ofyou hopefully are aware about Duff & Phelps, who Duff & Phelps is andwhat it is we do, and we can add some additional background on that to theextent you find that helpful. One of theplaces that we feel we can add real value in the market place and the fairvalue debate is as an independent financial advisory and investment bankingfirm. We come at these issues from apragmatic point of view. We understand both the practicalities of dealing withvaluation issues and the very technical aspects of dealing with valuation. We also understand the issues from thebusiness point of view. Many of us grewup in the accounting and auditing environment, so we understand the auditingand regulatory aspects of the issues. Iam a member of FASBs Valuation Resource Group, I worked on the AICPA Net AssetValue Taskforce dealing with private equity aspects, I am on the InternationalPrivate Equity Valuations board, I drafted the PEIIG guidelines which arrived inthe private equity space back in 2003 and I have been involved in the ongoing fairvalue debate. I was a panelist at the SEC [Mark-to-Market Roundtable] back inNovember. As these issues continue tocome forward, I think that one of the things that many people struggle with --and one of the things I struggle with, being very much involved with it -- isthe misinformation that seems to be out there in the press. I think that misinformation comes from acouple of different places. Some of it comesfrom a lack of understanding, some from different agendas, and some from justthe complexity of what were talking about. So I think its helpful to go back in time a little bit and see how wegot to where we are today. As Warren said, things arechanging in real time and changing rapidly.
In less then a week, from a point whereI thought things had died down on the question of fair value, they have now heatedup to a level that is probably as hot as at any time since fair value enteredthe news. The heat was felt at the HouseFinancial Services Subcommittee hearings last Thursday [March 12], FASBs boardmeeting on Monday [March 16], and FASBs issuance yesterday [March 17] of twoproposed FASB Staff Positions (FSPs) that deal with questions on fair value,distressed markets and active markets and on how bank assets should bevalued. Well come back to that in amoment.
If we go back in time a little bit,one of the things that can be slightly frustrating to those of us who are closeto the issue is to hear people -- and sometimes some very smart people -- mischaracterizesome of the facts. One key fact is thatFAS 157, which is much maligned, did not change or add any asset that is orshould be recorded at fair value. It did not change which assets are requiredto be recorded at fair value. FAS 157 became effective in 2007, although somebanks started applying it in 2006, but it did not change which assets arerequired to be recorded at fair value. Whatit did was change the disclosure of how you arrived at fair value andharmonized the definition of fair value. I think most people would conclude that disclosure and addedtransparency are good things.
FAS 157 provides a commondefinition of fair value where, in the accounting literature in the past, fairvalue was used in 60 different places and without a uniform definition. Certain bank and private equity assets havebeen required to be recorded at fair value for years, if not decades. You can go back to as early as the 1940Investment Company Act that required assets to be recorded at fair value. So, fair value is not a new concept and it wasnot implemented by FAS 157; FAS 157 harmonized the definition and added additionaldisclosure requirements. FAS 157,because of the added disclosure and the harmonization of the definition beingimplemented together in a unique time in the economic history of the world, hascreated a number of challenges and in many ways became a lightning rod for someof the problems and some of the questions surrounding the current economiccrisis.
Accounting standards dont talkback, and so, rather than management saying I made a bad loan, its easier tosay that the accounting standards dont let me tell you whats really happening. There are obviously some personal dynamicsgoing on around the debate. As a result,when FAS 157 came out and started being implemented, and a historically liquid marketbecome illiquid, the move from liquid to illiquid caused some additionaldynamics. FAS 157 has a disclosureconcept using Levels one, two and three. Many in the press and many of the pundits on TV have somehow associated Levelsone, two and three with good, okay and bad, whereas Levels one, two and threeare intended to classify the quality of the input into the valuationdetermination; they do not characterize the quality of the asset itself, theycharacterize the observability of the input.
A Simple example of level one: We lookat the stock market price of IBM, for example, and theres a ticker price everysecond because theres a trade every second. Thats an active market, and we can value an asset as its price timesthe quantity owned. Very simple, verystraightforward. Theres a rule based inFAS 157 that says thats how you value something in an active market, pricetimes quantity. (P times Q)
Level two inputs raise the questionof dealing with an asset that either has observable prices, that wecan see inthe marketplace, or is very much like another asset that actively trades in themarket place, and we can use those inputs to value an asset. Many of the loans,historically, that did trade in the market, but that had broker quotes, , wereso-called level two inputs and thats how they were valued in that marketplace.
If you look at the private equityworld, almost all of its assets were level three because they ae illiquid. The inputs that go into valuing them areunobservable. That doesnt mean thatprivate equity invests in bad companies -- it just means that private equity investsin companies whose fair value is determed using less observable inputs. Thatswhere weve arrived, I think, at a mistake in the debate around fair value thatsomehow has caused some people to tend to believe all of Level three isbad. No, Level three can be every bit asgood a quality of an asset. It just doesnot trade as regularly [as others], or the inputs to determine its value arenot as readily available.
So,we moved forward into a phase where the assets became less liquid. You had management, and to some extentauditors, who, because of the way theyve always looked at things, were morefocused on those Level two inputs. And,over time those Level two inputs became somewhat less reliable because therewere not as many transactions taking place. That all came to a head in the fair valuedebates in September and October of 2008. Lehman went down; on September 29, 2008 the Dow dropped over 700 points,then we had a couple of blips back up and since then weve been trailing back downagain.
Well, in those first ten days ofOctober there was major pressure on FASB and on the SECon the fair value question. We had the SEC and FASB joint press release that came out onOctober 1 and then we had perhaps the quickest issuance of financial accountingstandards in the history of the world when FSP 157-3 was issued on October 10after basically a ten day comment period. (The second quickest issuance of standards inthe history of the world will be these two FSPs if they are enacted on April 2. We get at least two weeks or so todeal with them which well come to in just a moment.) FSP 157-3 re-highlights a couple of keyfactors: one, all available information should be taken into account in comingto a fair value assessment; and, two, if broker quotes are not based on realtransaction data they should not be used. It tries to address what I call the stickiness of observabletransaction pricing and saying that you ought to be able to move away from thatand instead take into account other inputs such as cash flow.
As we move forward to today, thestickiness of Level two has remained. AsI believe is outlined in the proposed FSPs, it has remained for a coupledifferent reasons. Bob Herz, thechairman of FASB, testified last week at the House hearing and he raised acouple of those issues. He mentioned thatsome banks or some financial institutions may not have the internal staff orinternal abilities to move from using broker quotes to value assets to using cashflow models to determine the value. They may not have the internal systems tobe able to do that; they may have been unwilling to hire third party expertsthat could help validate models and focus on those issues. You also have, letssay, management and boards that again somehow feel that looking at observableinput is somehow better and not as risky.
Question: Can youdefine what you mean by stickiness there?
Larsen: Let me use an example which Ill come backto a couple of times in the discussion. Lets say we have a bond with a par value of 100 and we look at the underlyingcash flows in todays environment;because of the dislocation of the market and becauseof the recession, we conclude that the actual cash flows will really be 90. Ignore the time value [of money] for themoment. Lets say we now furtherdiscount those cash flows with an appropriate risk-adjusted rate of return -- wetake risk and time value of money into account -- and now we come to a value of80. Now, lets say that somebody elsehas a similar bond or the same bond and for their own purposes they sold thebond at 20. Well, we can see the 20 -- thatsan observable transaction. We looked onBloomberg and saw that the bond sold for 20. Management and to some extentauditors, as Bob Herz said, look at that 20 and thats a Level two input that somehownow should take precedence -- just push everything to the 20. FAS 157 doesnt say to do that; FSP 157-3 saysdont do that. But the practice is, inpart, that because these assets used to be Level two theres still thatpresumption that we should be using Level two inputs. Now, if we had an active market and all ofthese assets were being traded at 20, 20 would be the right fair value number. But if we have one transaction or even fivetransactions at 20, but only five bonds were sold there and we hold 100,000bonds, then that 20 may not be the best indication of fair value. Thats where we get the debate aboutmark-to-market and fire sale pricing, with some people arguing that FAS 157 andfair value requires fire sale pricing. Well,that is blatantly false. It [FAS 157] doesnot require [fire sale pricing]; in fact, it prevents you from using fire salepricing. If you use fire sale pricing you are not acting in accordance withgenerally accepted accounting principles. Thats just blatantly wrong. Youhave to come up with your best assessment of fair value and the fact is that thereare those who see that 20 and latch on to it. Thats what I term as stickiness.
Question: Auditorshave latched on to it?
Larsen: Managementlatched on to it and, to some extent, auditors have latched on to it aswell. Again, because until mid-2007there was more volume, it made a lot of sense. Now, maybe when the price was up at 80 or 90or even 99 there was a lot more volume. But as things became less liquid the predisposition was to still use thatthe ovservable price. Auditors look atwhat management did. So, managementstarts but then theres always that tension along the way. Now, there are thirdparties that do come in. This is one ofthe pieces of business that Duff & Phelps is in. We help companies validate their assessmentsand then we will stand by management when they discuss it with the auditors, ortheir boards, or third parties.
So, we had the fair value debateinto October. We had part of the initialTARP funding,; and the TARP legislation which mandated the SEC perform a study to answer six questions. The SECissued its report on December 30. Someof those questions revolved around whether FAS 157 should be suspended and if theaccounting standard setting process was appropriate.
Hirschhorn: Justto restate one thing David said, in todays market place people perceive FAS157 as fair value. Thats thestandard. Well, fair value has alwaysbeen in place. People interpreted itdifferently, as they do any piece of documentation. But FAS 157 did not create the whole fairvalue conversation, it [fair value] was always there. People, banks, hedge funds, private equity they all were required under GAAP to mark their assets to fair value prior toFAS 157.
Larsen: Thats another key point that washighlighted by the SEC study. The SECstudy looked at 30 large financial institutions and provided data in the report. If you listened to the House hearings lastweek, you also heard a good deal of data on the community banking area. And the banking area is really where thedebate has become hot and heavy mark-to-market is bad for everybody, eventhough were really talking about banks, the impact on regulatory capital andthe impact on bank assets.
For the 30 banks in the SEC study, which are 30 of the largest nationalinstitutions, only 45% of their assets were recorded on a fair value basis. 55percent were recorded at amortized cost. So when people start pontificating that FAS 157 or mark to market orfair value requires us to value our assets at a fire sale price, [they shouldrecognize that], first and foremost, the accounting is driven by the individualasset and by the entity. The bankaccounting is very specialized but it really puts assets into three differentbuckets. For the 30 largest banks, 55percent of their assets were at cost. Well come back to that in a minute.
In the community banking area,where there has been a lot of pressure on congressmen from the local communitybanks, 99 percent of their assets are at cost. But -- and theres always a caveat in all of this -- the banking accountingrules say that if you have a loan held at cost that is other than temporarilyimpaired (OTTI) -- if you hit the OTTI rules (which are under discussion to bemodified slightly), which basically are that you cannot hold the loan tomaturity or you do not have the intent to hold it to maturity and you wont getfully repaid -- if you hit that trigger, then you have to fair value the loan. That happens once, where you fair value theloan one time at that point in time when you hit that other than temporaryimpairment trigger. You can then go onand if it gets impaired again, then you again fair value it once. Well, that fair value assessment is done usingthe concepts and principles of FAS 157. So, if you have sticky Level two pricing, then some people say oh, thatmeans I have to go down to the 20 price (in my previous example) even wherethe actual cash flows say its really worth 90 and the discounted cash flowssay its worth 80. Thats where peopleare saying well, FAS 157 mandates that I have to write it down to 20. FAS 157 does not mandate that. FAS 157 gives you the guidance to use yourjudgment. People have applied FAS 157and have selected the 20 in their application of FAS 157. That practice, then, blew up again with theHouse Financial Services Committee last Thursday.
Question: In yourexperience, how widespread is the problem of people using the 20 rather thanthe 80? Is this something you see a lot,where people say we have to do this?
Larsen: It isanecdotal at one level. I sit on FASBsValuation Resource Group. I have hadmore than one of the large accounting firms tell me that this is going on. It is a judgment question and it is a realworld question. The question that isbeing answeredwhen we say were recording at fair value is, and the definitionof fair value is, what is the price that I could sell my asset for in anorderly transaction today? If you thinkabout that and I see a transaction that happened at 20, theres a lot of logicto say, hey, thats what the market is telling me the market is doing -- thatswhere its trading. Youve got to havesome evidence to move away from that.
Im not throwing any stones atmanagement or how they apply the standards; Im saying that it isunderstandable why Level two is sticky. How widespread it is, is difficult to tell. We will see that if FSP 157-e, the proposedstandard, comes through. I think it will be applicable for [2009] Q1 and one ofits provisions is to identify the impact it had. Now, its really hard to identify that impact because markets aremoving dynamically. If wetake one singleasset recorded at 20 on December 31 and now we get away from the stickiness oflevel two pricing and weve determined that the fair value is 30 on March 31 how much of that 10 point uplift is because of moving away from sticky Leveltwo and moving to looking at cash flows, and how much of it is because of changesin the market? There are a lot ofdifferent things that make the number a little bit fuzzy.
Hirschhorn: Wevealso seen, because of what we do within the space, that each quarter there arefairly robust conversations with funds and auditors on this because as astarting point people go to the sticky price. There will be a cash flow analysis, a DCF analysis done of the underlyingcredit, and there may be a difference in price. 20 may be the right price, but the right price may be the 80 that Daviddiscussed. We will have discussions withthe auditor primarily around that. At the end of the day, the decision has tobe made on booking of the fair value. Sowe get involved in those conversations; they have been fairly routine.
Larsen: Andanother anecdote. Im not saying itswidespread, but it demonstrates the issue. Its a real live situation where we had a discussion where a client or acompany bought 100,000 bonds in a private transaction at the price of 85. They did that, lets say, in October. Then in mid-November, theyre watching theirBloomberg screen and all of a sudden 6 bonds pop up at a price of 30. They say, Ive got 100,000 already, now Imgoing to buy the six, they buy the six for 30. They get to the end of the year and now theyre valuing the 106,000bonds that they own and the auditor says, well, clearly the price is 30,thats the last trade. The client is saying, but the underlying economics supportthe 85. The auditor says, No, you haveto use the last transaction price. Andthen they say, Well, I just bought another bond in January for 85; it happenedto be a one-off, I wanted to add to my position, so I bought another one at 85in January. And the auditor said, thenAt the end of January, the price was 85, but at the end of December it is 30. That thought process, I would argue, isnot compliant with FAS 157; its a misapplication of FAS 157 by the auditor. Its a major firm auditor but its not an auditorin one of the major cities. Thats asingle anecdote that shows the stickiness and the mindset of level two.
All of these questions about OTTIand the stickiness of level two pricing have been on the table for some time-- Idiscussed them at the SECRoundtable back in November, theyre part of the SECstudy that came out on December 30, we discussed them at the Valuation ResourceGroup in February -- so they have been under discussion and analysis. I thinkCongress felt like things were not moving quickly enough, and so thats why theyveput the pressure on to say act in the next three weeks. And so, FASBs board met on Monday and cameout with the two FSPs late last night.
In a nutshell, the first FSP, theproposed FSP/FAS 157-e (that is how it will be defined until such time as it isissued in final form), gives the background in a maybe more succinct fashionthan I just did. It is actually titled DeterminingWhether a Market is Not Active and a Transaction is Not Distressed. Maybe coming at things from the negative is uniqueto accounting, but rather then defining active or defining distressed theyredefining not distressed and not active. Whether its Boolean logic or whatever I dont know, but thats the waytheyve chosen to do it; some of the comment letters that come out in the nexttwo weeks I am sure will address that and say, well, why dont you just comeout and tell us what it is, as opposed to telling us what its not.
FASB in the proposed FSP identifiesatwo step process. Again, the reason theyredoing this;I believe in looking at it and understanding the issue; is FASB seesthe human behavior application problems with the stickiness of level two, Ithink their solution is relatively elegant -- to the extent that you can usethe term elegant in the context of an accounting standard -- because itidentifies a two step process.
They identify seven differentfactors that are kind of the first trigger. If those seven factors or portions of those seven factors exist, thenyou can conclude that a market is not active.
Those seven factors include: • few recent transactions based on volume; • price quotations are not current; • price quotations vary substantially over time oramong market makers; • indices that used to be correlated are no longercorrelated; • there are abnormal or significant increases inliquidity premiums; • there are abnormally wide bid-asked spreads; • there is little public information.
Again, this is a proposed standard andI expect it to be tweaked, if not significantly changed. I know we will issue a comment letter to saywhere we think it needs to be tweaked a little, but in general I think it givesa fairly good outline. So, if you havethose factors -- coming back to my 20 bond example, theres been no trade orthereve been very few trades or there was a big private trade that wasntpublic in October but there were a just a couple of trades in November -- thatwould logically qualify fromstep one as not active.
Now, you also ask the question wasthere sufficient time to allow for usual and customary marketing of that asset? Say it usually takes a week to sell abond -- Im just contriving that example because maybe it really takes twoseconds to sell a bond but lets say it usually takes a week -- if that 20price came up after a week and it normally takes a week, then I take that intoaccount. And, if there were multiplebidders that came up with that price of 20, then I take that into account. If there were not multiple bidders or if therewasnt sufficient time so, in other words, I only had a minute to sell thatbond rather than my normal week -- then that is a distressed price and itsthrown off the table.
Hirschhorn: David said if it meets the test you takeit into consideration. That doesnt meanyou take it. You still apply judgment.
Larsen: So thats where its kind of a two stepprocess. Step one is identifying if youhave those factors showing that the market is not active. If, in addition, theres not enough time tosell an asset and theres only a single bidder, you throw it off the table. That is a fairly major way to say hey,industry -- dont be stuck on that level two pricing. Its fairly dramatic to say knock it off thetable.
In most cases there is sufficienttime, even in this distressed market, and there are enough bottom feeders outthere that there are multiple bids. Inthat case, the FSP then goes on to say, ok, we use the observable price as aninput but we dont use it as the exclusive input. And whether we use it as an input or throw itoff the table, we still then augment the observable price by the use ofsomething like an income approach or a discounted cash flow. So coming back to my example, where my risk-adjustedcash flows came to 80 or 75, I would say, alright, my cash flow data using anappropriate discount rate comes to 80. Ivegot this transaction, there was enough time to see it through, and there weremultiple bidders at 20. Im going to weightthose, the weighting is going to be judgment- based and theres going to haveto be a basis for it but its absolutely reasonable to come up with a fairvalue estimate of 65. The weightingwould be driven by how much volume is really going on in the market for theasset;, how many bids were there, how often did those happen?
That new guidance, if its passedsubstantially in the form it was presented last night, will be effective for Q1(2009) as its currently drafted. So, thatwould say that anybody who, because of managements own choice to stick withthe sticky level two pricing or because of pressure from the auditor to stickwith the sticky two pricing, would say hey, Ive really got to rethink using leveltwo pricing, and I need to augment that by the use of discounted cash flows. Now, discounted cash flows are not a panacea;discounted cash flows have to be thoroughly vetted, they must be robust and havegood analysis behind them. In many cases,firms will conclude that they need to use a third party expert to assist themin that process.
The other big change is the OTTIrules as proposed by FASB last night. Itsan adjustment to three different accounting rules. That FSP deals with FASB Statements 115 and124, and EITF 99-20. The fact that itsmodifying an EITF from 1999 says that these rules relate all the way back to99 and earlier. So again, its not a Statement157- vintage 2006/2007 issue; the bank accounting has been in existence foryears. What the change does, in a nutshell, is this: it says that if youve gotan asset that you plan to sell, you determine the fair value as I have describedcoming up with the 65 in my example and thats the amount that is recorded; anyadjustments in fair value flow 100 percent through your income statement.
If you planto hold the asset to maturity -- if you have the ability to hold it to maturityand you plan to hold it to maturity, i.e., you do not plan to sell it -- thenyou split that fair value assessment. Again,weve determined that fair value is 65 -- but we split that 35 reductionbetween the 10 of actual cash flow loss which is run through the Profit andLoss statement (P&L), and we take the 25 of extra fair value adjustment andit is run through other comprehensive income. It still impacts total equity but the other comprehensive incomeadjustment, at least under the current regulatory environment, doesnt hitregulatory capital like the P&L impact does. So, and thats a point that I didnt make atthe beginning, the whole regulatory capital question -- the regulators donthave to match things to GAAP; they have chosen to, but they dont have to. And so, when they say fair value accountingis causing problems with regulatory capital, fair value accounting is reportingthe best assessment of fair value. Theregulators are choosing to use that to tell a bank what they can and cant lend. But the regulators could change their rules.
Let me stop there and see if thereare questions or where you have questions.
Question: Why doyou think this is elegant?
Larsen: I thinkits elegant because it uses factors that allow judgment to continue to beexercised. And it doesnt come at itwith an iron fist rule that would arguably be arbitrary. As I was thinking of it, I was almost thinkingwell. maybe the only answer is to mandate the use of DCF, but to mandateanything where judgment is required is a little bit scary. So Im saying that the proposed FSP staystrue to the principles outlined in FAS 157 and it really gives some factorsthat need to be considered. Itencourages people to do what they should have been doing in the first place,using judgment.
Hirschhorn: FAS 157is a principles-based document. Whathappened, I believe, is that FASB tried to give people an idea of how theywanted it to be applied. However, It got pulled into a rules-based culture. So, by giving seven criteria or factors youtry to keep to the principles. It doesnt say you have to meet all seven; youfollow these and it says after evaluating all factors and considering thesignificance and relevance of each, the reporting entity shall use its judgmentin determining. So its saying thatyouve got these (factors) -- look at them and come up with a decision. It doesnt say, if A happens this is whatyou have to do. And so I think itsstill principle-based.
Question: This isfascinating because it seems that the public at large and Congress have gottenthe opposite message. This seems to bewhat youre saying, that with this guidance FASB seems to be saying were nottelling you to use these fire sale prices, quite the opposite. Is that true?
Larsen: Thats where public opinion, those whohave written about it in a newspaper, some of the big pundits that I repeatedlysee on TV, absolutely have the rules wrong. Accounting standards do not allowthe use of fire sale pricing in determining fair value. They do not allow it. It is peoples exercise of judgment that resultsin using that 20, for our previous example. The pundits say that 20 is a fire sale price. I dont know if it always is or it isnt. If its the best assessment of fair value itshould be used in a fair value estimate. If its a fire sale price it should not be used. The fact that human behavior allowed lasttransaction pricint to be used without fully exercising the principles and thejudgment, thats the problem -- its the application of the standard, its notnecessarily the standard itself.
Question: What then about the arguments from thebanks, and is there a distinction here without a difference? The banks arguelike clockwork that this is pro-cyclical, that FAS 157 is pro-cyclical. From what youre saying, this hogwash on itsface. But in fact thats the effect ofit, so whats the answer to that?
Larsen: Taking our simple example; I think often whenpeople talk about fair value, they talk about the sound bite that says oh,mark-to-market require fire sale pricing, oh, the actual cash loss is adollar and youve forced me to write it down by 90 percent. Again, those arent the rules. Where the banks get caught up, I think, are acouple of different places. Question one is, is the asset recorded at fairvalue or not? Weve already determinedthat most of them are not. When theresa real impairment, then we have to move to the fair value model. First and foremost, they have given a loanthat is impaired. Thats the wholepremise we are not even in the fair value boat unless theres been an impairment. The new rules allow thatimpairment charge to be split, part of it going through the P&L and part ofit through comprehensive income.
But, the other question is, what isthe fair value in an illiquid market? That requires the exercise of judgment. The fact that management has been using 20 raises the big question of is that where management really believes the fair value to be? That if they actually redid this loan, resoldthis loan, sold this loan today, that they would get 20, even at a hypotheticalprice? If thats what it is, the factthey loaned money out at 100 and are only going to get 20 back because this isa trading asset, we should be reporting it at 20. And for anybody to say, well, if we nowstart reporting it back at 100 that somehow makes the world a better place,when deep down management knows its really going to be 20, that does not helpthe economy or the regulators or anybody at all. However, if we have gone too far and weveallowed the pendulum to push us too far to that sticky pricing and when marketsare illiquid we should be able to use the underlying cash flows and reasonableadjustments and maybe fair value is really 40 and that means some of the capitalthat was deemed lost, may not have been eroded -- again, judgment goes into determiningfair value. But its been easy toeffectively blame the problems of bad loans on an accounting standard thatcant talk back.
Question: How much support do you guys have forsomething like separating the regulatory capital from the GAAP accounting?
Larsen: We dont get involved on the regulatorycapital side too much. I just make theobservation, and this was reiterated by the Deputy Controller of the Currencyin the hearings last week, that they dont have to follow the GAAP numbers,they choose to follow them. Theregulator is looking at risk, and there is an argument that in good times youought to scale things back from the regulator point of view and in bad timesyou may need to loosen them a little bit. Again, regulators have the ability to do that. Im not a regulatory expert to be able todetermine what all that means and how you impact prudential regulation and implementthe Baselregulations. I just make the observationthat theres nothing that says regulators have to be in lockstep with GAAP.
Question: Thatwould be a natural response, for FASB to be saying to the SEC at this point, hey, look you guys have usright in the middle here and when are you going to get us out? You should be the one who in some sensedecides how the whole thing plays out.
Larsen: Which is to say, if youve got two kidson the playground and one of them is getting beat up by the bully, the otherone doesnt necessarily say come hit me, too. They run for cover.
Hirschhorn: The split we just spoke about just cameout in the proposed regulations yesterday, so Im sure theres a lot going ontoday and the next couple of days within the banks to see how it affectsregulatory capital.
Larsen: What I haventstudied enough yet because it just came out late last night, is if you justapply it prospectively or if you do anything with that retrospectively? Because again, if you say that from nowforward I split my fair value assessments because of OTTI that happened alongthe way -- what do I do with historical fair value adjustments? Those are the ones where youve got all thosepeople pounding on the table saying we were forced to write this down to a firesale price. Again, no one was ever forced to write it down to a fire sale price. People wrote it down to their best assessmentin the collective judgment of management and the auditor of what fair value was,but that all flowed through the P&L. Now, were splitting it between other comprehensive income and the P&L,but I dont know yet what happens to the historical piece.
Again, Duff & Phelps often getscalled in to augment what management has done, to independently vet it and westand there as an additional discussion point when the auditors come in. That can help support a very healthy informeddiscussion because these are our judgmental assessments. People can look at things differently but byusing an appropriate analysis, using all the available information, when itcomes to a reasonable estimate of fair value, then that fair value estimate iswhat would be receive if the asset were sold in an orderly transaction in thecurrent market?
One of the key interesting points includedin the proposed FSP as a final sentence, says that an orderly transactionwould reflect all risks inherent in the asset, including a reasonable riskpremium -- so again that takes it down from my 90 to the 80 or the 75 -- forbearing uncertainty that would be considered by willing buyers and willingsellers in pricing the asset in a nondistressed transaction at the measurementdate. So, it effectively restates theway most of us read FAS 157 when it came out in the first place, but the humanbehavior part and the dislocation of the markets caused kind of a perfectstorm that allowed Level two to become more sticky then it was ever intendedto be.
Question: Youmentioned that youre going to send a comment letter about the rules and thingslike that. Just from your first glance,is there anything in particular that sticks out to you that might need to betweaked? Or anything in particular thatyoure planning to recommend to be tweaked?
Larsen: Its verypreliminary to determine what we will say. I think that, at first blush, [weneed to] make sure that none of the words can be misinterpreted, or identifywords which could be misinterpreted, and [point out any possible] unintendedconsequences. Heres an example of anunintended consequence that well have to identify and address; it will takemore study, but Ill say this as a potential one: in the private equity arena,valuing limited partnership interests is a question that is on the table.Thats something that the AICPA [has looked at]; I was a member of a task forcethat came out with a draft issues paper in January and it was discussed atFASBs VRG meeting in February, and FASB said they would put it on their agenda. I think its gotten to be a back burnerbecause of these [other] issues. The wayI would interpret FSP 157-e is that the secondary market would fall into thoseseven characteristics and therefore secondary pricing would be thrown out. On the surface that really helps the LPcommunity. But if you dont have anytype of price information you need to perform a discounted cash flow anlaysis. So if you then say, ok -- rather then using NAV, you limited partners now need to come up withyour own models of all the cash flows for all of your private equity investments,that would be a major operational change at the LP level. Im not saying it says that, but some peoplecould interpret it that way. So thatwould be an example of a potential comment we would make; to say, we likethis, this word needs to be looked at but this is the potential impact andmaybe this needs to be addressed or carved out in some way, and/or the LPguidance needs to come out at the same time because some people could interpretit to say, ok, LPs, youve got to deal with a discounted cash flow model forall of your 400 limited partnership interests. And most LPs wouldnt have the ability to do that.
Hirschhorn: TheLPs wouldnt have the ability to do that because while the LPs may have theknow-how to value companies, they dont necessarily have the access to go intoand review every underlying investment;they dont have the information. Limitedpartners get limited information. Theyget the information thats called for under the PPM, the offering document, butthey do not get detailed financial statements of the underlying companies, theyusually get a synopsis. So, one of theissues that David is working on, with the AICPA and the LPs, is that the auditorsare calling for a robust review of the GP reports and I think robust is theterm. So, within that, what defines robust?If the tide turned toward what David just spoke about, in my opinion it wouldbe virtually impossible for the LPs to do that. They dont have the information to do that kind of work.
Larsen: So thatsan area where Duff & Phelps gets involved in the industry debate. Our clients ask us to pragmatically assistthem in implementing these rules, understanding them, articulating them andcoming to appropriate fair value conclusions. We work with corporate clients on valuing those types of assets orvalidating the assessments that they have made, were in the hedge fund space,were in the private equity space, were in the BDC space, were in the limitedpartner space, helping with policies, procedures and focusing on their needs inunderstanding, executing, implementing and developing processes and thenvalidating any estimates.
Hirschhorn: Actually,we were talking the other day about why people call Duff & Phelps to helpthem out. The reason, we believe, isbecause we have strong GP relationships, strong LP relationships, strongaccounting relationships and the accounting regulatory knowledge. Were a group that has a lot of different experiencewith the constituencies that people are looking for. We dont have boundaries regarding providingadvice. We can provide advice; we cutacross all of the Big 4, for instance. On the best practices basis, we see the best practices across the widespectrum and we provide and give clients advice as to best practices. That is a differentiation point, because ofthe independence that David spoke about that we provide.
Larsen: And weve got the technical strength.Weve got PhDs, weve got people that understand each of these different, verycomplex assets. When you get into a CDO, a CLO, all of them are looking at the cashflow streams. On one level one can say,its easy to look at the cash flows and put in a discount rate, I can do thatin Excel. No, these are complex cashflows that have different parameters that must be applied to them that reallytake some of the PhD type talent to understand them, and thats what we bringto the table. We understand the audit side, we understand the investor side, weunderstand the management side, we understand the technical valuation side. So, we bring all of that to the table inassisting clients. And again, we have aseat at the table at FASB, advising FASB through the VRG, and we makeappropriate comments along the way to make sure that the debate is wellunderstood, not just from an auditors perspective or from a companys perspectivebut that holistic perspective because were in enough of the differentsituations to be able to see various perspectives and we want to have solutionsthat make sense. FASB is coming up with the principles, but how are theseprinciples executable on a pragmatic basis?
Question: I imaginethis is going to make this whole thing go away.
Larsen: In my mind, that this [the proposed FSPs] substantiallyfixes the two major problems that when you cut down to the actual issues thatare there. It substantially if not fullyfixes those two major problems, which are: What do I do on the OTTI side, andhow does that impact things? And, how do I deal with stickiness of Level twopricing? I distill it down to those twoissues. And the proposed FSPs substantially fixes them. To translate that intothe understanding for the House subcommittee and the public at large is aHerculean task, if not an undoable task. In my mind, it absolutely fixes the perceived problem, but because theseare technical matters, because they are principle-based, because they areaccounting rules, not very many people really comprehend what the impact is.
Lets take last night. Iheard just a snippet of Bill Isaac; I didnt see his full interview. Hes one of those who somehow believes thatfire sale pricing is required -- which its not, but he says that. The little bit I caught of him, it soundedlike he said, well, this fixes things. He has said more than once that real credit losses should be taken intoaccount. From the banking side, realcredit losses are still flowing through the P&L, and the other piece willbe recorded down as other comprehensive income, which ostensibly doesnt hitregulatory capital, so it seems to solve the banks problem. It solves the problem from the perspective of,were not telling FASB what to do and notthrowing the baby out of the bath water. Weve got an independent standard setter, were getting appropriatepublic comments and were dealing with that, and then were dealing with the behavioralissue of the application of the standard.
But it doesnt change those who saymark-to-market accounting is the evil villain in the economy. It wont change that, because again fairvalue accounting is embedded in multiple different uses, appropriately so, andso it wont change that piece. So, howdo you translate that, because there is in some ways mass hysteria againstmark-to-market. Some of you can have animpact on that. Others can have animpact on that. Were clearly open to talkand well continue to talk about it to identify what the real issues are. Its just now a question of do people careabout the facts.
Hirschhorn: Its [a matter of] getting the rightmessage out. Theres a lack ofunderstanding. Are people calling forthe end of fair value and the end of FAS 157 and that makes the world good again? Thats not the right answer, but so muchis in the message that needs to be delivered. I was actually listening thismorning on the TV and they didnt talk about fair value at all. I thought that maybe since it came out lastnight they could have rounded somebody up for this morning, but I expect that overthe next couple of days it will be interesting to hear how this is spokenabout. The fact this happened makes it incrediblyrelevant; its always been relevant but this just brings it to the floor.
Larsen: For thosewho somehow believe that eliminating fair value accounting will some how help--if we eliminate FAS 157, all we do is remove disclosure and have lesstransparency, and we go back from an exit market concept of fair value beingdetermined base on what would I receive if I sold this asset in an orderlytransaction today to a definition of willing buyer/willing seller which, goingfull circle, is what is now embedded/re-embedded in 157-e all killing FAS 157 woulddo is reduce transparency. Stopping fairvalue accounting, in my mind, is tantamount to taking freshness dates off ofthe milk bottle in the grocery store and somehow thats going to allow thegrocery store to sell more milk. If Im a buyer, Ive got to find out how freshthat milk is, otherwise Im not going to buy it, Ill look for something else.
Published March 31, 2009.