“late late night after the show at leeds university…”
Most of what’s been written about the financial crisis and the firms that were forcibly acquired, failed, or bailed out tends to focus on “fair value” as the feckless culprit.
“MtM [mark-to-market] accounting itself is flawed… There are difficulties in establishing real values of many instruments. It creates volatility in earnings attributable to inefficiencies in markets rather than real changes in financial position…Valuation for all but the simplest instruments today requires a higher degree in a quantitative discipline, a super computer and a vivid imagination. For complex structured securities and exotic derivatives, the only available price is from the bank that originally sold the security to the investor. Prices available from the purveyor of the instrument (a concept known as mark-to-myself) strain reasonable concepts of independence and objectivity…In the global financial crisis, with the capital markets virtually frozen, the extent of losses on bank inventories of hard-to-value products and commitments (structured debt and leveraged loans) was difficult to establish.”
We know that the banks’ “independent” external auditors had a hard time establishing both fair values and the “extent of losses on bank inventories of hard-to-value products and commitments.” We know this because their clients did not tell us about the extent of the losses until it was too late. There were no “going concern” warnings for any of the financial institutions that went bankrupt, were taken over, or were nationalized via bailout.
We also know that the auditors did a poor and inconsistent job of establishing fair values and forcing disclosure of the “extent of losses” on banks’ investments because their regulator, the PCAOB, told us so.
Inspection teams also observed instances where firms’ procedures to test the fair values of financial instruments, including derivative instruments, loans, and securities, were inadequate. In these instances, deficiencies included (a) the failure to gain an understanding of the methods and assumptions used to develop the fair value measurements of financial instruments that were illiquid or difficult to price, (b) the reliance on issuer-supplied pricing information without obtaining corroboration of that information, and (c) the reliance on confirmation responses from third parties or counterparties that included disclaimers as to their accuracy and appropriateness for use in the preparation of financial statements.
How do the auditors, one step removed and ten steps behind, determine fair values of complex instruments especially in illiquid markets if even the super-bankers couldn’t get it right? This question supposes that it’s the auditors’ obligation to determine the values and that the bankers didn’t get it right.
Neither is true.
What are the auditors’ obligations with regard to clients’ fair value measurements and disclosures? Auditors do not establish fair values. Instead, their role is to, “test management’s fair value measurements and disclosures.” But that obligation is broader than just taking the word of the “masters of the universe.”
The auditor should consider using the work of a specialist if the auditor does not have the necessary skill and knowledge to plan and perform audit procedures related to fair value. Observable market prices may exist to assist in testing fair values. Where they do not and other valuation methods are used, the auditor’s substantive tests of fair value may involve (a) testing the significant assumptions, the valuation model, and the underlying data, (b) developing an independent estimate of fair value for corroborative purposes or, where applicable, (c) reviewing events or transactions occurring after the period covered by the financial statements and before the date of the auditor’s report.
I say it’s outrageous to see ongoing material “disputes” regarding the fair value of complex derivatives between counterparties, especially if they are clients of the same auditor. Critics have suggested that I condone breaches of client confidentiality. Without betraying client confidentiality, they ask, how can distinct audit teams compare the values assigned to either side of same transaction?
One of my commenters explained it:
Just how many PhD’s with CDS valuation expertise do you think PwC has lying around in New York? The valuation of these instruments and the testing of the assumptions would have been sent to a centralized derivative valuation group to review and test. Such a team would have had a fairly standard set of guidelines and testing approach regardless of the team sending it. After validating the inputs, they would have likely put it through their own sausage machine / valuation tool and compared the results. I think there would be a high probability that the same analysts would have been reviewing the same instrument for both GS and AIG. And when they notice that GS is using market derived inputs for the referenced MBS while AIG is using the historical average default rates and ignoring the market you would have hoped they might speak up. And when the partner (finally) heard the rumblings of a problem, even after it has been filtered through the manager / senior manager “make-it-go-away” screen, he would have asked “who else deals with this cr_p in the firm? GS… ah, [insert name of old white guy here] is an old buddy of mine, I’ll just give him a call and ask him what they do…”
When one excuses the auditors for not getting fair value right, there’s a follow-on argument that claims no one got it right. No one could possibly get it right. That’s why the crisis occurred. That’s what the scoundrels that benefited most from the crisis would like you to believe.
Reality is the opposite.
Much has been written about how well Goldman Sachs made out as a result of the crisis. But there are others. Some are getting prosecuted like Bank of America’s Ken Lewis for hiding losses to further their interest in millions of bonus dollars. That’s why some are starting to use the word “fraud” when speaking of Lehman’s collapse.
On February 11th, Bloomberg’s Jonathan Weil asked why no one is prosecuting Lehman Brothers executives for fraud:
It is so widely accepted that Lehman Brothers Holdings Inc.’s balance sheet was bogus that even former Treasury Secretary Hank Paulson can say it in his new memoir. And still, the government hasn’t found anyone who did anything wrong at the failed investment bank…In his new book, “On the Brink,” Paulson doesn’t point fingers at specific Lehman executives for violating any rules. He displays amazing candor, though, in describing how Lehman’s asset values were a gross distortion of the truth. It doesn’t take much imagination to figure out they didn’t get that way all by themselves.”
A reader, I’ll call him David the CFE, repeats a story to me to illustrate this point:
“Casey Stengel probably said it best when he said after the Mets 40-120 season, ‘Gentlemen, not one of you could have done this on your own. This was a team effort.’ “
Losing $156 billion requires a team effort.
When former Lehman Managing Director Arthur Doyle reviewed Larry McDonald’s book on Lehman, he asked the same questions about fraud and Lehman executives:
“The most important questions of all are not even asked in “A Colossal Failure of Common Sense,” or in any other account I have so far seen of the Lehman failure. Simply put, how did Lehman’s published financial statements, as recently as its final 10-Q published in July of 2008, show a positive net worth of $26 billion, when the bankruptcy liquidators are saying that they are looking at a negative net worth of $130 billion? Doesn’t any or all this constitute securities fraud? And shouldn’t there be criminal liability for the executives who signed the firm’s 10-K and 10-Q’s, who under Sarbanes-Oxley are responsible for material misstatements made in those documents?”
Bloomberg’s Weil has a theory about why these crimes are not being prosecuted:
“There’s been much talk the past two years about moral hazard, which is the risk that companies and their investors will behave more recklessly when they believe the government will bail them out. Less has been made of a similar hazard: The danger that powerful companies won’t follow the law when their executives believe the government won’t hold them to it…The latter risk threatens not only our economy, but our democracy. There’s every reason to believe both kinds are growing.”
David the CFE and I have another theory:
The crimes are too numerous to prosecute without indicting the whole system and most of the major players. And because they were part of the problem before they were theoretically part of the solution, culpability also attaches to Paulson and Tim Geithner.
David the CFE’s theory is premised on some of the oldest tricks in the book for manipulating revenue recognition and, therefore, reported profits and incentive compensation payouts including stock options – roundtrips, parking, and channel stuffing. In another variation on the theme, global trading company Refco used a round trip loan to repeatedly hide a related-party transaction incurred to delay disclosure of significant uncollectible accounts. It’s not like these techniques haven’t been used before (by AIG, for example) to offload risk and smooth earnings at quarter- and year-end.
“This case shows that the Commission will pursue insurance companies and other financial institutions that market or sell so-called financial products that are, in reality, just vehicles to commit financial fraud,” said Stephen M. Cutler, director of the SEC’s Division of Enforcement.
With regard to the financial crisis, these revenue recognition fraud techniques may have been most useful in establishing “observability” of market prices for otherwise illiquid assets. Establishing “market prices” via fraudulent, sham transactions amongst the market participants before quarter-end and year-end reporting periods would have allowed assets to remain on the books longer at inflated values and, therefore, to inflate profits and bonuses. “Market prices” that appeared to support existing valuations sustained the myth. The investments were not written down until long after the market for subprime real estate securities started to wilt.
David the CFE explains this theory in the case of Lehman Brothers:
Nassim Taleb says about banks: “Banks hire dull people and train them to be even duller. If they look conservative, it’s only because their loans go bust on rare, very rare occasions. But bankers are not conservative at all. They are just phenomenally skilled at self-deception by burying the possibility of a large, devastating loss under the rug. Taleb further states: “Executives will game the system by showing good performance so they can get their yearly bonus.”
Lehman paid out $5.2 billion in bonuses in 2006 and $5.7 billion in bonuses in 2007. Did this result from the executives at the bank gaming the system to increase their bonuses? An example of burying a large loss under the rug can be found in this excerpt from Lehman Brothers in its 2006 10-K:
We held approximately $2.0 billion and $0.7 billion of non-investment grade retained interests at November 30, 2006 and 2005, respectively. Because these interests primarily represent the junior interests in securitizations for which there are not active trading markets, estimates generally are required in determining fair value. We value these instruments using prudent estimates of expected cash flows and consider the valuation of similar transactions in the market.
Junior interests in securitizations. Lehman and other firms purchased mortgages that would effectively be resold by them as collateralized debt obligations. Each of Lehman’s securitizations was broken into tranches in which senior interests received greater preference with respect to collections of interest and principal than junior interests that were entitled to greater profits, if such profits were realized. A junior interest in a securitization is the lowest level of the tranches for collateralized debt obligations. Generally, only the bottom 3% of a securitization was labeled as equity.
During 2006, housing prices dropped nationally by at least 5% from the spring of 2006 to Lehman’s Nov. 30, 2006 and the default rate was increasing as well. With prices of houses dropping and the default rate increasing, there was a risk of large losses when the buyer defaults. Thus, the junior interests in securitizations that Lehman was purportedly investing in were probably already worthless at the time that Lehman invested in them or at November 30, 2006.
An auditor would have to suspect a material loss is being hidden and that collusion between several departments at Lehman Brothers and management’s participation in the deception was possible. Ernst and Young, Lehman’s auditors, were probably unwilling to consider such a possibility because auditors accept as dogma that collusion between many employees and multiple departments is unlikely no matter what the motive, i.e., $5.2 billion in bonuses. Auditing standards also do not consider collusion likely. Apparently, auditors did not consider the possibility that two different groups at Lehman Brothers such as the underwriters who sold the securitization IPOs and the trading departments would collude to hide a $1.3 billion loss in a junior equity position that could not be sold.
Hiding losses on CDOs and mortgages purchased for securitization. A reasonable question to ask was: If Lehman Brothers started the fiscal year ending Nov. 2007 with $57 billion of CDOs and held them for the year, what would their estimated loss be? Also: What would the additional loss be with $32 billion in CDOs and/or mortgages purchased?
Presumably, the losses would be in the range of $10 billion to $30 billion. By Nov. 2007, everyone knew of the problems with CDOs. Bear Stearns had already closed two hedge funds investing in CDOs. Merrill Lynch had made huge write downs and forced out its CEO. My guess is that Lehman Brothers engaged in schemes to fool the auditor in order to avoid disclosing losses from their securitizations and investments in CDOs.
Lehman probably pulled a variation of the old “telecom swap.” In the “telecom swap” cases, one telecom company would sell telecom capacity to another telecom and then purchase the same amount of telecom capacity from the other party. The firm selling the capacity would book the amount received as revenue and the firm purchasing the capacity would book the amount received as a fixed asset. It worked very well in creating fictitious profits for those firms.
That same trick could be used by financial institutions in the case of CDOs/CDSs. Let’s say Financial Institution A sells collateralized debt obligations with a true fair market value of 90 million to Financial Institution B for 100 million dollars in cash. Financial Institution B purchases collateralized debt obligations with a true fair market value of 90 million dollars from Financial Institution A for 100 million dollars in cash.
And then those phony trades are shown as the “observable” similar transactions in the market.
Did the auditors check for this item? Probably not. Why not? Because it’s an example of collusion between Lehman and other companies. Auditors don’t check for collusion no matter how many times they get fooled by it!
Incentivizing fraud. Auditors, especially inexperienced ones, think management has to actually tell someone if they want to overstate their income. Auditors and the judges that try these cases want to find “smoking gun” memos and emails that say, “Overstate income so we can all get our bonuses and keep our jobs.” But all top management really has to do is tell each unit head that: (a) you and your employees will get large bonuses if your unit reaches its profit goals and, (b) you will not receive a bonus if your unit doesn’t achieve its goals. Then management promotes only those who meet those goals – regardless of how they meet them.
In other words, each manager within the Lehman brokerage unit had a major incentive to reach his profit goals. And, each employee who worked for those managers also has that incentive because his bonus and promotions are based on meeting those goals, too. Thus, management doesn’t have to direct its employees directly to commit fraud. They can claim plausible deniability because they rather passively allow the employees create their own frauds. Employees who understand the system will game that system by working with others in the organization and outside the organization to produce fake profits.
 AU 328, Auditing Fair Value Measurements and Disclosures, paragraphs 20 and 23; AU 332, Auditing Derivative Instruments, Hedging Activities, and Investments in Securities, paragraph 06. Also, in December 2007, in response to the auditing challenges presented by the subprime credit crisis and the transition to the new fair value accounting standard, the PCAOB staff issued Staff Audit Practice Alert No. 2, Matters Related to Auditing Fair Value Measurements of Financial Instruments and the Use of Specialists (December 10, 2007), which provides auditors with information about auditing fair value measurements and disclosures.
 AU 328.23; AU 332.40
Here’s the real Jane’s Addiction with one of my faves, Dave Navarro. (We once rode up in an elevator together at the W Union Square, NYC):