Lehman Bankruptcy 10 Year Anniversary
Six years ago I wrote a paper for a graduate accounting course on Lehman’s financial reporting. I thought I’d share it here because it’s an interesting story that is not well understood.
GAAP’s ultimate goals of fairness and accuracy in reporting require more than mere technical compliance.
— Judge Gerard E. Lynch, In re Global Crossing
The narratives leading to Lehman Brothers Holdings Incorporated’s (Lehman) Chapter 11 bankruptcy on September 15, 2008 were numerous and complex: generous homeownership subsidies (Slivinski 2009), amoral investment banks (Lewis 2008), rapacious mortgage originators (Berndt and Gupta 2008), industry-captured regulators (Miller 2009), hubristic central banks (Bernanke 2004), misunderstood correlations in structured finance, which has also been referred to as securitization or the shadow banking system. (Jaffee, et al. 2009), and corrupt political systems (Johnson 2009). As a result, diverse views on Lehman’s relationship with the wider financial crisis have emerged. These opinions range from viewing Lehman’s failure as a key cause of the financial crisis (Heider, Hoerova and Holthausen 2009) to dismissing it as a microeconomic footnote of a macroeconomic story (de Soto 2009). The focus of this paper is not on the moral culpability or the importance of Lehman’s business practices with relation to the financial crisis but rather on Lehman’s external financial reporting and the assurance services provided to it by Ernst & Young LLP (E&Y).
Lehman’s external financial reporting is salient due to certain “colorable claims”  detailed in the bankruptcy examiner’s report (Valukas 2010, 990). A colorable claim is “a claim strong enough to have a reasonable chance of being valid if the legal basis is generally correct and the facts can be proven in court.” (Cornell University Law School n.d.). The U.S. Trustee Program is an agency within the Department of Justice whose mission is to “promote integrity and efficiency in the nation’s bankruptcy system by enforcing bankruptcy laws, providing oversight of private trustees, and maintaining operational excellence.” (Department of Justice 2010). The U.S. Trustee appointed Anton Valukas, chairman of the prestigious law firm Jenner & Block (Jenner & Block 2012), as the examiner to “investigate the acts, conduct, assets, liabilities, and financial condition of the debtor” and to “file a statement of any investigation conducted . . . including any fact ascertained pertaining to fraud, dishonesty, incompetence, misconduct, mismanagement, or irregularity in the management of the affairs of the debtor, or to a cause of action available to the estate” ( U.S.C. § 1106(a) 2011). The 2209 page report filed by the examiner (the Valukas report) is the most researched explanation of Lehman’s bankruptcy publicly available and is based entirely on primary sources: internal Lehman e-mails, interviews, and financial statements (Valukas 2010, 32). No other detailed account of Lehman’s bankruptcy or E&Y’s involvement has emerged, thus this paper will primarily be based on the contents of the Valukas report.
None of the parties involved have publicized evidence that contradicts the colorable claims asserted in the report: that Lehman Brothers materially misrepresented its balance sheet in public filings (Valukas 2010, 891), failed to disclose its accounting treatment of certain repurchase agreements as sales (Valukas 2010, 985), and Lehman’s external auditor, E&Y, was aware of the misrepresentations and lack of disclosure but did not notify the audit committee or Lehman’s investors (Valukas 2010, 1027). These assertions fall squarely in the province of financial reporting and assurance and will be analyzed through that lens in the first half of the paper. The second half of the paper reviews the academic literature, media coverage, and legal consequences related to the Valukas report, Lehman’s financial reporting, and E&Y’s external audit.
The Valukas report raises two financial reporting issues: can a company report a repurchase agreement as a sale of securities rather than a financing transaction pursuant to ASC 860 (previously SFAS 140)? Furthermore, if a public corporation does report certain repurchase agreements as sales, must it disclose this alternative accounting treatment in the public filings’ Management’s Discussion and Analysis (MD&A)?
There are two definitions for repurchase agreements (Choudhry 2011, 146). The legal definition is that a repurchase agreement is the sale, with transfer of title, of a security to a buyer. The seller simultaneously promises to buy the security back from the buyer with a markup. The markup compensates the buyer for foregoing the use of their cash. The economic definition sees this markup as the interest component of a loan secured by the instrument.
Typically the cash received and paid back by the seller in a repo transaction is 2% less than the value of the securities (Valukas 2010, 767). This “haircut” protects the buyer from changes in the value of the collateral, it is identical to a down payment on a loan to reduce the loan to value ratio. Market participants use repurchase agreements (repos) for short-term investing and financing since most repos have a maturity of less than one year.
The U.S. GAAP accounting treatment of repurchase agreement generally adopts the economic definition by requiring in SFAS 140 (now ASC 860–10–55–51) that companies recognize these transactions on financial statements as secured borrowing or lending and not as sales or purchases. This accounting treatment is consistent with the FASB’s conceptual framework which states that the financial reporting of instruments such as repurchase agreements should provide information that helps investors “assess the amount, timing, and uncertainty of future net cash flows for an entity” (FASB 2010, 1).
Lehman complied with U.S. GAAP for the “tens of billions of dollars of repo transactions [used] in its normal course of business for financing purposes (“ordinary repo” or “traditional repo” transactions).” (Jenner & Block 2012, 751) The collateral securities remained on Lehman’s balance sheet, cash increased, and a corresponding liability was created to represent the obligation to repay the cash. Lehman’s 10-K and 10-Q disclosures stated that all repo transactions were treated as liabilities in its financial statements. Unbeknownst to the investing public a certain subset of repo transactions, called “Repo 105” and “Repo 108”, were treated as sales: cash increased, securities inventory decreased, and a forward agreement to repurchase was recognized. Interest expense was still recognized in the income statement.
The terms Repo 105 and Repo 108 refer to the haircut these transactions used. As their names imply, Repo 105 and Repo 108 had a 5% and an 8% haircut respectively. Economically they were otherwise indistinguishable from Lehman’s other repo instruments as they “were conducted with the same collateral and substantially the same counterparties” (Jenner & Block 2012, 770). Repo 105 transactions used government bonds as collateral while Repo 108 transactions used publicly traded equity securities.
These repurchase transactions had a greater haircut than the traditional 2% due to a perceived loophole in the accounting standards for sales of financial assets. SFAS 140 Paragraph 9 (ASC 860–10–40–5) outlines three conditions which must be met for a sale to occur:
1. The transferred assets are legally isolated from the seller.
2. The purchaser can pledge or exchange transferred assets freely.
3. The seller does not have a repurchase agreement or call option tied to the transferred assets.
Lehman proved to its auditors that the transferred assets were isolated by obtaining a true sale opinion from a law firm in the United Kingdom, Linklaters. They were unable to obtain such an opinion in the United States (Jenner & Block 2012, 784). In practice this meant that all Repo 105 and Repo 108 transactions were performed by Lehman Brothers International Europe (LBIE) based in London rather than Lehman Brothers Holding Inc. (LBHI) based in New York. Securities were transferred to London using standard intercompany repurchase agreements. Since Lehman reported as a consolidated entity “the LBIE‐only Repo 105 practice impacted LBHI’s publicly reported balance sheet and leverage ratios” (Jenner & Block 2012, 787). This true sale opinion in the UK arguably fulfilled the first and second requirement of a sale.
The third requirement is the most complicated to explain since Lehman had to literally prove that its repurchase agreements did not fit the definition of a repurchase agreement. The following table explains the corresponding terms to avoid confusion.
SFAS 140 Paragraph 218 (ASC 860–10–40–24b) states that the borrower’s right to repurchase the securities exists if the transferor is protected by “obtaining cash or other collateral sufficient to fund substantially all of the cost of purchasing identical replacement securities during the term of the contract so that it has received the means to replace the assets even if the transferee defaults. Judgment is needed to interpret the term substantially all and other aspects of the criterion that the terms of a repurchase agreement do not maintain effective control over the transferred asset.” In other words the borrower’s right to repurchase the collateral (which now legally belongs to the lender) exists if they have borrowed enough from the lender to replace the collateral in case the lender defaults (since the collateral securities would become part of the lender’s bankruptcy estate).
The crux of Lehman’s loophole is what the interpretation of “substantially all and other aspects of the criterion” is. The FASB provides guidance in the next sentence of the SFAS 140 Paragraph 218 (ASC 860–10–55–37): “arrangements to repurchase or lend readily obtainable securities, typically with as much as 98 percent collateralization (for entities agreeing to repurchase) or as little as 102 percent overcollateralization (for securities lenders), valued daily and adjusted up or down frequently for changes in the market price of the security transferred and with clear powers to use that collateral quickly in the event of default, typically fall clearly within that guideline. The Board believes that other collateral arrangements typically fall well outside that guideline.” Lehman interpreted this as a bright line rule where reducing the overcollateralization to 105 percent would automatically make the repurchase agreement disappear. I think that the reasonable reader would recognize that the verbiage used by the FASB indicates that this was not a bright line rule: “judgment,”, “interpret”, “typically”, “as little as”. The bright line phrases “fall clearly within” and “fall well outside” refer not to the 98–102 range but rather to the “arrangements to repurchase or lend readily obtainable securities” and “other collateral arrangements”.
Lehman used the reverse logic of the standard by asserting that their right to repurchase does not exist if they did not borrow enough to replace the securities (i.e. if they borrow $100 cash by repoing $105 of securities then the $100 cash is insufficient to replace the securities if the lender defaults), thus the third condition for a sale was met. Had the overcollateralization been 120% then the accounting treatment could have been more ambiguous under this standard. To justify a material haircut the collateral would have to be impaired or illiquid. Since the “Repo 105 transactions were structurally and substantively identical to ordinary repo transactions” in all regards except for the 3 or 6 percentage points difference a reasonable accountant would conclude that the Repo 105 and 108 transactions fall within the FASB’s guidelines and should have been treated as repurchase agreements.
Lehman’s loose interpretation of the standard dates back to the publication of SFAS 140 in September 2000 when senior executives, external auditors, and attorneys met to analyze the pronouncement. Over the course of several months they developed an accounting policy for Repo 105 and Repo 108 transactions based on their interpretation of SFAS 140 Paragraph 218 (Jenner & Block 2012, 766). The policy laid dormant until stress on Lehman’s short term funding made managing publicly reported leverage ratios a priority. As interviews with traders indicate it was not until 2007 that these special repo transactions were used in large amounts: “in mid‐2007 “there was a definite change” and the firm began “trying desperately to reduce its balance sheet.” […] Repo 105 transactions were used as a shortcut for meeting quarter‐end balance sheet targets (i.e., avoiding balance sheet limit breaches) and reaching the firmwide net leverage ratio target” (Jenner & Block 2012, 845). This information was never disclosed to analysts or investors despite the emphasis placed on “deleveraging” and “selling assets” during earnings calls (Jenner & Block 2012, 847). Concretely Repo 105 had the following effect on Lehman’s leverage ratios (net assets divided by tangible equity capital).
Lehman’s external auditor, Ernst & Young, defined materiality for Lehman’s 2007 audit as “any item individually, or in the aggregate, that moves net leverage by 0.1 or more (typically $1.8 billion)” (Jenner & Block 2012, 889). The Securities and Exchange Commission (SEC) specifically requires registered public corporations like Lehman to identify, in the MD&A section, commitments that are likely to materially affect liquidity and “any off-balance sheet financing arrangements” (Item 303(a)(2)(ii) of Regulation S‐K). This indicates that the use of Repo 105 and Repo 108 should have been disclosed it its SEC filings, yet “a sophisticated reader of Lehman’s financial statements would not have been able to ascertain from Lehman’s 2007 Form 10‐K or its first and second quarter 2008 Forms 10‐Q the amount of Lehman’s Repo 105 usage, nor even ascertain the fact that Lehman was engaged in these transactions […]” (Jenner & Block 2012, 984). As a result Lehman’s filings were materially misleading even if the Repo 105 and Repo 108 were accounted for correctly (Jenner & Block 2012, 987).
The CEO of Lehman, Dick Fuld, and the three CFOs in charge between December 2004 and Lehman’s bankruptcy all deny having any involvement with the Repo 105 and Repo 108 programs (Jenner & Block 2012, 917–944). Internal documents and interviews contradict Fuld’s statement and “no reasonable dispute exists that each of Lehman’s Chief Financial Officers from late 2007 to September 2008 possessed some knowledge of and/or involvement with multiple aspects of Lehman’s Repo 105 program […]” (Jenner & Block 2012, 921). On the other hand the Board of Directors was not aware of the accounting manipulation because Ernst & Young did not notify the audit committee and management did not disclose the use of Repo 105 in presentations to the Board (Jenner & Block 2012, 947). In fact, Ernst & Young’s lead partner on the Lehman engagement, William Schlich, knew about Repo 105 since its creation and was “comfortable with the Policy for purposes of auditing financial statements” but E&Y did not draft or prepare Lehman’s Repo 105 Accounting Policy (Jenner & Block 2012, 949). From the evidence available the Examiner concluded that there are colorable claims against the CEO and CFOs for breaching their fiduciary duties as well as the public auditor Ernst & Young for the lack of disclosure (Jenner & Block 2012, 990).
This leads us to a discussion from the assurance services perspective. The report attempts to answer several questions with regards to Lehman’s external auditor: did E&Y’s partners have a positive obligation to disclose substantive fraud allegations to the firm’s audit committee? Was E&Y’s lead partner on the Lehman audit correct in “becom[ing] comfortable with [Repo 105] for purposes of auditing financial statements”? Lastly, did E&Y have any responsibility for the information contained in quarterly filings and Management’s Discussion and Analysis?
On May 16, 2008 Matthew Lee, “then‐Senior Vice President in the Finance
Division responsible for Lehman’s Global Balance Sheet and Legal Entity Accounting” ( (Jenner & Block 2012, 956), blew the whistle in a letter to the CFO. The Board of Director’s “Audit Committee explicitly instructed Lehman’s Corporate Audit Group and Ernst & Young to keep the Audit Committee informed of all of Lee’s allegations” (Jenner & Block 2012, 946). On June 12th E&Y partners William Schlich and Hillary Hansen interviewed Lee. Hansen’s notes of the meeting describe Lee’s assertions regarding the use of Repo 105 at the end of the quarter to reduce leverage (Jenner & Block 2012, 1034). When Schlich met with the Audit Committee the next day he did not disclose Lee’s allegation as instructed (Jenner & Block 2012, 960) and Schlich did not follow up on Repo 105 before E&Y “signed a Report of Independent Registered Public Accounting Firm for Lehman’s second quarter 2008 Form 10‐Q on July 10, 2008” (Jenner & Block 2012, 961). This was a direct violation of the Public Company Accounting Oversight Board (PCAOB) auditing standards: “Fraud involving senior management and fraud (whether caused by senior management or other employees) that causes a material misstatement of the financial statements should be reported directly to the audit committee.” (Auditing Standard 316.79)
E&Y’s lead partner on the Lehman engagement, William Schlich, was aware of the Repo 105 policy before the whistleblower came forward. In fact, Ernst & Young “clearly. . .concurred with Lehman’s approach” although they did not review the Linklaters true sale opinion (Jenner & Block 2012, 949) . In 2007 Lehman employees created a “netting grid” that included Repo 105 transactions and was used by E&Y “in connection with its 2007 year‐end audit of Lehman” (Jenner & Block 2012, 952). This work sheet provided all the information E&Y needed to determine that Repo 105 transactions made Lehman’s balance sheet materially misleading. The above facts led the Examiner to assert “that sufficient evidence exists to support colorable claims against Ernst & Young LLP for professional malpractice arising from Ernst & Young’s failure to follow professional standards of care” (Jenner & Block 2012, 1027).
On Thursday, March 11, 2010 the Valukas report was released to the public and the media soon ran stories on its contents. On March 12th the New York Times published a news article titled “Lehman Bros. Hid Borrowing, Examiner Says” (De la Merced and Sorkin 2010). The front page article begins with a factual error: “Lehman Brothers used accounting sleight of hand to conceal the bad investments that led to its undoing.” Repo 105 was not used “to conceal bad investments,” in fact it was unrelated to Lehman’s real estate operations or its poorly timed acquisition of developer Archstone. Repo 105 was used to artificially lower Lehman’s reported leverage and the transactions only used government securities and publicly traded stocks. The article compounds the error while getting closer to the truth in the fourth paragraph: “Lehman used what amounted to financial engineering to temporarily shuffle $50 billion of troubled assets off its books in the months before its collapse in September 2008 to conceal its dependence on leverage, or borrowed money.” Again, troubled assets were not removed from the books, it was U.S. Treasuries and other low risk investments. The article becomes even more specific at the end on what troubled assets the authors believe were a part of Repo 105 transactions: “mostly illiquid real estate holdings.” The authors of the article are conflating two distinct items: the collateral used for Repo 105 and the collateral used for emergency loans from other banks and the Federal Reserve (an entirely separate issue also discussed in the Valukas report). Other than this surprising mistake the article downplays the accounting fraud with euphemisms: “sleight of hand”, “gimmicks”, “engineering”, “maneuvers”, “tool”, and “aggressive accounting”. The article does not mention that Ernst & Young committed malpractice or that filing materially misleading financial statements with the SEC is a crime.
The Wall Street Journal (WSJ) published an incoherent description of Repo 105 on March 13th: “Lehman was able to book what looked like an ordinary asset for cash as an out-and-out sale, drastically reducing its leverage” (Craig and Spector 2010). In other words the article literally claims that a balance sheet account was booked as an income statement account. This description would be a humorous misunderstanding if it wasn’t for the entirely accurate description in fine print of a footnote under the graphic:
As for the legal repercussions of Lehman’s Repo 105 accounting, on July 28, 2011 the NYT published an article titled “Lehman Case Hints at Need to Stiffen Audit Rules”. The article describes a class-action lawsuit against Lehman’s executives and E&Y. The judge in the case dismissed the charges against E&Y regarding its 2007 audit, finding that the Netting Grid was not sufficient evidence that E&Y did not have a reasonable basis for its audit opinion. Nevertheless the judge did not dismiss claims against E&Y relating to the whistleblower, Matthew Lee, and the Audit Committee. The rest of the lawsuit is awaiting trial. Confusingly the article discusses GAAP as if it contained auditing standards and was written by the PCAOB.
The perceived ambiguity in FAS 140’s language led the SEC to not pursue Lehman or E&Y (Eaglesham and Rappaport 2011). In December 2010 New York State Attorney General Andrew Cuomo filed a lawsuit against E&Y to seize $150 million from E&Y; the same amount the firm charged Lehman for seven years of audits. The case is awaiting trial in the Manhattan state court (Dye 2012).
Due to the Valukas report, the FASB amended FAS 140 (ASC 860) with Accounting Standards Update №2011–03 in April 2011. This amendment removed the “the criterion relating to the transferor’s ability to repurchase or redeem financial assets.” (FASB 2011, 2) This removed the perceived bright line rule and improved the standard since the borrower’s ability to repurchase assets similar to the collateral is irrelevant to the economic substance of the secured lending transaction. This case highlighted the dangers of loosely interpreting accounting standards to manipulate financial statements and the difficulty of prosecuting accounting fraud when it is only incidental to a major bankruptcy.
U.S.C. § 1106(a). Duties fo the Trustee and Examiner. Washington, D.C., January 2011.
Bernanke, Ben S. The Great Moderation. February 20, 2004. http://www.federalreserve.gov/boarddocs/speeches/2004/20040220/default.htm (accessed January 21, 2012).
Berndt, Antje, and Anurag Gupta. “Moral Hazard and Adverse Selection in the Originate-to-Distribute Model of Bank Credit.” SSRN. October 24, 2008. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1290312 (accessed January 21, 2012).
Cornell University Law School. Legal Information Institute — Colorable Claim. n.d. http://www.law.cornell.edu/wex/colorable_claim (accessed January 22, 2012).
de Soto, Jesus Huerta. Money, Bank Credit, and Economic Cycles. Auburn: Ludwig von Mises Institute, 2009.
Department of Justice. U.S. Trustee Program Mission. 2010. http://www.justice.gov/ust/eo/ust_org/mission.htm (accessed January 22, 2012).
Heider, Florian, Marie Hoerova, and Cornelia Holthausen. “Liquidity Hoarding and Interbank Market Spreads.” European Central Bank Working Paper Series. December 2009. http://www.ecb.int/pub/pdf/scpwps/ecbwp1126.pdf (accessed January 21, 2012).
Jaffee, Dwight, Anthony W. Lynch, Matthew Richardson, and Stijn Van Nieuwerburgh. “Mortgage Origination and Securitization in the Financial Crisis.” In Restoring Financial Stability, by Viral Acharya and Matthew Richardson, 61–82. New York: Wiley, 2009.
Jenner & Block. Anton R. Valukas. January 2012. http://www.jenner.com/people/AntonValukas (accessed January 22, 2012).
Johnson, Simon. “The Quiet Coup.” The Atlantic. May 2009. http://www.theatlantic.com/magazine/archive/2009/05/the-quiet-coup/7364/ (accessed January 21, 2012).
Lewis, Michael. The End. November 11, 2008. http://www.portfolio.com/news-markets/national-news/portfolio/2008/11/11/The-End-of-Wall-Streets-Boom (accessed January 21, 2012).
Miller, David. “Revolving doors, accountability and transparency: Emerging regulatory concerns and policy solutions in the financial crisis.” Organisation for Economic Co-operation and Development (OECD). July 6, 2009. http://www.oecd.org/dataoecd/22/15/43264684.pdf (accessed January 21, 2012).
Slivinski, Stephen. “Region Focus.” Richmond Federal Reserve. January 12, 2009. http://www.richmondfed.org/publications/research/region_focus/2008/fall/pdf/cover_story.pdf (accessed January 21, 2012).
Valukas, Anton R. Report of Examiner Anton R. Valukas. Examiner’s Report, Chicago: Jenner & Block LLP, 2010.