Lehman Brothers and Repo 105 · Prep

The Professor's Questions

The Role of Auditors, Law Firms, and Management.

If you were an investor in Lehman Brothers, how would you evaluate the riskiness of Lehman’s business?
Why did Lehman use the repo 105 transactions?
What institutions, if any, were at fault for Lehman’s use of repo 105's and how would you go about improving the institutions?

My Answers

Q1. Investor view of risk

I would not start with the net leverage they advertised. I would start with how the firm dies.

Lehman had no stable retail deposits. It rolled short-term loans every day. In 1998, a rumor after LTCM already showed how fast that funding can run. After Bear Stearns in March 2008, that was not a theoretical risk. It was the whole business.

Then I would look at what sat on the thin equity. From 2004 to 2007, assets grew about $300 billion; equity grew $6 billion. That is 50 times leverage on the new dollar. The 2007 record — $19.3 billion of revenue, $4.2 billion of profit — came from packing and trading exotic mortgage paper, the high-margin stuff. Fuld said smart risk management is never putting yourself where you cannot fight another day. The portfolio said the opposite: they had switched from moving assets for clients to holding them.

Then I would distrust the ratio itself. Callan told the market net leverage was 15.4, down from 16.1. Lowitt later said 12 times, "low double digits," "we will operate conservatively." Einhorn had been short since July 2007 and said they were about 40 times on tangible equity, holding things that should not be levered at all. By June 2008 Lehman was the most shorted firm on the Street. He was looking at the right object.

Matthew Lee, who signed the balance sheet for a thousand entities, gave the inside version in May 2008: on the last day of each month the books held about $5 billion more net assets than management actually ran; tens of billions of accounts were unsubstantiated; tens of billions of inventory probably could not be bought or sold at the marks, in that size, in that market. That is not a footnote. That is the risk: a confidence business, financed overnight, sitting on sticky marks.

If I had also seen Exhibit 2, I would have done one more piece of arithmetic. Ernst & Young’s own papers said a 0.1 move in net leverage was material, typically $1.8 billion. So tangible equity was about $18 billion. Repo 105 at Q2 2008 was $50.4 billion — 28 times that threshold, about 2.8 turns of net leverage. The "12.1" was closer to 14.9 without the quarter-end sale. The "15.4" in Q1 was closer to 18. The rating agencies were shown the cleaned number. S&P still praised the "strong funding/liquidity profile" on June 2, then the firm pre-announced its first quarterly loss as a public company one week later.

So my scorecard as an investor: (1) funding model — overnight confidence, no deposits; (2) asset quality — sticky, levered, marked in an illiquid market; (3) governance of the number — if the ratio only looks right on the last day of the quarter, it is not a risk metric, it is a press-release metric. Repo 105 is the proof of (3). Even without knowing its name, Einhorn and Lee were already pointing at it.

Q2. Why Repo 105

They used it because they needed the leverage ratio to fall, and they did not want to sell the sticky inventory at a real loss.

From mid-2007 the market stopped staring at revenue and started staring at the balance sheet. Rating agencies, analysts, and short sellers were all reading one number: net leverage. Fuld made McDade "balance sheet czar." Businesses were ordered to cut the ratio to avoid a downgrade. McDade wanted them to sell inventory. Traders found the assets hard to sell, or did not want the discount. So they "rented the balance sheet": Repo 105 moved inventory off for a few days, cash paid down other liabilities, equity sat still, and the ratio printed lower. King said there was always quarter-end pressure, but from mid-2007 it jumped. "To the powers that be, Repo 105 counted as a balance sheet reduction." McDade’s own phrase: "another drug we are on."

The accounting trick is small and specific. A normal repo is financing: 2% haircut, asset stays, cash and a liability both go up. SFAS 140 says that at 98–102% collateral you still have effective control, so you cannot call it a sale. Lehman took a 5% or 8% haircut, claimed it had surrendered control, and booked a sale. Exhibit 4: the same $50 billion, treated as financing, leaves net leverage at 17.6 after you repay; treated as a sale, it falls to 15.6. That two-turn cosmetic is the whole product.

They then hid the product. No public disclosure. The obligation to buy tens of billions back in a few days did not appear. The trades were dumped into the derivatives footnote. US law firms would not call them sales. Linklaters would, if the securities were "sited" in the UK and done under English law. So American paper was walked through LBIE in London and walked back after the reporting date.

Watch the dose. $24.5 billion at end-2006. $38.6 billion at end-2007. $49.1 billion in Q1 2008 — the Bear quarter, a $10.5 billion jump. $50.4 billion in Q2. Sum of the last three quarter-ends: $138 billion. They had told Ernst & Young the cap would be $20–30 billion. They blew through it. They used more of the drug in the days just before reporting dates. They took the cleaned ratio to S&P, Moody’s, and Fitch in May–June 2008 and argued that no negative action was justified. Repo 105 was not in the deck. The Fed and the SEC did not know either.

So the motive is not "we found a clever way to finance ourselves." Ordinary repo already did that. The motive is window dressing under pressure: hit the number the agencies and the earnings call were going to read, without taking the loss McDade said they should take. Richards, in the prison letter, describes the same slope from the other industry: once you start managing the number, even small changes in procedure leave you exposed. Lehman’s small change was 3 extra points of haircut and a London opinion letter. After that, every quarter needed a bigger dose.

Q3. Who was at fault, and how to improve

I would not pick one villain. I would say the system passed a form and failed a function. Management designed the drug and raised the dose. The law firm made the form legal in a jurisdiction that would say yes. The auditor blessed the policy, knew what it was for, heard $50 billion from a whistleblower, and still did not say the words "Repo 105" to the Audit Committee. The Audit Committee had every SOX chair in the room and never heard the fact. Rating agencies and the Fed were shown the cleaned ratio. Lee followed the Code of Ethics and was fired in about two weeks.

Management. They are first. Goldfarb is alleged to have invented it. Fuld set the growth model, pushed leverage, and pushed Antoncic out of the room when risk got tense. They told E&Y the cap was $20–30 billion, then printed $50.4 billion. They never disclosed. They took the number to the agencies. McDade at least named it a drug and wanted real sales; the traders were rewarded for the cosmetic. That is tone at the top: the ratio was the goal, the inventory was not.

Linklaters, and the US firms that refused. This is the session title. US counsel would not call Repo 105 a sale. That refusal was information. Lehman shopped the answer to London, then walked American securities through LBIE so the opinion would attach. A legal opinion that is "legalistic" and "form-driven" — Kelly’s words — is not a control. It is a ticket. Improving law firms: opinions should state what the structure is for, not only whether a haircut clears a paragraph. If the only commercial purpose is to print a quarter-end ratio, the letter should say so, or not be written. Forum-shopping a true-sale opinion for US paper should itself be a disclosure event.

Ernst & Young. They approved the Policy in 2001. Smith’s 2001 memo told them, in words, that the point was to net down the balance sheet. Jain flagged reputational risk in 2006 and it died in the hierarchy. Kelly told Schlich the justification was form-driven and that volumes spiked at quarter end. Materiality in their own papers was 0.1 of net leverage, about $1.8 billion; $50 billion is not a rounding of 31 times to 32 times. Most damning: on June 12 Lee told them about $50 billion coming off and going back. Cruikshank had ordered them to report every assertion, in the letter or not. Schlich was silent on June 13, silent on July 8, silent on July 22. On July 10 they issued an unqualified review on the 10-Q. The engagement letter said fraud involving senior management goes straight to the Audit Committee. They measured GAAP comfort — process. They did not measure "did the board hear the ugly fact" — outcome. Turley’s later line, "we feel very good about the work we did," is the paper talking.

How to improve auditors: (1) materiality for leverage window-dressing is the move in the ratio the market is using, not net income; (2) a whistleblower interview is not closed until every new fact is in the Audit Committee minutes, in the whistleblower’s words; (3) alumni moving from the audit firm into the CFO chair should trigger extra skepticism, not extra comfort; (4) "the accounting wasn’t challenged" cannot be the end of the disclosure analysis.

The Audit Committee. Independent members, a financial expert as chair, private sessions with the engagement partner — the process inputs of Sarbanes-Oxley were all green. They still did not know Repo 105 existed. You cannot oversee a number you have not been told is synthetic. Improve: the committee should receive, as a standing paper, quarter-end vs mid-quarter balance sheet, and a list of all true-sale opinions in use. If the auditor says a whistleblower letter is "pretty ugly" but "shouldn’t result in significant issues," that sentence is itself an issue.

Rating agencies, the Fed, the SEC. They are downstream. They were not told. Still, a 73% price drop, a first-ever quarterly loss, and Einhorn’s 40-times claim were already on the tape. An agency that praises "de-leveraging" without asking "sold or rented?" is measuring the press release. Improve: for broker-dealers who live on wholesale funding, require a gross and net leverage before quarter-end window dressing, or simply forbid calling a 5% haircut repo a sale.

The Code of Ethics, as an institution. This is where Soltes’s preface belongs. Lee did exactly what the paper asked. Outcome: dismissal, a gag, unpaid severance. Ninety percent of people say they would report; fewer than half did. Lehman is that gap in one career. Improving speak-up is not a new hotline. It is: the reporter still has a job, the board heard the fact that was not in the letter, and the next quarter’s ratio is published two ways. Those are outcomes. Training completion would have changed nothing.

Richards is the human version of the same slope. He also had a Big Four comfort letter. He also had an internal GC investigation that found nothing. He also told himself it was "just timing, two or three days." Repo 105 is timing, two or three days. The gray is where senior managers live, he wrote, and it is a dangerous place. The fix is not to pretend the world is black and white. The fix is what Soltes calls uncomfortable dissonance: a person outside the quarter-end huddle who can still kill the trade. Antoncic was that person, and she was asked to leave the room. Lee was that person, and he was fired. Jain was that person, and the email stopped at his manager.

If I have to put it in one classroom sentence: no single institution invented the lie; each institution certified a piece of the form, and so nobody owned the function. Improve them by changing what they count as success — not "was there a policy, an opinion, an independent committee, an audit fee, a code of ethics," but "did the number the CEO said on the call match the number without the drug, and did the person who said otherwise still work there."

Case Summary

In May 2008, Matthew Lee had been at Lehman for fourteen years. A senior vice president in the finance division, he was responsible for the balance sheets of more than a thousand legal entities around the world. He read the company's Code of Ethics over and over: anyone who discovers a possible violation has a duty to report it immediately; there will be no retaliation for good-faith reports. He hesitated for a long time, and after seeing a lawyer he concluded that this was his duty. On May 16 he handed his letter to Controller Martin Kelly, head of capital markets product control Gerard Reilly, CFO Erin Callan, and Chief Risk Officer Christopher O’Meara. The letter made six points: at month-end the books carried roughly US$5 billion of net assets that no one was really managing; billions of dollars of accounts across global entities did not reconcile; billions of dollars of inventory were nearly unsellable at their marked prices; systems and staff had not kept pace with the expansion since the 1994 IPO; the finance function in Mumbai might be efficiently distributing wrong numbers; and some internal audit executives lacked the skills to audit what had become a complex business. The words Repo 105 did not appear anywhere in it.

About two weeks later, his boss called him in and told him he had been laid off as part of a broader downsizing. No other reason was given. He fought it, then accepted a severance agreement with a confidentiality clause. Lehman went bankrupt in September 2008, and he never received all the money in the agreement. E&Y later said that Lehman had investigated the letter itself, and that in July 2008 management reported to the audit committee that the allegations were unfounded and there were no material issues; E&Y's own review was never finished, because the company had gone bankrupt.

Founded in 1850, Lehman was one of Wall Street's oldest investment banks, and once one of its most profitable. It was sold to Shearson/American Express in 1984 and spun off in a 1994 IPO. Richard Fuld, who came up as a trader, became chairman and CEO. Its historic strengths were fixed-income underwriting and trading. Investors worried that its fixed-income exposure was too large, so the firm diversified: fixed income fell from 55% of revenue in 1995 to 39% in 2007, with equities and advisory filling the gap. In the mid-2000s it deliberately chased faster growth, shifting from a low-risk brokerage model to a high-risk banking model — the balance sheet went from "moving" assets to "holding" them, with the risk staying on its own books. From 2004 to 2007, assets grew by about US$300 billion while equity grew by only US$6 billion. Fortune's Shawn Tully described Wall Street's game bluntly: take shareholders' small pile of cash, lever it up enormously, and use the debt to stack up securities; as long as interest rates fall, the portfolio swells, and a thin layer of equity can carve out huge returns.

The high profits came from riskier stuff. Like its peers, Lehman went ever deeper into packaging and trading exotic mortgage-backed securities. These high-margin products helped it post record profits of US$4.2 billion on revenue of US$19.3 billion in 2007. In Fuld's own words: smart risk management means never putting yourself in a position where you can't fight another day. But it had no retail deposits underneath it; it relied on short-term loans that had to be rolled over again and again. Once creditors panicked, the funding would stop. When LTCM nearly collapsed in 1998, rumours had already gone round that Lehman was in trouble; Fuld put down that crisis of confidence, and it was seen as the defining performance of his career.

In March 2008, the equally highly levered Bear Stearns collapsed. Since mid-2007, analysts and rating agencies had been watching investment banks' leverage; after Bear, it became an urgent task: the market had to see leverage going down. CFO Callan said she was fighting "hand-to-hand combat" against the rumours. On the March 18 first-quarter earnings call she said: we have deliberately brought leverage down; net leverage is 15.4x, versus 16.1 at year-end. On June 2, S&P cut the rating from A+ to A but still praised its funding and liquidity profile. Less than a week later, Lehman pre-announced its first quarterly loss since going public, US$2.8 billion, replaced its CFO, and announced its third capital raise of the year, US$6 billion. The new CFO, Ian Lowitt, said on June 16: gross leverage has fallen from 31.7 to 24.3, and net leverage from 15.4 to 12; the appropriate level of leverage right now is the low double digits; we will run the business conservatively.

Not everyone believed it. Hedge fund manager David Einhorn had been short since July 2007 and particularly distrusted the published leverage numbers: the problem was overall leverage and the loan portfolio; on tangible equity they were levered about 40 times, and they held things that should never be levered at all. By June 2008, Lehman was the most heavily shorted investment bank on Wall Street. Over the summer Moody's moved its outlook to negative while still writing that it was actively adding capital and deleveraging, and that its liquidity management remained sound. In August it cut 1,500 jobs, about 6%. In the first eight months of the year the share price fell 73%. On September 10 it reported another loss, of US$3.9 billion, and planned to sell most of its investment management business. Fitch put the rating on negative watch. Three days later, New York Fed president Geithner called Wall Street together and made it clear the government would not bail it out. That weekend there was no buyer. On September 15, Lehman filed for Chapter 11 bankruptcy, citing "significant liquidity problems." With more than US$600 billion in assets, it was the largest bankruptcy in US history.

Repo was a form of short-term financing everyone in the industry used. In 2007 the US and European repo markets were worth about US$11 trillion, double their 2002 size. The standard deal has two steps: first, borrow cash using assets on the books as collateral (a typical haircut is 2%, i.e. Repo 102); then repay the loan, take the assets back and pay a little interest. In accounting terms this is a financing: cash and short-term liabilities both increase, and the collateral stays on the balance sheet. SFAS 140 says that with collateral between 98% and 102%, the borrower retains effective control and the deal cannot be treated as a sale. To be a sale, it must meet three conditions for surrendering control — and the standard itself notes that most repo market participants are not used to treating these transactions as sales.

Lehman invented Repo 105: raise the haircut to 5% or 8%, claim that it had surrendered control, and therefore book the deal as a sale. Booked as a sale, assets on the balance sheet fall (by the amount lent plus the haircut), the cash received goes to pay down other liabilities, equity is unchanged, and leverage looks lower. Exhibit 4 illustrates it: the same US$5 billion, treated as a financing, leaves net leverage at 17.6 after paying down debt; treated as a sale, net leverage drops to 15.6. From the fourth quarter of 2007 to the end of the second quarter of 2008, the practice removed more than US$138 billion of net assets in total. Volumes were ramped up in the days before each reporting date to push down net leverage. It was never publicly disclosed. The financial statements counted it as derivatives, mixed into the note on "more traditional" derivatives; readers could not see an obligation to repurchase tens of billions of dollars of securities within days.

From mid-2007, what the market watched shifted from revenue to the balance sheet. Chief Risk Officer Madelyn Antoncic had already urged caution, pulling back and more study at the executive committee in late 2006; in 2007 she was sidelined — whenever a tense risk topic came up, Fuld had her leave the room along with the deal team. At the same time, senior executives pushed every business to sell assets and push down net leverage, to avoid a downgrade. When traders found sticky assets hard to sell, or didn't want to sell them at a discount, they could "rent the balance sheet": use Repo 105 to move inventory off temporarily and still hit their net leverage targets. Trader Mitchell King recalled that there had always been quarter-end pressure, but it rose noticeably from mid-2007; the goal was to be profitable and shrink the balance sheet at the same time; to the people in charge, Repo 105 counted as balance-sheet reduction. In March 2008 Fuld appointed veteran Bart McDade as "balance sheet czar." McDade wanted traders to be disciplined, and in an email called Repo 105 "another drug we are on." His position was: sell the inventory. E&Y's 2007 walk-through workpapers noted that net leverage was an important ratio in rating agency analysis and was also included in Lehman's earnings press releases. On its calls Lehman repeatedly stressed deleveraging and never said how much of the number came from Repo 105. In May and June 2008 it met Fitch, Moody's and S&P and used the balance-sheet reduction and deleveraging it had achieved, and was about to achieve, to argue against negative rating action; Repo 105 appeared nowhere in the materials. The New York Fed and the SEC didn't know either.

E&Y had been Lehman's auditor since it became independent in 1994, and also did other accounting, advisory and tax work. From 2001 until the bankruptcy its fees exceeded US$150 million; one analysis put the decade's total at over US$185 million, an average of about US$31 million a year. For seven years running Lehman was one of E&Y's fifteen largest clients, ranking eighth by US audit fees in 2007. Quite a few accountants worked on this one account alone. Several of Lehman's senior finance executives were E&Y alumni: CFO Christopher O’Meara (2004–2007) had been a senior manager in E&Y's financial services practice and jumped ship while auditing Lehman in 1994; the CFO before him, David Goldfarb (2000–2004), had been a senior partner in E&Y's financial services business from 1979 to 1993, and some pointed to him as the person who invented Repo 105 internally. On paper it complied with Sarbanes-Oxley: E&Y reported directly to the audit committee, the members were independent, the chair was a financial expert, and the committee met the engagement partner alone without management present. The engagement letter stated that any fraud involving senior management, or fraud causing a material misstatement of the financial statements, would be reported directly to the audit committee.

In 2001, Lehman showed its internal accounting policy statement to the then engagement partner Kevin Reilly, as well as Matthew Kurzweil and William Schlich, hoping to use Repo 105. E&Y approved the policy. It did not help draft it and did not advise on the transactions. But the audit partners knew perfectly well what it was for. On August 19, 2001, Lehman's Kristine Smith sent the three of them a memo: this accounting treatment lets Lehman turn repos into inventory transactions, giving it greater capacity to net down the balance sheet. Lehman also told E&Y that no US law firm would characterise Repo 105 as a "sale," but that the UK firm Linklaters had given an opinion that, under strict conditions, it could be characterised as one — the securities had to be "located" in the UK, and the transactions had to be done in the UK under English law. Later Lehman put billions of dollars of US fixed-income government securities into Repo 105 as well: a US entity would first transfer the securities to its UK affiliate, LBIE, which then did the transaction. Schlich later said he was unaware of such transactions; Smith and Kelly insisted in testimony that they had explicitly told Schlich US securities were included.

In September 2006, Bharat Jain, a member of the E&Y audit team, worried that this could be a reputational risk and wrote to his superior Jennifer Jackson: what do we think about how much Lehman should be doing? Have we compared it to competitors, are there industry publications, is there regulatory guidance? Jackson later said that raising reputational risk was not an everyday occurrence and she couldn't immediately think of another example; but she couldn't recall whether she had raised the matter with Schlich. Lehman had initially told E&Y by email that it planned to cap volumes at roughly US$20–30 billion. It began exceeding the cap in 2007, and by the end of the second quarter of 2008 it had reached US$50.38 billion. E&Y received many internal documents stating the volumes. Kelly, who had just become financial controller in December 2007, told Schlich that volumes spiked at every quarter-end, and that the accounting rationale for sale treatment was "legalistic" and "form-driven," so Lehman had to rely on the Linklaters opinion. Kelly later said that E&Y was still "comfortable with the GAAP treatment, for the same reasons Lehman was comfortable." The 2007 audit workpapers defined materiality as anything that, individually or in aggregate, moved net leverage by 0.1 or more (typically US$1.8 billion). That was Lehman's own threshold; E&Y never expressed a separate view on the materiality of Repo 105. In 2009 Schlich admitted to the bankruptcy examiner that E&Y had no "bright-line" materiality rule, and that in a balance-sheet context it depended on facts and circumstances. E&Y's chief executive, Jim Turley, later defended the firm: every institution manages its balance sheet, and managing it is not in itself window dressing; what matters is that investors have an accurate picture of leverage; these repos — and the accounting was never challenged — would probably have changed leverage only from 31 times to 32 times, and Lehman would still have been the bank with the third-highest leverage.

Lee's letter reached E&Y. Schlich immediately told the chair of the audit committee and insisted that management inform the SEC and the New York Fed. He later said he had discussed the whistleblower letter with the committee at least three times in June and July 2008. The letter itself did not mention Repo 105, but it raised a series of possible accounting irregularities. The committee chair, Thomas Cruikshank, asked E&Y to assist the company's internal audit investigation, to meet Lee, and to report every allegation that came up during the investigation — whether or not it was in the letter — to the committee. Schlich wrote to colleagues: we are also dealing with a whistleblower letter that "on its face is pretty ugly," and it will take a great deal of time; based on what he had seen so far, he was confident it would not turn into a material issue for the financial reporting; that, plus some very difficult off-balance-sheet accounting issues, had everyone under a lot of pressure.

On June 12, 2008, Schlich and another partner, Hillary Hansen, interviewed Lee. On the spot, Lee added an allegation that wasn't in the letter: at the end of the second quarter of 2008, Lehman had used Repo 105 to take US$50 billion of fixed-income securities off the balance sheet, and put them back a few days later. The next day, June 13, Schlich briefed the audit committee on the investigation's progress and did not mention Repo 105. At the July 8 committee meeting, and when internal audit reported its findings to the committee on July 22, he still said nothing. Internal audit's conclusion: "no material issues identified." The audit committee never knew about Repo 105, never knew it was undisclosed, never knew US securities were being shifted to a UK affiliate to do it, and never knew that its size, its growth and its quarter-end timing had been flagged as problems. Meanwhile, on July 10, E&Y issued an unqualified review report on Lehman's 10-Q — less than a month after interviewing Lee.

In the epilogue, Turley said in the summer of 2010: at the end of the day, we feel good about the work we did at Lehman. He argued that the partners "took some comfort in the fact that the issue was a disclosure issue around Repo 105," since the accounting was aligned with US rules. Later, on television, he placed Lehman's failure in the context of a global financial meltdown, mortgage defaults, falling asset prices and a liquidity crisis, and said that without governments stepping in, many financial institutions would have filed for Chapter 11. He acknowledged that the system needed reform.

Key Data

Dimension	Figure / fact	Source
Scale	Bankrupt 2008.9.15, assets >$600B, largest in US history	Text p.4
Growth and profit	2007 revenue $19.3B, profit $4.2B; 2008 first loss $2.8B, then another $3.9B loss	Text p.3–4
Leverage	2004–2007 assets +$300B / equity +$6B; reported net leverage 2006Q4 14.5 → 2008Q1 15.4 → 2008Q2 12.1; gross leverage 2008Q1 31.7 → Q2 24.3	Text p.2; Exhibit 2
Repo 105	Haircut 5% or 8% (ordinary repo 2%); 2006Q4 $24.5B → 2008Q2 $50.4B; three-quarter total >$138B	Text p.5; Exhibit 2
Window-dressing effect	Illustration 17.6 → 15.6; backing out from 0.1 = $1.8B, Q2's reported 12.1 corresponds to roughly 14.9 "without window dressing"	Exhibit 4; Text p.9
Funding model	No retail deposits, reliant on short-term rollovers; 1998 LTCM confidence crisis; after Bear Stearns in 2008.3, the market had to see deleveraging	Text p.3
Market pressure	Einhorn short from 2007.7; most-shorted by 2008.6; share price -73% in first eight months; S&P cut from A+ to A on 6.2	Text p.3–4
Audit	E&Y since 1994; >$185M over ten years, ~$31M a year on average; 8th by US audit fees in 2007; approved the policy in 2001	Text p.7 and note c
Law firms	US firms would not call it a sale; Linklaters opinion under English law; US securities routed to the UK via LBIE	Text p.8
Whistleblowing	Lee's letter 2008.5.16 (Exhibit 1); laid off about two weeks later; revealed the $50B at the 6.12 interview; audit committee never told about Repo 105; unqualified 10-Q review 7.10	Text p.2, 9–10
Materiality	0.1 move in net leverage ≈ $1.8B; Q2 volume $50.38B ≈ 28× the threshold	Text p.9
Supplementary: CA	FY2000 Q2 revenue inflated 53%; EPS $0.60 → $0.05 after removing accelerated recognition	Letter from Prison Exhibit 4

Net leverage without window dressing: the case says "a 0.1 move in net leverage typically corresponds to $1.8B" → tangible equity TE ≈ $1.8B / 0.1 = $18B. Q2 2008 volume of $50.4B corresponds to 50.4 / 18 = 2.80 turns of net leverage. Reported 12.1 + 2.80 ≈ 14.9. Q1 2008: 15.4 + 49.1/18 ≈ 18.1.

Three-quarter total: $38.6 + $49.1 + $50.4 = $138.1B.

Key People

Person	Role	Position / part in the story	Relationships
Matthew Lee	Finance SVP, responsible for balance sheets of 1,000+ entities	Wrote the letter under the Code of Ethics; revealed the $50B of Repo 105 at the interview; laid off soon after	Reported to Kelly / Callan / O’Meara; interviewed by E&Y
Richard Fuld	Chairman and CEO	Set the tone of high growth and a principal-holding model; sidelined the CRO; appointed a "balance sheet czar"	Ultimate decision-maker on the executive committee
Erin Callan / Ian Lowitt	Successive CFOs	Reported net leverage falling 16.1 → 15.4 → 12 to the market; Callan was also a recipient of Lee's letter	Spoke to investors and rating agencies
Madelyn Antoncic	Chief Risk Officer	Urged caution in late 2006; shut out of tense risk meetings in 2007	Sidelined by Fuld
Bart McDade	"Balance sheet czar"	Wanted a smaller balance sheet, called Repo 105 "another drug," argued for selling inventory	Appointed by Fuld
Mitchell King	Senior trader	Testimony: quarter-end pressure rose noticeably from mid-2007; Repo 105 counted as balance-sheet reduction	Execution level
David Goldfarb	Former CFO (2000–2004), former E&Y senior partner	Named as the person who invented Repo 105 internally	Audit-firm alumnus inside the audited company
Christopher O’Meara	CFO (2004–2007), later CRO	E&Y alumnus; one of the recipients of Lee's letter	Same as above
Martin Kelly	Financial controller from 2007.12	Told Schlich volumes spiked at quarter-end and the rationale was "legalistic / form-driven"; first recipient of Lee's letter	Internal finance voice speaking to E&Y
William Schlich	E&Y engagement partner	One of those who knew about the policy's approval; after interviewing Lee, did not raise Repo 105 with the audit committee; signed off the 10-Q on 7.10	Owed a duty of communication to the audit committee
Bharat Jain / Jennifer Jackson	E&Y audit team member / senior manager	Raised reputational risk in 2006; Jackson unsure whether she escalated it	Hierarchy within the firm
Linklaters	UK law firm	Gave the "sale" legal opinion US firms would not	Made the form hold up
Thomas Cruikshank	Audit committee chair	Asked for every one of Lee's allegations to be reported	Had oversight power, but never heard the key sentence
Jim Turley	E&Y CEO	Afterwards: the accounting was never challenged, 31x would become 32x, we feel good about the work	Institutional defence
David Einhorn	Hedge fund manager	Short from 2007.7, distrusted the published leverage	External pressure / one of the few investors who saw through it
Stephen Richards	Former head of global sales at CA (supplementary reading)	Backdated contracts, seven-year sentence; his handwritten letter spells out the pressure, the grey zone and the board's distance	Soltes's comparison case

Who has decision rights: Fuld and the executive committee set leverage targets and risk appetite. Who puts up the money: short-term creditors and shareholders; the rating agencies are the gatekeepers. Who pushed back: Antoncic, McDade (against treating the drug as balance-sheet reduction), Lee, Einhorn, Jain. Who are the external sources of pressure: analysts, rating agencies, short sellers, and short-term funding markets that became hypersensitive after March. Who should have stopped it but waved it through on form: E&Y, Linklaters, the audit committee (kept in the dark), and the rating agencies and regulators (shown laundered numbers).

Core Tensions

Where the professor's question lies: was this management's window dressing, or an institutional failure in which the audit firm, the law firm, the committee, the rating agencies and the regulators all waved it through? On one side: under one reading of SFAS 140 the accounting could be treated as a sale, Turley says 31 becomes 32, and Lehman would have collapsed anyway. On the other: $50.4B was 28 times the materiality threshold, volumes were ramped up at quarter-end and never disclosed, there was silence to the committee after the whistleblower spoke, and the number the market relied on to judge "low double-digit leverage" was manufactured.
Reported net leverage of 12.1 vs roughly 14.9 without window dressing: was the deleveraging real, or staged for the rating agencies and the earnings calls? Callan / Lowitt told 16.1 → 15.4 → 12 as a story of discipline; Repo 105 grew from $24.5B to $50.4B, and the sticky assets were not sold as McDade demanded.
The Code of Ethics requires reporting and promises no retaliation vs Lee laid off two weeks later. On paper the speak-up system was complete; in practice the outcome was a confidentiality agreement and compensation he couldn't fully collect after the bankruptcy.
Legal in form vs misleading in substance. Linklaters' English-law opinion, the 5% haircut, the quick pass through the UK via LBIE; US law firms refused to write the same sentence. Even Kelly called it legalistic and form-driven.
The auditor "comfortable with GAAP" vs its duty to communicate with the committee. The engagement letter required issues involving senior management to be reported directly to the committee; Cruikshank explicitly asked for every allegation to be reported; after hearing about the $50B, Schlich stayed silent and issued an unqualified review within 28 days.
Richards's grey zone vs Lee's timing. CA recognised contracts two or three days early, and Richards said it was "just timing"; Lehman moved securities off for a few days and moved them back — also timing. Both paths slid from "everyone manages the performance numbers" into practices there was no coming back from.