Skip to content

G6 · Lehman Brothers (2008): thirty-to-one, funded overnight

In January 2008 Lehman Brothers reported its fourth consecutive year of record earnings, a 20.8% return on equity and a share price near $60. Eight months later it filed the largest bankruptcy petition in US history. Nothing about the collapse required secret information to suspect: the leverage, the Level 3 assets, the commercial-property book and the dependence on overnight repo funding were all printed in the 10-K. What was hidden — and what the court-appointed examiner later dug out — was how much worse the true picture was. This case puts you at the desk on 16 June 2008, the day Lehman reported its first quarterly loss as a public company, and asks the question every analyst of a levered balance sheet must answer: is this cheap, or is the book value an opinion?
Period FY2003 – 15-Sep-2008 (aftermath to 2012–2025)
Decision date 16-Jun-2008 — Q2 FY2008 results day, four days after Lehman closed a $6bn equity raise
Sector US investment bank / securities broker-dealer
Themes Gross vs net vs tangible leverage · Level 3 fair values · commercial real estate & Archstone · repo funding and liquidity pools · KPI definition changes · Repo 105 window-dressing · liquidity vs solvency
Modules this case reinforces 04.5 Leverage, solvency & liquidity · 04.3 Returns on capital · 04.7 Quality of earnings · 07.1 Banks & NBFCs · 06.6 Reverse DCF & expectations · 09.1 Why numbers lie · 09.4 Cash-flow games · 05.5 Capital allocation · 03.5 US filings
Difficulty Advanced — do it after M04, lesson 07.1 and M09
Time ~3.5 hours (about 90 minutes on Section 3 before you read on)

Conventions in this case

Lehman's fiscal year ended 30 November, so Q2 FY2008 = March–May 2008. All dollar figures are US\(; "\)bn" = billion. Per-share figures are as reported, i.e. already adjusted for the 2-for-1 stock split of 28-Apr-2006 (10-K FY2007, Item 5). Numbers in square brackets like [1] refer to the Sources list. Figures marked ≈ are rounded or estimated; every ratio not printed by Lehman was computed by us and checked in Python.

1. The scene (as of 16-Jun-2008)

The business. Lehman Brothers Holdings Inc. (NYSE: LEH), founded in 1850 and public since 1994, was the fourth-largest US investment bank. A broker-dealer is a firm that trades securities for clients and for itself, underwrites new issues and finances other investors' positions; Lehman reported three segments — Capital Markets (fixed income and equities trading), Investment Banking (advisory and underwriting) and Investment Management (including Neuberger Berman). FY2007 was a record: net revenues $19.3bn, net income $4.2bn, return on average common equity 20.8%, 50% of revenue from outside the US, $282bn of assets under management and 28,556 employees [4]. The press release boasted of being named Fortune's most admired securities firm [4].

Underneath the P&L was an extraordinary balance sheet: $691bn of assets on $22.5bn of equity at 30-Nov-2007 [1]. Two growth lines stand out. Total assets rose 37.2% in FY2007 while equity rose 17.2%. And "real estate held for sale" — direct property and property-company investments — more than doubled from $9.4bn to $21.9bn [1]. In October 2007 Lehman and Tishman Speyer closed the take-private of Archstone-Smith, then the second-largest listed US apartment REIT, at \(60.75 a share; after an immediate property sale the enterprise value was ≈\)22.2bn, of which only ≈\(5.1bn was equity, and Lehman held ≈\)5.4bn of the capital structure (≈$2.4bn of it equity) at closing [15, pp. 364–367].

How it was funded. A broker-dealer finances most of its inventory with repos (sale-and-repurchase agreements: you "sell" a bond for cash today and agree to buy it back tomorrow or next week at a fixed price — economically a collateralised loan). The lender protects itself with a haircut (it lends, say, $95 against $100 of collateral). At 30-Nov-2007 Lehman had $181.7bn of repos, $53.3bn of securities loaned and $23.0bn of other secured borrowings — $258.0bn of collateralised financing, 37.3% of total assets — plus \(28.1bn of short-term unsecured borrowings [1]. Its stated policy was to fund illiquid positions with "cash capital" (equity plus liabilities with more than a year to run) and to hold a **liquidity pool** at the holding company sized to survive twelve months without issuing unsecured debt; the pool was ≈\)34.9bn at year-end [1].

The industry and the regulators. The five large US investment banks (Goldman Sachs, Morgan Stanley, Merrill Lynch, Lehman, Bear Stearns) had no deposit base to speak of and were supervised at group level only through the SEC's voluntary Consolidated Supervised Entities (CSE) programme [20]. On 14–16 March 2008 Bear Stearns ran out of funding and was sold to JPMorgan with Federal Reserve support; on Sunday 16 March the Fed created the Primary Dealer Credit Facility (PDCF), lending to dealers like Lehman against investment-grade collateral from 17 March [19]. Lehman's stock fell almost 20% the day after Bear's rescue [18], closing at $31.75 on 17-Mar-2008 versus $62.19 on 2-Jan-2008 [14, p. 11].

Management. Richard S. Fuld Jr. was Chairman and CEO. On 12-Jun-2008 Lehman announced that Herbert (Bart) McDade would replace Joseph Gregory as President and COO, and Ian Lowitt would replace Erin Callan as CFO [9].

What the market believed. Two stories competed.

  • Management's: Lehman had reported a profit in every quarter through February 2008, including $489m in Q1 FY2008 in the week of Bear's collapse [8]. It raised $4bn of convertible preferred stock in April and, on 9-Jun-2008, pre-announced a $2.8bn Q2 loss together with a \(6bn raise (\)4bn of common stock, 143m shares at ≈$28, and $2bn of 8.75% mandatory convertible preferred with a $28.00 initial price) [5][6]. By 31 May it had cut gross assets by $147bn in a quarter, lifted the holding-company liquidity pool to $45bn and reported net leverage of 12.0x [8]. Fuld's line on 16 June was that the firm was taking the steps needed to restore its credibility [8].
  • The critics': David Einhorn of Greenlight Capital, short the stock, argued in public speeches on 8-Apr and 21-May-2008 that Lehman's marks were too generous, its Level 3 accounting hard to reconcile, its Archstone stake under-written-down and its leverage — measured on tangible common equity — above 40x [17][18]. Barron's had written in January that on REIT prices "the value of the Archstone equity could be zero" (as quoted in [15, p. 204]), and a March 2008 Portfolio.com article accused Lehman of changing the definition of a key capital measure (quoted in [15, p. 206]).

The price. The stock closed at $27.50 on 10-Jun-2008 [14, p. 11]; new investors bought \(4bn of stock at ≈\)28 [5][6]. We use $28 as the decision price. On 11 June a Bloomberg columnist noted that Lehman's \(19.2bn market capitalisation was ≈\)7bn below its $26bn book value (quoted in [15, p. 205]). In January 2008 the same shares had traded as high as $65.73 [14, p. 2] and at 1.5x book.

Trader's lens: short-dated funding of long-dated, illiquid assets

Lehman ran a gigantic carry trade: it owned assets that paid more than overnight repo costs and earned the spread, levered ~30x. The position is short a put on its own credit. On a normal day you collect carry; on the day counterparties refuse to roll, you must deliver cash you do not have, and the only assets you can sell quickly are the liquid ones — leaving the illiquid, hard-to-mark positions behind. The liquidity pool is your margin buffer; the haircut schedule is your margin model; your credit rating is the trigger that moves both. Ask of any levered lender — Indian NBFCs included — what is the tenor of the funding versus the tenor (and liquidity) of the assets, and who can pull the plug on which day?

2. The numbers then

Everything in this section was public on 16-Jun-2008 except where marked. Note the definitions Lehman used, because they matter in Section 3:

  • Gross leverage = total assets ÷ total stockholders' equity.
  • Net assets (Lehman) = total assets − cash and securities segregated for regulatory purposes − collateralised lending (reverse repos, securities borrowed) − goodwill and intangibles.
  • Tangible equity capital, TEC (Lehman) = stockholders' equity (including preferred) + junior subordinated notes − goodwill and intangibles. Net leverage = net assets ÷ TEC.
  • Tangible common equity, TCE (ours, the conservative measure) = common stockholders' equity − goodwill and intangibles. No preferred, no hybrid debt.
  • Level 3 assets: assets carried at fair value whose valuation relies on unobservable inputs, i.e. on management's models (US GAAP SFAS 157; the Ind AS/IFRS equivalent is the IFRS 13 hierarchy — see 02.9).

Table 2.1 — Five years of growth (FY ends 30-Nov; $bn unless stated) — source [1], Selected Financial Data and cash-flow statement

Item FY03 FY04 FY05 FY06 FY07
Net revenues 8.6 11.6 14.6 17.6 19.3
Net income 1.7 2.4 3.3 4.0 4.2
Diluted EPS ($) 3.17 3.95 5.43 6.81 7.26
Pre-tax margin (%) 29.3 30.4 33.0 33.6 31.2
Compensation ÷ net revenues (%) 49.9 49.5 49.3 49.3 49.3
Return on average common equity (%) 18.2 17.9 21.6 23.4 20.8
Total assets 312.1 357.2 410.1 503.5 691.1
Net assets (Lehman definition) 163.2 175.2 211.4 268.9 373.0
Total stockholders' equity 13.2 14.9 16.8 19.2 22.5
Tangible equity capital (Lehman) 10.7 12.6 15.6 18.6 23.1
Long-term borrowings 35.9 49.4 53.9 81.2 123.2
Gross leverage (x) 23.7 23.9 24.4 26.2 30.7
Net leverage (x) 15.3 13.9 13.6 14.5 16.1
Book value per share ($) 22.09 24.66 28.75 33.87 39.44
Cash flow from operations — — (12.2) (36.4) (45.6)

Growth FY03→FY07 (our calculation): net revenues CAGR 22.2%, net income 25.3%, total assets 22.0%, equity only 14.3%. The gap between asset growth and equity growth is the rise in leverage.

Table 2.2 — The broker-dealer balance sheet, 30-Nov-2007 ($bn) — source [1], Consolidated Statement of Financial Condition and Note 3

Assets $bn Liabilities & equity $bn
Cash & equivalents 7.3 Short-term borrowings (incl. current LTD) 28.1
Cash & securities segregated (regulatory) 12.7 Short positions ("sold, not yet purchased") 149.6
Inventory owned ("financial instruments owned") 313.1 Repos 181.7
— mortgage & asset-backed 89.1 Securities loaned 53.3
— government & agency 40.9 Other secured borrowings 23.0
— corporate debt 54.1 Payables (brokers, customers, accrued) 80.3
— corporate equities 58.5 Bank deposits 29.4
— real estate held for sale 21.9 Long-term borrowings 123.2
— commercial paper 4.0 Total liabilities 668.6
— derivatives (net) 44.6 Preferred stock 1.1
Reverse repos 162.6 Common equity 21.4
Securities borrowed 138.6 Total equity 22.5
Receivables 43.3
PP&E and other 9.3
Goodwill & intangibles 4.1
Total assets 691.1 Total liabilities & equity 691.1

Of the $291.2bn of assets carried at fair value, $42.0bn (13.4% of inventory) was Level 3; real estate held for sale was carried separately at the lower of cost and fair value less costs to sell [1]. Level 3 had nearly tripled during FY2007 (net of derivative liabilities, from $13.6bn to $38.9bn), and $12.5bn of that increase — about half — came from transfers into Level 3 rather than purchases [1].

How to read a broker-dealer balance sheet in six passes

  1. Strip out the matched book. Reverse repos and securities borrowed (\(301.2bn) are loans *to* clients, largely funded by repos and securities loaned (\)235.0bn). They are low-risk but huge — which is why gross leverage overstates risk and why firms invented "net assets".
  2. Net the inventory. $313.1bn long versus $149.6bn short (mostly government bonds and equities). The risk sits in what cannot be hedged or sold quickly: mortgages, property, leveraged loans, private equity.
  3. Read the fair-value hierarchy note. Level 1 (quoted), Level 2 (observable inputs), Level 3 (models). Track the Level 3 roll-forward: purchases, transfers in, and gains/losses recognised on Level 3 positions.
  4. Map the funding by tenor. Overnight/term repo, commercial paper and short-term debt, customer payables (prime brokerage cash can leave in a day), deposits, long-term debt, equity.
  5. Decompose the equity. Common vs preferred vs hybrid debt vs goodwill. Only tangible common equity absorbs losses without anyone's permission.
  6. Look beyond the date. Commitments to lend, derivative collateral triggers on a downgrade, off-balance-sheet vehicles — and whether the quarter-end balance sheet looks like the average one.

Table 2.3 — Five quarters to the decision date ($bn unless stated) — sources [4][8]; Level 3 [1][2]; liquidity pool [1][8]; exposures [8, Ex. 99.2]

Item May-07 (Q2 FY07) Aug-07 (Q3) Nov-07 (Q4) Feb-08 (Q1 FY08) May-08 (Q2 FY08)
Net revenues 5.51 4.31 4.39 3.51 (0.67)
Net income 1.27 0.89 0.89 0.49 (2.77)
Diluted EPS ($) 2.21 1.54 1.54 0.81 (5.14)
Total assets 605.9 659.2 691.1 786.0 639.0
Net assets 337.7 357.1 373.0 396.7 326.9
Total equity (incl. preferred) 21.1 21.7 22.5 24.8 26.3
— of which preferred 1.1 1.1 1.1 3.0 7.0
Tangible equity capital (Lehman) 21.9 22.2 23.1 25.7 27.2
Book value per share ($) 37.15 38.29 39.44 39.45 34.21
Gross leverage (x) 28.7 30.3 30.7 31.7 24.3
Net leverage (x) 15.4 16.1 16.1 15.4 12.0
Level 3 assets — — 42.0 42.5 n/a on 16-Jun ¹
Holding-co liquidity pool — ≈36 ≈35 ≈34 ≈45
Residential mortgage exposure — — 32.1 31.8 24.9
Commercial real estate exposure ² — — 51.7 49.0 39.8
High-yield acquisition finance — — 23.9 17.8 11.5

¹ The Q2 10-Q, filed 10-Jul-2008, later showed $41.3bn [3]. ² Commercial mortgages plus real estate held for sale (net investment). In Q2 FY2008 Lehman took \(4.0bn of gross mark-downs (\)9.3bn gross in H1; $17.0bn gross from FY2007 through May 2008, $9.3bn net of hedges), partly offset by $1.9bn of gains on its own debt as its credit spreads widened [8, Ex. 99.2].

Table 2.4 — Derived ratios (our calculations)

Ratio Nov-07 Feb-08 May-08 May-08 pro forma ³
Tangible common equity, TCE ($bn) 17.3 17.7 15.2 19.2
Total assets ÷ TCE (x) 40.0 44.3 42.1 33.6
Level 3 assets ÷ TCE (x) 2.43 2.40 — 2.22 ⁴
Fall in Level 3 values that wipes out TCE (%) 41.1 41.7 — 45.1 ⁴
Resi + CRE + HY loans ÷ TCE (x) 6.24 5.56 5.02 3.97
Fall in those assets that wipes out TCE (%) 16.0 18.0 19.9 25.2
Hybrids (preferred + junior sub notes) ÷ TEC (%) 25.3 31.0 44.1 —

³ Adding the June raise: +\(4.0bn common equity, +\)6.0bn total equity and cash; 143m new shares. Lehman's own pro forma was 20.0x gross and 10.1x net [3]. ⁴ Using the Feb-08 Level 3 figure, the latest available on 16-Jun.

Profitability and valuation (our calculations from [1][4][8]). FY2007 return on average assets was 0.70% on average leverage of 28.6x (≈20.0% on average total equity); in FY2006 it was 0.88% on 25.4x (≈22.3%). ROA fell; leverage rose to hold ROE up. Interest expense absorbed 67.4% of FY2007 total revenues. At \(28 the stock traded at 0.82x May-08 book value (\)34.21), ≈1.0x tangible book (≈\(27.5 per share before and after the raise), 3.9x FY2007 EPS — and trailing-twelve-month EPS was *negative* at −\)1.25. On 28-Jan-2008 ($60.63) it had traded at 1.54x book, 1.87x tangible book and 8.4x earnings.

Liquidity versus solvency

Solvency asks: at fair value, do assets exceed liabilities? Liquidity asks: can you pay what falls due today, tomorrow and next week? A solvent firm can die of illiquidity (the run comes first); an insolvent firm can stay liquid for months (nobody has noticed). For a dealer the two are coupled: doubts about the marks (solvency) make lenders raise haircuts and pull lines (liquidity), which forces sales at distressed prices that validate the doubts. Your job is to size both the equity cushion and the funding runway — and to notice when one is being used to reassure you about the other.

3. You are the analyst

Answer these using only Sections 1–2 (and, if you like, the three filings and two Einhorn speeches in the Sources, all public by 16-Jun-2008). Write your answers down before opening Section 7.

  1. Leverage three ways. Compute gross leverage, Lehman's net leverage and total assets ÷ TCE at Nov-07, Feb-08 and May-08 (and pro forma for the June raise). Which denominator would you use to judge solvency, and why do the three tell such different stories?
  2. Loss absorption. What percentage fall in the value of Lehman's Level 3 assets would wipe out its TCE at Nov-07? Repeat for the "sticky" book (residential + commercial real estate + high-yield acquisition finance) at May-08, pro forma for the raise. Compare with the cumulative gross write-downs already taken.
  3. Archstone. At closing, equity was ≈\(5.1bn of a ≈\)22.2bn capital structure. Einhorn said comparable REIT prices had fallen 20–30% since the deal was announced. Roughly how much of the equity survives a 20% fall in enterprise value? How big was Lehman's position relative to its TCE?
  4. The moving definition. Until FY2007 Lehman capped preferred stock and junior subordinated notes at 25% of TEC; from FY2008 it dropped the cap (footnote in [8]). Estimate net leverage at Feb-08 and May-08 on the old definition. What do you infer?
  5. Where the ROE came from. Decompose FY2006 and FY2007 ROE into ROA × leverage. Then: what does a 1% pre-tax write-down of Nov-07 total assets do to TCE?
  6. Funding map. Using the Feb-08 balance sheet [2], compare the liquidity pool with short-term secured and unsecured funding, and long-term capital with obviously illiquid assets (Level 3 + real estate held for sale). Which comparison reassures you, which worries you, and what does the 10-K say happens to collateral on a two-notch downgrade?
  7. Price-implied expectations. At $28 and ≈1.0x tangible book, what sustainable return on tangible equity (ROTE) is the market implying if the cost of equity is 12% and long-run growth 4%? Alternatively, if a "clean" Lehman deserved the 1.75x P/TBV implied by an 18% ROTE, what hidden loss is the price discounting?
  8. Earnings quality. (a) FY2007 cash flow from operations was −\(45.6bn against net income of +\)4.2bn. Red flag? (b) Einhorn alleged that Q1 FY2008 Level 3 assets showed a \(228m *gain* in the 10-Q versus an ≈\)875m loss described on the earnings call, a $200m write-down on $6.5bn of newly disclosed CDOs, and a $400–600m gain on an Indian power company (KSK Energy Ventures). How would you test these claims yourself, and what is management's best defence?
  9. Decision. Long, short or avoid at $28 — and with what position size and instrument? Write three things that would change your mind and the three numbers you would monitor weekly.

4. What happened

Table 4.1 — Timeline

Date Event Source
29-May-2007 Lehman and Tishman Speyer agree to buy Archstone-Smith at $60.75/share [15, p. 365]
Aug-2007 Lehman closes its BNC Mortgage subprime origination unit; Q3 FY07 transfers $9.9bn of mortgage assets into Level 3 [4][2]
5-Oct-2007 Archstone deal closes (EV ≈$22.2bn after an immediate property sale) [15, pp. 366–367]
29-Nov-2007 Einhorn presents on risk at the Value Investing Congress [15, p. 204]
Jan-2008 Dividend raised to \(0.68/yr; buyback of up to 100m shares authorised; ≈\)765m spent in Q1 at ≈$59/share [1][2]
16/17-Mar-2008 Bear Stearns rescued; Fed opens the PDCF; LEH closes $31.75 on 17-Mar [19][14]
18-Mar-2008 Q1 FY08 net income $489m [8]
8-Apr-2008 Einhorn's Grant's speech: Lehman's marks and Archstone questioned [17]
21-May-2008 Einhorn's Ira Sohn speech ("Accounting Ingenuity"); LEH closes down $2.44 [18][15, pp. 714–715]
9/12-Jun-2008 Q2 loss of $2.8bn pre-announced; $6bn raise priced and closed [5][6][7]
12-Jun-2008 President/COO and CFO replaced [9]
16-Jun-2008 Decision date: Q2 results; liquidity pool $45bn, net leverage 12.0x [8]
1-Jul-2008 LEH $20.96 [14, p. 11]
7–9-Sep-2008 Fannie Mae and Freddie Mac placed in conservatorship (7-Sep); talks with Korea Development Bank end (9-Sep); LEH $7.79 [15, p. 619][14]
10-Sep-2008 Q3 pre-announced: $3.9bn loss, $5.6bn net marks; plan to spin $25–30bn of CRE into "REI Global", sell a majority of investment management, cut dividend to \(0.05; liquidity pool "\)42bn"; LEH $7.25 [10][14]
12-Sep-2008 LEH closes $3.65; FRBNY convenes Wall Street CEOs over the weekend [14, pp. 11–12]
14-Sep-2008 Barclays deal collapses when UK shareholder-approval requirements are not waived [14, p. 12]
15-Sep-2008 LBHI files for Chapter 11 at 1:45 a.m.; ≥$138bn of LBHI debt accelerated; ≈7,000 derivative master agreements affected; Dow −504 points [11][14, p. 13]
16-Sep-2008 Reserve Primary Fund "breaks the buck" on Lehman paper; AIG rescued [14, pp. 13–14]
19/22-Sep-2008 SIPA liquidation of the US broker-dealer (LBI); sale of US businesses to Barclays for $1.54bn closes 22-Sep [12][21]
26-Sep-2008 SEC ends the voluntary CSE programme ("voluntary regulation does not work") [20]
11-Mar-2010 Examiner Anton Valukas publishes his report, revealing Repo 105 [14][16]
21-Dec-2010 New York Attorney General sues Ernst & Young, alleging it helped Lehman commit accounting fraud; E&Y says it will vigorously defend [23]
6-Mar-2012 Chapter 11 plan effective; all common and preferred shares cancelled; no distributions expected to shareholders [13]
28-Sep-2022 LBI liquidation closes: customers repaid $106bn in full; unsecured creditors of LBI recover ≈41.3% [21]

How the run worked. The examiner's summary is the best short description of the mechanism: Lehman held roughly \(700bn of mostly long-term assets on ≈\)25bn of capital, funded largely short-term; it had to borrow tens or hundreds of billions in repo every day, and the moment counterparties declined to roll, it could not open for business [14, p. 3].

flowchart LR
  A[Doubts about marks<br/>and losses] --> B[Equity and ratings<br/>under pressure]
  B --> C[Repo haircuts rise,<br/>clearing banks demand collateral]
  C --> D[Liquidity pool encumbered<br/>and drained]
  D --> E[Forced sales at<br/>distressed prices]
  E --> A

By June 2008 significant parts of the reported liquidity pool had become hard to monetise — cash left with clearing banks as "comfort" deposits that were later formally pledged — and by 12-Sep-2008, two days after the pool was publicly reported at ≈$41–42bn, it held less than $2bn of readily monetisable assets, according to the examiner [14, pp. 9–10]. The government concluded that it could not inject capital and that Lehman's collateral could not support a loan large enough to prevent collapse [14, p. 12].

Stock-price outcome.

Date LEH close ($) vs $28 decision price vs 2-Jan-2008
2-Jan-2008 62.19 +122.1% —
10-Jun-2008 27.50 (1.8%) (55.8%)
1-Jul-2008 20.96 (25.1%) (66.3%)
9-Sep-2008 7.79 (72.2%) (87.5%)
12-Sep-2008 3.65 (87.0%) (94.1%)
6-Mar-2012 shares cancelled (100%) (100%)

Prices from the examiner's Morningstar series [14]; cancellation from [13]. For a June-2008 buyer the 1-, 3- and 5-year outcome was a total loss (we have not verified any post-bankruptcy over-the-counter trading in the old shares). From the FY2007 high of $86.18 (Dec-2006–Feb-2007 quarter [1]) the fall to $3.65 was 95.8%.

What the examiner found — and did not find. The report's conclusions are findings of colourable claims (enough evidence for a claim to be brought), not court judgments:

  • Repo 105. Lehman treated certain repos, over-collateralised at 105% (bonds) or 108% (equities), as sales rather than financings, used the cash to pay down other liabilities just before quarter-end, and reversed the trades days later. Quarter-end volumes were $38.6bn (Q4 FY07), $49.1bn (Q1 FY08) and $50.38bn (Q2 FY08). Without them, reported net leverage of 16.1x, 15.4x and 12.1x would have been 17.8x, 17.3x and 13.9x [16, pp. 739–748]. No US law firm would give a true-sale opinion; the programme ran through Lehman's London broker-dealer under an English-law opinion from Linklaters [16, p. 740]. The practice was not disclosed to investors, rating agencies, regulators or the board [14, pp. 7–8]. An April 2008 email from McDade, later President, called it "another drug we r on" [16, p. 742]. The examiner found colourable claims of breach of fiduciary duty against Fuld and three CFOs, and of professional malpractice against Ernst & Young [16, p. 750]. Fuld's lawyer said he did not know what the transactions were or how they were accounted for; E&Y said the FY2007 statements were fairly presented under GAAP [22].
  • Valuations. The examiner found insufficient evidence that residential whole loan, RMBS, CDO or derivative valuations were unreasonable in Q2–Q3 FY2008, and insufficient evidence that the commercial mortgage book was unreasonably valued — but sufficient evidence that certain principal real-estate investments and the Archstone bridge equity were unreasonably valued from Q1 FY2008 onward. He found no colourable fiduciary-duty claim over valuations [15, pp. 214–215, 363].

5. Signals: knowable then vs hindsight

Knowable on 16-Jun-2008 (public documents) Only knowable in hindsight
Total assets ÷ TCE of 40x (Nov-07) and 44x (Feb-08); Einhorn published the arithmetic [1][2][17] Repo 105 cut reported net leverage by 1.7–1.9 turns at each quarter-end [16]
Level 3 = 2.4x TCE, half its FY07 growth from transfers in [1] The liquidity pool was partly encumbered from June; < $2bn readily monetisable by 12-Sep [14]
CRE exposure $39.8bn ≈ 2.1x pro forma TCE; Archstone bought at the top with 23% equity [8][15] Examiner: Archstone bridge equity unreasonably valued from Q1 FY08 [15]
TEC definition changed in FY08; hybrids reached 44% of TEC — ≈16x net leverage on the old basis vs 12.0x reported (our estimate) [8] Internal risk limits repeatedly exceeded in 2007–08 [14, p. 4]
Buybacks at ≈$59 in Q1 FY08, then $10bn of emergency capital at lower prices by June [2][5][8] Paulson's private warnings to Fuld (Fuld does not recall them) [14, p. 10]
$258bn of collateralised financing + $28bn short-term debt vs a \(35bn pool (Nov-07); +\)4.6bn collateral on a two-notch downgrade [1] Government's view that it lacked authority and collateral to lend [14, p. 12]; KDB and Barclays outcomes [14][15]
Einhorn's specific, checkable allegations on Level 3, CDOs and KSK [18]; CFO replaced [9] Examiner: a subsidiary, Lehman Commercial Paper Inc., was insolvent or borderline from Sept-2007 [15, p. 215]
Stock at ≈1.0x tangible book: the market already doubted the marks [8] —

The strongest bull case on 16-Jun-2008. It was not stupid:

  • Liquidity had been fixed by policy. After March, the Fed would lend to primary dealers against investment-grade collateral [19]. The Bear-style overnight run was supposed to be off the table. Lehman's pool had grown to $45bn, its long-term debt had a 7.1-year average maturity at Nov-07, and it had already completed its FY2008 unsecured funding plan [1][8].
  • Deleveraging was real. Gross assets fell $147bn in one quarter; residential, commercial and leveraged-loan exposures were down 20–35%; pro forma leverage was 20.0x gross and 10.1x net [3][8].
  • Capital had been raised, \(6bn of it at ≈\)28 by sophisticated investors who read the same filings [5][6].
  • Valuation: 0.84x book, ≈1.0x tangible book and 3.9x FY2007 EPS for a franchise that had earned 20%+ ROE for three years, with a valuable asset-management arm (the September plan claimed a majority sale would add more than $3bn of tangible book value [10]).
  • The marks were hedged: cumulative net write-downs were $9.3bn against $17.0bn gross [8]. And the examiner's later work suggests that the extreme bear view — all the marks were fiction — was too strong [15].

Guarding against hindsight bias

Merrill Lynch, Morgan Stanley and Goldman Sachs ran versions of the same model; the examiner notes no major investment bank survived with that model [14, p. 4], but two of them survived as institutions after converting into bank holding companies in September 2008 [14, p. 4 fn. 11]. A short at $28 was right; a short at $60 in January would have survived a March squeeze; a short in July 2008 had to live through a period when the SEC restricted certain short sales in financial stocks [15, p. 715]. The analysis — thin tangible equity, model-marked assets, overnight funding — was knowable. The timing and the policy response were not. Grade your answers on the former.

6. Lessons

  1. Measure leverage on the equity that absorbs losses, then translate it into a loss threshold. "30x" is abstract; "a 16–20% fall in the property, mortgage and leveraged-loan book wipes out common equity" is a decision. → 04.5 Leverage, solvency & liquidity, 07.1 Banks & NBFCs
  2. Liquidity kills first; map asset liquidity against liability tenor. A liquidity pool is only as good as its unencumbered part, measured against funding that can leave overnight — the same asset–liability mismatch that brought down IL&FS and DHFL. → 04.5, 08.1 Banks & lending, Case I4 IL&FS & DHFL
  3. When assets are Level 3, book value is management's opinion. Test marks against observable proxies (indices, listed peers — Einhorn used REIT prices for Archstone) and watch transfers into Level 3 and gains recognised there. → 04.7 Quality of earnings, 09.3 Expense & asset red flags
  4. When a company redefines a KPI during stress, recompute the old one. The TEC change alone was worth ≈4 turns of net leverage by May 2008. → 09.1 Why numbers lie, 04.6 Ratio dashboards
  5. Quarter-end is a photograph the subject can pose for. Compare period-end with average balances, read the secured-funding notes, and distrust ratios that improve exactly at reporting dates. → 09.4 Cash-flow games, 03.3 Notes to accounts, Case G4 Enron
  6. Capital allocation at the peak is a governance signal. Buying Archstone in 2007, raising the dividend and buying back stock at ≈\(59 in early 2008, then issuing equity at ≈\)28 is value destruction you can see without any hidden information. → 05.5 Management & capital allocation
  7. For financials, "cheap on book" is a hypothesis about the marks. Justified P/B = (ROE − g)/(r − g) only works if both ROE and B are real. → 06.5 Relative valuation, 06.6 Reverse DCF
  8. Treat critics' work as testable hypotheses, and go to primary documents — in any jurisdiction. Einhorn checked Lehman's KSK Energy mark against a draft red herring prospectus filed in India; management fighting short sellers is not a rebuttal. → 03.1 The disclosure universe, 11.6 Behavioural finance

India notes

The Lehman pattern — long, illiquid, hard-to-value assets funded with short-term wholesale money whose providers watch ratings — is the template for Indian NBFC/HFC stress (IL&FS and DHFL in 2018–19, case I4) and for bank failures driven by concentrated lending (Yes Bank, case I5). When you read an Indian lender, find the ALM (maturity-bucket) disclosure, the share of commercial paper and short-term borrowings in funding, undrawn lines, and — under Ind AS — the fair-value hierarchy note for anything marked with unobservable inputs. RBI's liquidity rules for NBFCs were tightened after IL&FS; lesson 08.1 states the current rules with sources (verify there rather than relying on this note). For how repo-style "sales" are treated on derecognition under Ind AS 109 (the IFRS 9 equivalent), see 02.9 and verify against the standard's application guidance before relying on it.

Trader's lens: equity as a deep, model-marked call

With ≈\(645bn of assets and ≈\)19bn of TCE (pro forma), Lehman's common stock was a call option on the balance sheet struck at ≈97% of asset value (Merton's view, lesson 06.7). Its delta to asset values was enormous — a 1% move in assets is a ≈34% move in TCE — and a big slice of the "underlying" was marked to model, like an exotics book marked to your own vol surface. The market was not pricing the call on Lehman's marks; it was pricing it on its own estimate of the marks, plus the jump risk of a funding run.

7. Model answers

Q1 — Leverage three ways | $bn / x | Nov-07 | Feb-08 | May-08 | May-08 pro forma | |:--|--:|--:|--:|--:| | Total assets | 691.1 | 786.0 | 639.0 | 645.0 | | Total equity | 22.5 | 24.8 | 26.3 | 32.3 | | **Gross leverage** | **30.7** | **31.7** | **24.3** | **20.0** | | Net assets ÷ TEC (Lehman) | 373.0 ÷ 23.1 = **16.1** | 396.7 ÷ 25.7 = **15.4** | 326.9 ÷ 27.2 = **12.0** | ≈10.1 | | TCE = common equity − goodwill | 21.4 − 4.1 = 17.3 | 21.8 − 4.1 = 17.7 | 19.3 − 4.1 = 15.2 | 19.2 | | **Total assets ÷ TCE** | **40.0** | **44.3** | **42.1** | **33.6** | (Pro forma: +$6bn of cash to assets, +$6bn to total equity, +$4bn to common equity; Lehman's own pro forma was 20.0x and 10.06x [3].) *Which denominator?* For solvency, TCE: it is the only layer that absorbs losses with no one's consent. Preferred stock and junior subordinated notes absorb losses too, but only after common equity is gone. "Net" assets rightly strip out the matched book, but net leverage also *adds* hybrids to the denominator — so between Nov-07 and May-08, as Lehman issued $5.9bn of preferred, net leverage fell from 16.1x to 12.0x while TA/TCE barely moved (40.0x → 42.1x). The three ratios tell different stories because two measure the size of the balance sheet and one measures the *quality* of the capital behind it. A good answer notices that the headline "improvement" came from the denominator's composition.
Q2 — Loss absorption Nov-07: TCE $17.27bn ÷ Level 3 $41.98bn = **41.1%**. A 41% average mark-down of the Level 3 book (pre-tax, ignoring hedges and tax shields) erases common equity. That sounds like a lot — until you look at the sticky book. May-08 pro forma: residential $24.9bn + CRE $39.8bn + HY acquisition finance $11.5bn = $76.2bn; TCE $19.18bn ÷ $76.2bn = **25.2%**. Before the raise it was 19.9%; at Nov-07 16.0% ($17.27bn ÷ $107.7bn). CRE alone ($39.8bn) is 2.1x pro forma TCE. Context: Lehman had already taken $17.0bn of gross ($9.3bn net) write-downs since FY2007 [8] — 89% of pro forma TCE in gross terms. The residential market was falling; Einhorn cited a ≈10% fall in the AAA CMBS index in one quarter [18]. So "25% more" on a book that was mostly whole loans, mezzanine, equity stakes and land was not a tail scenario. Note also the sign of the tax shield and hedges: they help, but hedges were described by Lehman itself as "ineffective" in Q2 (losses on some hedges exceeded gains on others [8]).
Q3 — Archstone Equity cushion ≈ $5.1bn ÷ $22.2bn ≈ **23%** of the capital structure [15, p. 367]. A 20% fall in enterprise value ($22.2bn × 20% ≈ $4.4bn) leaves ≈$0.7bn of equity — about 13% of the original $5.1bn; a 23% fall wipes it out; at 30% the losses reach the most junior debt. This is crude (it ignores cash flows, debt terms and the time value of a long hold), but it is the right *order of magnitude*: in a take-private at 77% debt, equity is a leveraged option on property values. Lehman's share at closing: ≈$5.4bn of the $22.2bn (24%), including ≈$2.4bn of equity (47% of all equity) [15, p. 367]. Against Nov-07 TCE of $17.3bn, that is ≈31% of TCE in one deal (≈14% in the equity layer alone) — consistent with Einhorn's "more than 20% of tangible common equity" [17]. The lesson is concentration: one late-cycle, highly levered transaction consuming a third of the loss-absorbing capital. (Hindsight: the examiner found the bridge-equity marks unreasonable from Q1 FY2008 [15].)
Q4 — The moving definition Old rule: hybrids count only up to 25% of TEC, so if the cap binds, TEC = (common equity − goodwill) ÷ 0.75. - Feb-08: 17.73 ÷ 0.75 = $23.6bn; net leverage = 396.7 ÷ 23.6 ≈ **16.8x** (reported 15.4x). - May-08: 15.18 ÷ 0.75 = $20.2bn; net leverage = 327.8 ÷ 20.2 ≈ **16.2x** (reported 12.1x in the 10-Q; 12.0x in the release). - Sanity check at Nov-07, when the old rule applied: 17.27 ÷ 0.75 = $23.0bn vs reported $23.1bn — the approximation works. Hybrids were 25.3% of TEC at Nov-07, 31.0% at Feb-08 and 44.1% at May-08. On a like-for-like basis, *net leverage did not fall at all* between November and May; the reported decline came from counting more preferred stock as equity. (Adding back the undisclosed Repo 105 — hindsight — takes May-08 to ≈18.7x on the old definition.) Inference: when a company changes a ratio's definition in the quarter it most needs the ratio to look good, recompute it the old way and ask why. Changing definitions is legal and was disclosed in a footnote; it is still information.
Q5 — Where the ROE came from | | FY2006 | FY2007 | |:--|--:|--:| | Net income ($bn) | 4.01 | 4.19 | | Average total assets ($bn) | 456.8 | 597.3 | | ROA (%) | 0.88 | 0.70 | | Average assets ÷ average equity (x) | 25.4 | 28.6 | | ROA × leverage ≈ ROE on total equity (%) | 22.3 | 20.0 | Lehman earned 70 basis points on assets and multiplied it 28–29 times. ROA was falling; leverage filled the gap. That is the DuPont signature of a business "reaching" for returns ([04.3](../../04-financial-analysis/03-returns-on-capital.md)). A 1% pre-tax write-down of Nov-07 assets = $6.9bn; after tax at the FY2007 effective rate of 30.3% [4], $4.8bn — **1.15 years of record profits, 28% of TCE**. Einhorn's version: at 44x, a 1% fall in asset values costs about 44% of tangible common equity pre-tax [17]. The volatility of 1% of the *asset* side becomes a very large volatility of equity.
Q6 — Funding map Feb-08 [2]: repos $197.1bn + securities loaned $54.8bn + other secured $24.5bn + short-term borrowings $34.5bn = **$311.0bn**, plus $72.8bn of payables to customers (largely prime-brokerage balances that can move). Liquidity pool ≈$34bn = **10.9%** of the secured + short-term total (14.5% at the Q2 pool of $45bn). Reassuring comparison: long-term capital $153.1bn vs Level 3 + real estate held for sale ($42.5bn + $22.6bn = $65.1bn) → **2.35x** coverage. On Lehman's own "cash capital" logic, illiquid assets were term-funded [1]. Worrying comparison: the model assumed that *Level 2* collateral — mortgage securities, corporate bonds, equities — could always be financed in repo at stable haircuts [1]. In a run, haircuts on exactly those assets widen and lenders shorten tenor or walk away; then "liquid" assets need cash capital too, and the pool must cover the gap. The 10-K itself quantified the ratings trigger: ≈$2.4bn of collateral callable at Nov-07, a further $0.1bn on a one-notch downgrade and **$4.6bn on two notches** [1]. And a pool held at the holding company does not help if cash is trapped in regulated subsidiaries or posted to clearing banks. A good answer names the variable that matters most: *the haircut and willingness-to-roll on $250bn+ of secured funding*, not the long-term debt maturity.
Q7 — Price-implied expectations Pro forma TBVPS = $19.18bn ÷ 695.7m shares = **$27.57**; P/TBV at $28 = **1.02x**; P/B = 28 ÷ 33.47 = 0.84x. Justified P/TBV = (ROTE − g) ÷ (r − g). With r = 12%, g = 4% (illustrative), P/TBV = 1.02 ⇒ ROTE = 4% + 1.02 × 8% ≈ **12.2%** — the market was pricing Lehman to earn roughly its cost of equity forever, against a FY2007 return on tangible equity of 25.7% [4] (which would justify ≈2.7x). Or flip it: assume the franchise is worth an 18% ROTE ⇒ justified 1.75x ⇒ "true" TBVPS = 28 ÷ 1.75 = $16.00. Hidden hole = ($27.57 − $16.00) × 695.7m ≈ **$8.0bn after tax**, ≈$11.5bn pre-tax at 30% — ≈15% of the $76.2bn sticky book. | Assumed sustainable ROTE | Justified P/TBV | Implied true TBVPS ($) | Implied pre-tax hole ($bn) | % of sticky book | |:--|--:|--:|--:|--:| | 12% | 1.00 | 28.00 | ≈0 | ≈0 | | 15% | 1.38 | 20.36 | 7.2 | 9.4 | | 18% | 1.75 | 16.00 | 11.5 | 15.1 | | 20% | 2.00 | 14.00 | 13.5 | 17.7 | The price was already discounting a further ≈10–15% write-down of the sticky assets *or* a permanently impaired franchise. That is the expectations-investing move ([06.6](../../06-valuation/06-reverse-dcf-and-expectations.md)): a long at $28 was a bet that losses would be *smaller* than that and funding would hold. Q2 said a 20–25% fall wipes out TCE — so the payoff was skewed: modest upside if the marks were right, total loss if they were not *or* if funding ran.
Q8 — Earnings quality (a) **Not by itself.** For a dealer, buying inventory is an *operating* cash outflow, financed by repo (a financing inflow). CFO was −$12.2bn, −$36.4bn and −$45.6bn in FY2005–07 [1] — the cash-flow signature of a balance sheet growing 37% a year. The CFO/PAT test from [04.4](../../04-financial-analysis/04-working-capital-and-cash-conversion.md) does not port to lenders and dealers; the right tests are asset growth vs equity growth, Level 3 growth, and funding tenor. (b) **How to test Einhorn's claims** (they were allegations by a short seller with a position): - Level 3: open the Q1 10-Q roll-forward [2]. It shows realised +$114m and unrealised +$114m — a **$228m gain** — made up of −$750m on mortgages (≈3% of the $25.0bn opening balance), +$695m unrealised on corporate equities (≈8% in a quarter when the S&P 500 fell) and +$204m on net derivatives. Compare with the ≈$875m write-down described on the call [18]. Lehman told Einhorn the difference reflected re-categorisation between Level 2 and Level 3 [18]. - CDOs: the Q1 10-Q footnote to "other asset-backed securities" describes $6.5bn of CDO interests, ≈25% rated BB+ or lower, written down $200m (≈3%) [2][18]. - KSK: Einhorn compared Lehman's explanation (a new third-party round) with an Indian DRHP filed 12-Feb-2008 that, he said, showed Lehman leading a January restructuring [18]. You would pull the DRHP from SEBI's website and read the capital-structure and pre-IPO sections yourself ([03.5](../../03-reading-filings/05-other-documents.md)). *Management's best defence:* Level 3 moves are shown before hedges held in other levels; transfers change category totals without changing P&L; equity gains were real events; and the firm said it expected further losses in Q2 (which came). *Our verdict on 16-Jun:* the explanations may each be defensible, but in aggregate the numbers moved in the direction that flattered earnings in a quarter when markets fell. That pattern, not any single item, is the signal ([04.7](../../04-financial-analysis/07-quality-of-earnings.md), [09.1](../../09-forensics/01-why-and-how-numbers-lie.md)).
Q9 — Decision and monitoring (rubric) A strong answer: **avoid, or a small short/put position sized for a fat left tail in *your* P&L too.** Reasons: tangible leverage ≈34x even after the raise; a 20–25% fall in the sticky book erases common equity; the price already assumed ROTE ≈ cost of equity, so upside was modest; the funding model depended on confidence and ratings (−$4.6bn on two notches); earnings quality and KPI definitions were deteriorating; management turnover at the top. If you short, size it as if the stock could double on a rescue or a buyer (it had short-squeeze risk and a possible sovereign investor — the KDB talks later became public [15, p. 619]) and prefer defined-risk instruments ([11.4](../../11-process/04-position-sizing-and-portfolio-construction.md), [11.7](../../11-process/07-fundamentals-meets-derivatives.md)). *Would change my mind:* (1) a large, arm's-length sale of CRE or Archstone at or above carrying value; (2) a strategic equity investor at a premium, not a discount; (3) term funding replacing overnight repo for mortgage collateral. *Monitor weekly:* CDS spread and rating-agency outlooks (funding trigger); the stock's P/TBV and any transactions in comparable assets (CMBS indices, REIT prices); disclosed liquidity-pool size *and composition*. A weak answer buys because "it's below book" or shorts because "it's levered" without quantifying either.

8. Discussion questions & extensions

  1. Was Lehman insolvent or illiquid on 12-Sep-2008? The government said its collateral was insufficient for a rescue loan [14]; the examiner found most — not all — of its marks reasonable [15]. Build the argument both ways, and say what evidence would settle it.
  2. Repo 105 was, the examiner said, possibly within the letter of SFAS 140 but materially misleading in effect [16]. Design three disclosure checks (for example, average vs period-end secured funding) that would catch quarter-end window-dressing at an Indian bank or NBFC today. Which notes and regulatory returns would you read?
  3. What would you monitor today in a similar company? Pick an Indian NBFC or a global broker and fill in the Lehman dashboard: TA/TCE, Level 3 ÷ TCE, sticky assets ÷ TCE, funding due within 30 days vs unencumbered liquid assets, collateral on a two-notch downgrade, and any change in KPI definitions in the last eight quarters.
  4. Short sellers as information. Einhorn's speeches were public, specific and checkable. Why did so few investors act on them? Relate your answer to social proof, authority bias and the costs of shorting (11.6).
  5. Extension. Repeat Tables 2.3–2.4 for Bear Stearns (FY ending November 2007) and Goldman Sachs using their 10-Ks, and compare with Case I4 IL&FS & DHFL. What distinguishes the survivors?

Sources

All accessed 21-Sep-2026. EDGAR documents are Lehman Brothers Holdings Inc. filings (CIK 806085, now listed as the LBHI Plan Trust).

  1. Lehman Brothers Holdings Inc., Form 10-K for FY ended 30-Nov-2007 (filed 29-Jan-2008) — Item 5 price ranges, Selected Financial Data, MD&A (liquidity pool, cash capital, ratings triggers), balance sheet, cash-flow statement, Notes 3–4 (inventory, fair-value hierarchy, Level 3 roll-forward). https://www.sec.gov/Archives/edgar/data/806085/000110465908005476/a08-3530_110k.htm
  2. Lehman, Form 10-Q for quarter ended 29-Feb-2008 (filed 9-Apr-2008) — balance sheet, Level 3 table and roll-forward, mortgage exposures, TEC/net leverage, buyback authorisation. https://www.sec.gov/Archives/edgar/data/806085/000110465908023292/a08-10156_110q.htm
  3. Lehman, Form 10-Q for quarter ended 31-May-2008 (filed 10-Jul-2008) — Level 3 at May-08, net leverage 12.06x, pro forma 20.00x/10.06x. https://www.sec.gov/Archives/edgar/data/806085/000110465908045115/a08-18147_110q.htm
  4. Lehman, Q4/FY2007 results press release (8-K Ex. 99.1, 13-Dec-2007). https://www.sec.gov/Archives/edgar/data/806085/000110465907088588/a07-30473_33ex99d1.htm
  5. Lehman, Q2 FY2008 pre-announcement and capital raise (8-K Ex. 99.1, 9-Jun-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908038647/a08-16256_1ex99d1.htm
  6. Lehman, Final term sheet, 8.75% Mandatory Convertible Preferred, Series Q (FWP, 9-Jun-2008) — $28.00 initial price. https://www.sec.gov/Archives/edgar/data/806085/000110465908038691/a08-16256_2fwp.htm
  7. Lehman, Form 8-K on issuance of Series Q preferred (12-Jun-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908039456/a08-16256_38k.htm
  8. Lehman, Q2 FY2008 results (8-K Ex. 99.1 and Ex. 99.2 exposure attachments, 16-Jun-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908039972/a08-15739_14ex99d1.htm and https://www.sec.gov/Archives/edgar/data/806085/000110465908039972/a08-15739_14ex99d2.htm
  9. Lehman, Form 8-K on management changes (event 12-Jun-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908040477/a08-15739_168k.htm
  10. Lehman, Preliminary Q3 FY2008 results and strategic restructuring (8-K Ex. 99.1/99.2, 10-Sep-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908057829/a08-22764_2ex99d1.htm
  11. Lehman, Form 8-K, Item 1.03 Bankruptcy (event 15-Sep-2008; filed 19-Sep-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908059632/a08-22764_48k.htm
  12. Lehman, Form 8-K on the Barclays asset purchase agreement (filed 22-Sep-2008). https://www.sec.gov/Archives/edgar/data/806085/000110465908059841/a08-22764_58k.htm
  13. LBHI, Form 8-K on the Chapter 11 plan's effective date (event 6-Mar-2012) — cancellation of common and preferred stock. https://www.sec.gov/Archives/edgar/data/806085/000090951812000100/mm03-1212_8k.htm
  14. Anton R. Valukas, Report of the Examiner, In re Lehman Brothers Holdings Inc., Vol. 1 (US Bankruptcy Court SDNY, 11-Mar-2010) — introduction and executive summary; stock-price series pp. 2 and 11. Copy hosted by Stanford University: https://web.stanford.edu/~jbulow/lehmandocs/VOLUME%201.pdf (the original Jenner & Block site was not reachable on the access date).
  15. Valukas, Examiner's Report, Vol. 2 (valuation, Archstone, survival efforts; also quotes Barron's 21-Jan-2008, Portfolio.com 20-Mar-2008 and Bloomberg 11-Jun-2008). https://web.stanford.edu/~jbulow/lehmandocs/VOLUME%202.pdf
  16. Valukas, Examiner's Report, Vol. 3 — Section III.A.4: Repo 105, pp. 732–750. https://web.stanford.edu/~jbulow/lehmandocs/VOLUME%203.pdf
  17. David Einhorn, "Private Profits and Socialized Risk", Grant's Spring Investment Conference, 8-Apr-2008. https://foolingsomepeople.com/wp-content/uploads/2015/12/Grants-Conference-04-08-2008.pdf
  18. David Einhorn, "Accounting Ingenuity", Ira W. Sohn Investment Research Conference, 21-May-2008. https://foolingsomepeople.com/wp-content/uploads/2015/12/TCF-2008-Speech.pdf
  19. Board of Governors of the Federal Reserve System, press release, 16-Mar-2008 (Primary Dealer Credit Facility). https://www.federalreserve.gov/newsevents/pressreleases/monetary20080316a.htm
  20. US SEC, Press release 2008-230, "Chairman Cox Announces End of Consolidated Supervised Entities Program", 26-Sep-2008. https://www.sec.gov/newsroom/press-releases/2008-230-chairman-cox-announces-end-consolidated-supervised-entities-program
  21. SIPC, "Lehman Brothers Inc.'s 14-Year Liquidation Successfully Concludes", 28-Sep-2022. https://www.sipc.org/news-and-media/news-releases/20220928
  22. CBS News/AP, "Lehman Cooked Books before Collapse, Report Finds", 12-Mar-2010 — includes the responses of Fuld's lawyer and Ernst & Young. https://www.cbsnews.com/news/lehman-cooked-books-before-collapse-report-finds/
  23. Kevin Reed, "E&Y sued over Lehman's audit", Accountancy Age, 21-Dec-2010 (updated 22-Dec-2010). http://www.accountancyage.com/aa/news/1934026/-sued-lehmans-audit
  24. LBHI, Quarterly Financial Report as of 30-Jun-2025 (8-K Ex. 99.1, 28-Aug-2025) — shows the LBHI Chapter 11 case still open and being administered. https://www.sec.gov/Archives/edgar/data/806085/000119312525191036/d11066dex991.htm

Further reading (not relied on for facts above): David Einhorn, Fooling Some of the People All of the Time (Wiley; later editions add the Lehman story); Financial Crisis Inquiry Commission, Final Report (2011), for the system-wide context.

← Previous: G5 Amazon (1997–2015) · Module index · Next: G7 Nokia & BlackBerry →