The Lehman Brothers Collapse: The Weekend That Broke 2008

At 6 p.m. on Friday 12 September 2008, black cars lined Liberty Street outside the Federal Reserve Bank of New York. Treasury Secretary Henry Paulson, New York Fed President Timothy Geithner, and SEC Chair Christopher Cox had summoned the chief executives of every major Wall Street house — Morgan Stanley, Goldman Sachs, Merrill Lynch, JPMorgan, Citigroup, Credit Suisse, UBS — to a war room on the thirteenth floor. The agenda fit on a single page. Lehman Brothers had run out of cash. A private-sector rescue needed to be assembled by Sunday night, because on Monday morning the firm would either be sold or it would file for bankruptcy. One CEO, arriving late, asked a Fed staffer how bad it was. "We have," the staffer replied, "until Asia opens."

Seventy-nine hours later, at 1:45 a.m. on Monday 15 September, Lehman Brothers Holdings Inc. filed a Chapter 11 petition at the United States Bankruptcy Court for the Southern District of New York. The schedules listed $639 billion in assets and $613 billion in debt — the largest corporate bankruptcy in American history by an order of magnitude, four times larger than WorldCom in 2002. Dick Fuld, who had run the firm for fourteen years, stayed in his thirty-first-floor office on Seventh Avenue until near dawn, signing documents and watching the stock ticker roll to zero. Lehman's equity had closed at $3.65 on Friday; the common was worth $0.21 by Monday's open and untradeable by Tuesday.

The failure of a 158-year-old investment bank would have been a singular event on its own. The cascade that followed — an AIG rescue on Tuesday, a money-market fund breaking the buck on Wednesday, a global commercial-paper freeze by Friday, TARP passed by Congress within three weeks — turned Lehman from a corporate bankruptcy into a historical marker. The 2008 financial crisis existed before 15 September 2008, but the weekend made it unignorable.

From Alabama cotton to global broker-dealer

Lehman Brothers began as a dry-goods store in Montgomery, Alabama, opened in 1844 by the Bavarian immigrant Henry Lehman. His brothers Emanuel and Mayer joined within six years, and by 1850 the partnership was buying and selling cotton — Alabama farmers paid for groceries in bolls, and the Lehmans moved the crop to Liverpool and New York. The firm relocated to Manhattan in 1858, pivoted into commodity brokerage after the Civil War, and helped found the New York Cotton Exchange in 1870 and the Coffee Exchange in 1882. Lehman Brothers joined the New York Stock Exchange as a member firm in 1887.

For most of the twentieth century Lehman was a conservative, mid-sized underwriter — a distant competitor to Morgan Stanley and Goldman Sachs rather than a rival. That changed in 1984 when American Express bought the firm, merged it with Shearson, and ran it as an embedded brokerage for a decade before spinning it back out as an independent public company in 1994. Richard Fuld Jr., a former commercial paper trader, became CEO in 1994 and remained in the chair for the next fourteen years. Under Fuld, Lehman's balance sheet grew from roughly $75 billion to just over $700 billion, revenue quadrupled, and the firm pushed aggressively into residential and commercial mortgage origination, private equity, and proprietary trading (Sorkin, 2009).

The commercial real estate pile

Lehman's exposure on 15 September 2008 was not principally subprime residential mortgages — that was where the firm's origination pipeline ran, but securitisation kept the warehouse flowing. The problem was commercial real estate equity. Between 2005 and early 2008 Lehman's Global Real Estate Group, led by Mark Walsh, put roughly $60 billion of the firm's balance sheet into commercial property deals, including a 50 percent interest in the $22 billion Archstone-Smith apartment REIT acquisition (closed October 2007, at the market peak), a partnership with homebuilder SunCal on California land, and the Real Estate Trust of Bahrain portfolio. When commercial property values rolled over in 2008, these positions lost 20 to 40 percent of book value with almost no offsetting liquidity.

(Sources: Valukas, 2010; Lehman Chapter 11 schedules; Fed FCIC testimony)

Lehman funded this illiquid asset base with $200 billion of overnight tri-party repo, and the repo book was rolled every morning by 10:30 against collateral posted to JPMorgan as clearing bank. The same dynamic that had killed Bear Stearns in March 2008 — overnight funding against declining-quality collateral — was present at Lehman at six times the scale. The difference was that Bear had imploded in 72 hours; Lehman died over six months of attrition.

Repo 105 and the leverage illusion

In the spring of 2008, as short-sellers and analysts began parsing Lehman's balance sheet line by line, the firm adopted an accounting technique it called Repo 105. At quarter-end the balance-sheet date, Lehman entered repurchase agreements in which it sold fixed-income securities to European counterparties with a contractual obligation to buy them back several days later, paying a 5 percent haircut. Because the haircut exceeded the 2 percent threshold in SFAS 140, Lehman booked the transaction as a true sale rather than a collateralised borrowing, removing the securities from assets and the corresponding cash liability from debt. The Valukas Report, the 2,200-page examiner's report filed by Anton Valukas of Jenner & Block on 11 March 2010, documented that Lehman used Repo 105 to move $38.6 billion off balance sheet at the end of Q4 2007, $49.1 billion at the end of Q1 2008, and $50.4 billion at the end of Q2 2008, reducing reported net leverage from roughly 17.3x to roughly 15.4x in Q2 (Valukas, 2010).

Ernst & Young, Lehman's auditor, had signed off on the treatment on the basis of a legal opinion from the London firm Linklaters — no US law firm was willing to provide one. The Valukas Report concluded there was "sufficient credible evidence" to support colourable claims against CEO Fuld, CFO Erin Callan, and others for breach of fiduciary duty for certifying financial statements that the report described as "materially misleading." No criminal charges were ever filed.

Bear Stearns set the expectation

By March 2008 the market had already taken a dress rehearsal. The Bear Stearns collapse and the Fed-backed JPMorgan rescue ended with JPMorgan paying $10 a share for a firm that had been worth $170 twelve months earlier, with $29 billion of Bear's least-liquid mortgage assets ring-fenced in a Fed-financed vehicle called Maiden Lane I. That transaction — the first use of Section 13(3) of the Federal Reserve Act since 1936 — communicated something subtle and damaging. It told every Lehman counterparty that the Federal Reserve would not let a large broker-dealer fail, and it told every Lehman executive that there was an implicit floor under the firm's debt.

Paulson later wrote that he was painfully aware of the moral hazard the Bear deal had created. "In the weeks after Bear Stearns," he noted, "people began to say 'they'll never let Lehman go' — and I knew that was the worst possible outcome" (Paulson, 2010). Geithner spent the summer of 2008 trying to persuade Fuld to raise equity at any price. Three times Fuld came close — a Korean consortium, a Warren Buffett preferred-stock deal, a Chinese sovereign wealth approach — and three times he walked away at the price.

Korea Development Bank walks

The clearest chance came from Seoul. Through the summer of 2008, Lehman negotiated with Korea Development Bank, the state-owned lender, for an equity injection that would have been large enough to stabilise the firm. KDB's chairman Min Euoo-sung, a former Lehman banker himself, had agreed in principle to inject $6 billion in return for roughly 25 percent of the firm, pricing Lehman's equity at around $18 a share. On 9 September the Korean financial regulator publicly signalled discomfort with the valuation. Lehman's stock fell 45 percent in one session. By Wednesday 10 September the talks were dead and Fuld pre-announced a $3.9 billion third-quarter loss — its largest ever — in a hurriedly assembled earnings call.

The weekend at the New York Fed

Paulson, Bernanke, and Geithner had a single operating theory going into the weekend of 12–14 September: Lehman was too interconnected to default on a Monday morning without a private sector buyer in place, and too politically toxic to rescue with public money. The model was the 1998 Long-Term Capital Management consortium — fourteen banks, each putting up $250 million, no Fed balance sheet. Paulson insisted at the Friday meeting that "there would be no government money in this deal," a line Geithner later described in his memoir as "the position we needed the room to hear" (Geithner, 2014).

Two potential acquirers materialised. Bank of America had been in diligence on Lehman for a week and had the scale to absorb it. Barclays, the British bank, had commercial ambition in US investment banking and a CEO — Bob Diamond — who wanted the deal. By Saturday lunchtime both had concluded that Lehman's commercial real estate book contained a $30 to $70 billion hole that no acquirer could swallow without a government guarantee. At which point the weekend began to move in two directions at once.

Bank of America pulled out of Lehman on Saturday afternoon and bought Merrill Lynch the same day for $50 billion in stock — a deal that foreclosed Merrill's own weekend crisis and removed the second-largest broker-dealer from the pipeline. That left Barclays. Through Saturday night the Barclays team, working out of Simpson Thacher, engineered a structure in which a Wall Street consortium would carve off Lehman's most toxic real estate assets into a bad bank while Barclays bought the rest. The structure was essentially workable.

It died on Sunday morning, London time, when the UK Financial Services Authority informed the Treasury that Barclays could not legally guarantee Lehman's trading obligations between signing and closing without a vote of Barclays shareholders under UK listing rules. Hector Sants, the FSA chief executive, relayed the message through Callum McCarthy, the FSA chairman. The Bank of England, led by Mervyn King, declined to waive the requirement. Paulson later described the phone call from London as the moment he realised no rescue was coming: "The British screwed us," he told his staff, a line he later repeated in his memoir (Paulson, 2010).

The authority question

Paulson's account — no political will, no private buyer, therefore bankruptcy — became the public narrative on 15 September. At his Monday afternoon press conference he said that "I never once considered that it was appropriate to put taxpayer money on the line in resolving Lehman Brothers." Bernanke's account is different. In his memoir and in testimony before the Financial Crisis Inquiry Commission, Bernanke argued that the Federal Reserve had no legal authority to lend to Lehman because Section 13(3) required "sufficient collateral" and Lehman's holding company, where the borrowing would have had to occur, lacked the pledgeable assets needed to secure a loan large enough to stabilise the firm (Bernanke, 2015).

These two framings have never been reconciled. If Paulson is right and the constraint was political, then the Fed could have lent but chose not to. If Bernanke is right and the constraint was legal, then Paulson's political explanation was a cover story for a decision that had already been forced by statute. The FCIC majority report concluded that the authority question was "unresolved on the face of the record," and the dissenting view — written by Peter Wallison — argued the opposite: that the Fed did in fact have the authority and the decision was political. The historiographical split has persisted for fifteen years.

The week after

What followed the filing was not a recession, which Lehman's failure did not cause on its own, but a global funding freeze that rolled through every corner of the short-term credit market. On Tuesday 16 September the Federal Reserve extended an $85 billion emergency loan to AIG — which had written $441 billion of credit default swap protection, much of it to European banks needing regulatory capital relief on their holdings of mortgage-related securities. On Wednesday 17 September the Reserve Primary Fund, a $62 billion money-market fund, announced that its net asset value had fallen to $0.97 after writing off $785 million in Lehman commercial paper. The fund had "broken the buck" for the first time in the industry's modern history, and redemption requests of $40 billion hit the desk in a single day.

Money-market funds, which held roughly $3.5 trillion of American corporate short-term debt, saw $172 billion of net redemptions across the next week. Corporate commercial paper rates for General Electric, IBM, and Johnson & Johnson — the highest-rated issuers on earth — doubled overnight. By Friday 19 September the Treasury had announced a temporary guarantee programme for money-market fund balances, and Paulson had begun drafting what would become the Troubled Asset Relief Program. TARP became law on 3 October 2008, seventeen days after the Lehman filing. What the Fed and Treasury had refused to do on the weekend of 13–14 September — write a government check to stabilise the financial system — they committed $700 billion to do three weeks later.

Lehman's European and Asian operations were sold within the week. Nomura Holdings bought the European investment banking and equities business on 22 September and the Asian business two days later, together for roughly $225 million plus retention payments of about $1 billion. Barclays bought the North American broker-dealer out of bankruptcy on 22 September for $1.35 billion plus the assumption of certain liabilities — the same business it could not buy intact five days earlier. The transfer of 9,000 employees, customer accounts, and trading positions happened over a single weekend at 745 Seventh Avenue, with Barclays staff moving in as Lehman staff moved out.

The post-mortem

The Valukas Report took 18 months and 70 attorneys to produce. It ran to 2,209 pages, cost the bankruptcy estate $38 million, and examined every major decision Lehman's senior management made from 2007 through the filing. The report's conclusions on Repo 105, Fuld's knowledge of the leverage position, and CFO Erin Callan's public representations have defined the historical record. Neither Fuld nor Callan was criminally charged. Fuld testified before Congress in October 2008 and again in September 2010, maintaining in both appearances that Lehman had been solvent when it filed and that the firm's demise was the result of a short-seller-driven panic rather than insolvency. The trustee's final report in 2012 put the recovery rate for general unsecured creditors at approximately 28 cents on the dollar; by the time the last distribution closed in 2020 the figure had climbed to roughly 41 cents.

The legislative response was the Dodd-Frank Wall Street Reform and Consumer Protection Act, signed 21 July 2010, whose Title II established the Orderly Liquidation Authority — a bankruptcy regime for systemically important financial institutions that was drafted, almost line by line, to prevent a future Lehman weekend from having to happen the same way. Title II gives the FDIC the power to place a failing bank holding company into receivership, transfer its operations to a bridge institution over a single weekend, and wind down the holding company without a Chapter 11 filing. It has never been used. Whether it would work in practice remains one of the open questions of post-crisis financial regulation.

The Lehman moment

Fifteen years on, the failure is often invoked as a benchmark — "the Lehman moment" — the instant a slow-moving crisis becomes a liquidity avalanche. The phrase carries a specific connotation. It refers to the second in which counterparty trust collapses, not the prior months of balance-sheet deterioration, not the failed rescue talks, not even the bankruptcy filing. It is the moment at which a bank's liabilities stop being accepted as collateral, and every other bank in the system recalculates who might be next. Something similar had happened in Britain a year earlier during the Northern Rock bank run, but on a smaller scale and with a different kind of depositor.

What made Lehman different was the global scale of the derivatives book and the timing. Thirty-five trillion dollars of notional derivatives had to be unwound through the ISDA protocol over the following weeks. German municipal banks, Japanese regional insurers, and Korean pension funds found themselves holding structured notes whose issuer had ceased to exist. In Hong Kong, retail investors who had bought Lehman-issued minibonds through local distributors protested for months outside bank branches. The Icelandic banking collapse that followed in October was mechanically a separate event, but it was the Lehman weekend that had told every international creditor to pull every line it could pull.

Fuld left his office at 745 Seventh Avenue for the last time on 21 September 2008. The sign was taken down that week. Nomura's name went up within the month in London and Tokyo. In New York, the Barclays logo replaced Lehman's on the trading floor by the end of September, on the same glass that had carried the Lehman name since 1994. The firm that had begun in an Alabama dry-goods store in 1844 had lasted 164 years as a business and 114 years as a member of the New York Stock Exchange. It had survived the Civil War, the Panic of 1907, the 1929 Crash, the Great Depression, and the 1994 spin-off. It did not survive the weekend of 13 September 2008.