Interesting Timeline of Lehman's Collapse (For Dummies)

[Asher]

I got a rather hefty word doc from my firm documenting how and why Lehman failed. The most interesting part of this is the timeline of their collapse, with details I didn't know about earlier.

I can't vouch for its entire accuracy, obviously, but from what I understand it is correct.

• Some German immigrants open up a general store. The economy hits a rough patch and they are paid in cotton. They enter the cotton trade since they had an abundance of the resource and decide that trading in futures is much more profitable than running a general store. The investment firm keeps the same name as the general store, Lehman Brothers. Investment banking moves away from trading and into advisory roles such as M&A, debt syndication and equity offerings. However, trading operations remain a profitable side business.
• 1933 The Glass-Steagall act ensures the separation of Banks and Investment Banks to avoid conflicts of interest. This act was actually passed to help “prevent the next great depression”
• May 1, 1975, fixed commissions for trading securities are abolished. This forces investment banks to find new models since the traditional “tell people what to buy or sell and then charge them for the trade” model quickly loses all value as commissions went down to nothing fairly rapidly.
• Early 1980s, Salomon Brothers’ Capital Markets group shows the rest of Wall Street that a trading desk can generate HUGE amounts of money by selling people products they don’t need and can’t possibly understand. Other Wall Street firms follow suit. For some history here, see Liar’s Poker. Salomon created many of the instruments you hear about on the news now, including the modern MBS and CDO industries.
• 80’s and 90’s - Proprietary desks, trading desks that make ‘educated’ bets on the direction a market will take instead of the traditional market neutral stance of conventional trading desks, gain more power and latitude within the investment banking corporate structure. Given the absurd bonus rules, Prop traders often make tons of money on bonuses.
• 1994, Lehman Brothers is spun off as a public company from its previous parent, American Express. This offers Lehman a new source of capital and lowers its WACC (weighted average cost of capital) that most firms use to determine the feasibility of new projects. This event created the Lehman we knew and loved.
• Late 90’s, early 00’s - Financial Innovation and low interest rates following the burst of the tech bubble allows for increased access to loans and poor lending practices. Credit rating agencies drop the ball when evaluating the quality of the new mortgage backed securities. Everyone thinks “the investments will never go bad because house prices can only go up.”
• In 2004, the SEC provides Lehman and other investment banks with an exemption to quadruple their leverage ratios. This means Lehman can move from 10 times leverage up to as much as 40 times in some cases. In other words, for every hundred dollars that Lehman bets, they are controlling as much as 4000 dollars. When the investment goes bad, Lehman is out 4K instead of just 100.
• 2000’s - Traders at Lehman Brothers rack up large positions in mortgage backed securities. Some of these are due to underwriting activities (Lehman’s fixed income business creates these MBSs for a nominal fee and cannot always sell it off right away, other times traders actually purchase the things to make a profit).
• 2006, executives at well-run investment banks warn senior management of the toxic nature of mortgage backed securities and other complex derivatives. These firms lower their exposure to such assets. JPMorgan’s voice of reason, Jamie Dimon, now serves as the firm’s CEO. He de-leverages JP Morgan while still “enabling” the other investment banks by providing counter party and CDS services. Basically, he bets on the other investment banks failing (with a CDS position) while controlling their money through counter-party and repo trades.
• March 2008, Bear Stearns sells to JPM while on the brink of bankruptcy.
• Merrill Lynch sells toxic mortgage backed securities at 20 cents on the dollar, providing the financing to Loan Star funds (the purchaser, a Texas based private equity group) with clauses that provide downside protection on the sale all the way to 4 cents on the dollar. In other words, Merrill sells the assets at 20 cents and then provides “insurance” in case the value drops lower. This model is no different than companies like Lehman selling a CDO and then providing a CDS in case the underlying asset defaults. Think of it as being self-insured without a cash reserve to cover insurance needs.
• Lehman chairman Dick Fuld gets into fight with notorious short seller David Einhorn blaming short selling for causing fear about capital markets and basically driving companies out of business.
• Early Sept 2008 - Lehman debt placement is cancelled. In other words, Lehman stops trying to borrow more capital because such borrowing requires a degree of financial disclosure that would show how bad of shape Lehman is in. Talks with a South Korean bank over a purchase break down when a price could not be agreed upon with CEO Dick Fuld.
• Early Sept 2008 - The market shows concern that Lehman will post a large quarterly loss. The Merrill Lynch ABS sale causes the market to question Lehman’s own solvency (we are way past liquidity issues at this point). These things can cause a downward spiral. Lehman’s credit rating is tightly coupled with its ability to generate new funds when needed. This is one of the benefits of a public company, you could always issue more stock. If the price of the stock goes down considerably, your ability to generate new cash is crippled. This will increase your likelihood of defaulting on your debt, which will decrease your credit rating. Credit rating agencies Moody’s and S&P are suddenly on the ball. The threat of a decreased credit rating decreases Lehman’s stock further as it would result in higher margin requirements on Lehman’s loans. Since Lehman is leveraged around 30x at this point, this would be devastating. There is no way Lehman could pony up this kind of cash.
• Tuesday Sept 9th - Lehman posts its first quarterly loss since becoming a public company. Dick Fuld (Lehman CEO) identifies his plan to sell the Investment Management Division. Talk proves to be rather cheap; the market is demanding Lehman shore up its books yesterday.
• Sept 10-12 - Market continues to hammer LEH stock.
• Sept 13th - Paulson, Bernanke, and the CEOs of all the major Wall Street firms meet in a downtown NY Fed building to discuss possible resolutions over the September 13th weekend. Group acquisition, single firm acquisition with downside protection from the treasury (similar to the BSC deal), full scale government bailout, and bankruptcy are all evaluated.
• Sept 13th & 14th - Over this same weekend, AIG discloses possible bankruptcy to the fed, removing its focus on Lehman as AIG is a much bigger fish to fry. Bank of America and Barclays are the most likely suitors for a Lehman takeover. Coincidentally, they are also the only interested parties. Barclays does not feel comfortable purchasing Lehman without proper due diligence and without Federal (i.e. “no risk”) downside protection on toxic securities owned by LEH. Barclay’s basically wants a Bear Sterns style deal but the Fed says no. Barclays pulls out of the deal. John Thain, CEO of Merrill Lynch, realizes that his firm could be next. Mr. Thain steps away from the meeting and calls Bank of America directly and proposes that they purchase MER instead of LEH. BofA backs out of the Lehman deal. Keneth D. Lewis, CEO of Bank of America, announces the acquisition of MER on Sunday evening. Robert Diamond, CEO of Barclays, asks Dick Fuld if a deal could be struck after Lehman files for bankruptcy in order for him to avoid purchasing the bad debt on Lehman’s books. He receives a positive response.
• Sept 15th - Monday morning Lehman files for bankruptcy.
• Sept 16th - Tuesday Robert Diamond announces that Barclays will be purchasing Lehman’s North American operations, without any of the toxic assets. The Brits are up in arms because before the end day Friday, Lehman moved $8 Billion out of their London office and into NY. This allowed them to pay the US salaries but not their London ones.
• Sept 17th -The Fed loans AIG $85 Billion and receives warrants that, when executed, will give the Fed a 79.9% stake in the insurance company.
• Sept 18th - Thursday, Secretary of the Treasury Paulson and Ben Bernanke announce a $700 Billion recovery plan to cure the rest of Wall Street’s issues. They also announce they will be fixing Tom Brady’s left knee.
• Sept 18th - A plan to ban short-selling is put in place. Coupled with the socialization of the financial world mentioned in the previous point, capitalism is officially dead. Long live the USSA.
• Sept 22nd - The following week Nomura purchases Lehman’s Asian operation and is in discussions to purchase its European operation.
• Late Sept/Oct – Lehman’s Investment Management Division, centered on Neuberger Berman, is bought by a private equity firm.
• Congress continues to fight with Paulson & Co. over the announced deals. Not a single side is discussing the important questions. Layth is furious.

[Asher]

More "for dummies"

Mortgage Backed Security
A Mortgage Backed Security is a type of asset backed security where the originator (often Fannie Mae, Freddie Mac, Lehman, Citi or one or two other major players) buys a bunch of “whole loans” (aka residential mortgages). The loans are placed in a big pool and payments on the loans (interest, capital, foreclosure recovery, etc) are pooled as the “income” part of the security. The MBS originators sell “shares” of the security to mutual funds, pension funds and other “buyers”.
Over the 00’s roughly 1 to 2 trillion dollars of MBS were created each year. However, an MBS is not as easy to price as other types of securities (such as commercial paper, munis or government bonds). The value of an MBS depends on the aggregate value of the loans. Each loan is valued based on:
• The size
• The interest rate
• The length of the loan
• The value of the property in the case of foreclosure
In addition, there are numerous costs for servicing the loans (actually gathering and tracking the payments), pushing for recovery in the case of default or foreclosure, selling the properties in the case of foreclosure and many others.
The risk on the security is also complicated because it depends on the individual risk of failure of each loan, plus aggregate risks. The risk factors include:
• The chance of default (borrower will simply stop paying)
• The chance of prepayment (borrower will pay early, reducing interest)
• The chance of delinquency (borrower will miss a payment)
• Average interest rate vs current interest rate (if interest rates go down, prepayment goes up as borrowers refinance. If interest rates go up, delinquency and default increase).
• Credit rating of the borrower
• Payment history of the borrower
The Lehman team, for example, had models that included upwards of 20 different variables which gave the risk of whether an individual loan, or pool of loans would fail. However, these models are only as good as the underlying data. Whole Loans were often bought based on just the size, rate and maturity. This created numerous opportunities for fraud and also more cost for “due diligence” in reviewing the loan packages. The feeling was that loan quality really didn’t matter all that much because the MBS market fed the CDO market and the CDO market was super hot. In other words, Lehman would turn over the loans so fast it wouldn’t matter how good or bad they were.
For further reading:
• http://en.wikipedia.org/wiki/Mortgage-backed_security

Collateralized Debt Obligation
Because an MBS can be made up of horrible quality loans, and no one would buy the MBS, banks came up with the idea of slicing up an MBS into “tranches”. Each tranche can be thought of as a bucket. As Lehman (or the MBS holder) receives money (interest, principal, whatever), they throw the money into the buckets in order of seniority. The first tranche gets all the money until the bucket is full. Then the second tranche gets money. The third tranche gets whatever is left. The belief was that, regardless of the underlying loan quality, enough money would come in to always fill the first tranche, and almost always fill the second tranche. The third tranche was riskiest, but was also priced accordingly (it was cheap and had a high interest rate).
A CDO basically let the rating agency give high ratings to part of an MBS even if the underlying assets were junk. The CDO also created nightmares of level 3 (hard to price or sell) assets for companies like Lehman who often simply kept the 3rd tranche.
For further reading:
• See if you can sneak into an Alvin Hall lecture. He explains this amazingly well.
• http://en.wikipedia.org/wiki/Collate...ebt_obligation

Credit Default Swap

A Credit Default Swap is a contract between two parties that can be thought of as insurance. The term insurance, however, has to be avoided because insurance is regulated whereas a CDS is a private contract and thus, unregulated. The seller of a CDS provides “insurance” against credit related default events on some particular object (security, company, whatever). The buyer pays the seller a premium and then only gets payment if a credit event occurs on the CDS object.

Take, for example, the numerous CDS contracts issued against Lehman. JP Morgan sold CDS protection to the Lehman customers who bought Lehman MBS and CDO products. If Lehman defaulted on payment, or went bankrupt, JP Morgan had to pay out the value of the CDS. Of course, a CDS is often much more complicated than this and can involve stock, debt or other considerations.
Most of the CDS’ that were called in over the fall were tied to either CDO or MBS products, or to the companies that issued the products. Imagine this chain of events, and you can see the leverage factor at play:
• Lehman issues an MBS for 10 Million.
• Lehman turns the MBS into a CDO and sells the first tranche to say… AIG. The second tranche goes to…call it “failed Euro Bank”.
• AIG buys a CDS from JP Morgan (counterparty to $58 trillion in swaps!) to protect against the CDO going bad.
• Failed Euro Bank buys a CDS from Citi (counterparty to $38 trillion in swaps!) to protect their part of the CDO.
• Hedge Fund “TBNL” buys a CDS against Lehman going bankrupt from BofA (counterparty to ANOTHER $38 trillion in swaps!).
We now have seven companies (Lehman, AIG, JP Morgan, Citi, Euro Bank, TBNL, BofA) all tied to the same underlying asset. Note that some of these companies (JP Morgan, Citi, BofA) are required to pay massive amounts if the MBS fails. AIG and Failed Euro Bank are also potentially in trouble, especially if JP Morgan, Citi or BofA can’t pay. Hedge Fund TBNL helps cause the problem because Hedge Fund TBNL has no exposure to the underlying asset and is simply betting that it will fail. Add a few dozen Hedge Funds doing the same thing and you suddenly have companies promising to pay 30 or 40 times (or more) of the underlying asset’s value if a credit event occurs.
Speculation exists that the government deal around Bear Sterns was actually to keep JP Morgan from having to pay out many trillions of dollars in CDS payments.
For further reading:
• http://en.wikipedia.org/wiki/Credit_default_swap
• http://asecondhandconjecture.com/ind...nd-nightmares/
• http://nickgogerty.typepad.com/desig...ment-risk.html

[Asher]

Naked Short Selling
Normal short selling requires that the seller of the shares has access to the actual shares before placing the sell order. Usually, the seller:
1. Borrows shares from a prime broker or investment institution
2. Pays a fee for the borrowing privileges
3. Sells the shares
4. Buys new shares to replace the borrowed shares.
In the case of a naked short sell, the seller skips steps 1 and 2 and simply places a sell order for shares he doesn’t own or have access to in the hope that he can find the shares somewhere in the market before the trade actually closes. The seller is “naked” because he cannot cover his obligations to produce the shares. If the seller manages to find shares, the deal goes through. If the seller cannot find shares, the deal is broken (“fail to deliver”) and the seller may have to pay fines. In normal practice, naked short selling just doesn’t happen except in the case of market makers who may need to resort to a naked short temporarily to set the price and bring liquidity to a market.
Naked short selling is a factor in the current situation because it applies significant downward pressure to the price of the stock in question. Even if the deal is broken on a “fail to deliver”, the sale has already been registered on the exchange and the downward pressure has already been applied. Various articles estimate (yet again… no hard numbers!) that sell orders for certain stocks may have been several times larger than the total available stock.
For further reading:
• http:// faculty.chicagogsb.edu/christopher.culp/research/pubs/NakedShorting.pdf

Counterparty
A counter party is simply “the other part of a deal”. Lehman was a counter party to a vast number of MBS, CDO and CDS deals.

Overnight Repo
A repurchase, or repo, deal involves “trading” assets for cash for short term (usually overnight). For most purposes, think of Repo as simply a short term loan. The asset holder needs cash to cover the margin on a trade. Therefore, the asset holder gets a loan from the counterparty by “selling” the asset with the contracted expectation that the counterparty will sell back the same asset at a fixed date for the same price (minus the repo fee or “repo rate”). The asset holder gets a temporary infusion of cash (usually to meet regulatory requirements). The counterparty gets any value from the asset and the repo rate. Lehman engaged heavily in the repo trade using the pool of whole loans that fed into MBS and CDO deals. If the loan wasn’t actively tied to the deal, Lehman would perform a repo trade on it to get cash to perform other trades.

Cost of Capital
Companies like Lehman run on capital. The more capital you have, the more deals you can enter. The cost of capital is how much of a fee you pay to “borrow” large amounts of capital.

Asset Tiers
Many reports discuss the difference between Level 1, Level 2 and Level 3 assets. In short, the level of the asset is determined by how you value the asset:
• Level 1 – Any asset that can be priced based on an active market in the same asset is a Level 1 asset. In other words, if you can sell it or buy it right now, it’s probably a Level 1 asset. Most stocks and government bonds are Level 1 assets. These are easily tradable assets where everyone can agree on a price range.
• Level 2 – Assets that are priced based on inactive markets, or models with real inputs (e.g. LIBOR) are classified as Level 2. These assets can’t be bought or sold as easily as Level 1 assets because the markets aren’t sufficiently liquid or active, or the commodity is just too rare. Level 2 assets include corporate and muni bonds (that trade infrequently), mortgage assets (pricing depends on default rates, interest rate, etc), and more complicated types of derivatives such as interest rate and currency swaps. These assets are more subject to “valuation gaming” than Level 1 assets, but they still are valued based on externally observable numbers.
• Level 3 – Assets that have no market, or are priced on arbitrary internal models are considered Level 3 assets. Assets such as exotic derivatives, structured products, log dated options and many types of mortgage products are Level 3 assets because they don’t trade regularly and the pricing depends entirely on complex models of future events.
As you can imagine, Level 1 assets are regulated, transparent and easy to understand. Level 2 assets are also likely to be regulated and reasonably transparent. However, pricing Level 2 assets is slightly more difficult and the values are subject to greater uncertainty. Level 3 assets are the least transparent and most difficult to price. All of the recent banking failures were related to write offs of Level 3 asset values.
At the end of Q2 2007, Lehman took a $700 million write down on Level 3 assets of $22 billion. Does this mean that Lehman took a loss of $700 million? The answer is that no one really knows because the models used to price Level 3 assets aren’t based on any verifiable market reality. One day, the bank may decide that the value of Level 3 assets is $22 billion. The next day, with a simple change in the model, the bank may say the value is $35 billion. If the assumptions of the model fail (as has happened), the value may suddenly drop to $2 billion.
As a side note, most traders and corporate executives were bonused based on the initial mark-to-model value of Level 3 assets. Since that value is rather arbitrary, the bonuses were large even though the company hadn’t seen the actual income off the investment.

Loan to Value
The Loan to Value ratio is simply the amount of the loan divided by the value of the collateral. In the case of a mortgage, the LTV is the loan amount divided by the estimate of the house value. Since the house value is simply an estimate (and changes over time), the astute observer will see LTV as a type of Mark to Model where the model is based on the best guess of the current house price. LTV plays into the current situation in two major ways, both related to the value part of the equation.
In the first case, many mortgages were originated with bogus valuations. Corrupt lenders and assessors would place absurd values on property. As soon as the loan paperwork was inked, the loan was sold to an investment bank, packed up as an MBS and was therefore gifted with a high rating. The investment banks were often late, lazy or unable to perform sufficient due diligence. Besides, one bad apple in an MBS is insignificant, whereas the cost of re-assessing all the properties was exorbitant and delayed the loan purchase and packaging. This attitude led to larger than expected defaults and smaller than expected recovery values. It’s hard to recover the value of a house when there never was a house on the lot to begin with.
In the second case, the value is based on “comps” or comparative sales of houses in the neighborhood. If the neighborhood had high turnover, then plenty of comps were available and the prices generally went up. In a stable neighborhood, comps are rare so assessors would have to make a guess as to what the house might be worth. In most cases, the guess was settled at or slightly above the purchase price for the mortgage. Lenders were more concerned with getting the deal signed than any kind of fair valuation, so any valuation that made the deal work was acceptable.

Mark to Market and Mark to Model
One of the largest challenges in the financial services world is determining how much something is worth. Asset valuation is critical for every aspect of the business from the front office (deciding what to buy or sell or create), to the back office (controlling capital reserves). Almost every aspect of a financial services company is tied into asset valuation, including risk management, regulatory compliance, capital management, and employee compensation. Over the years, two main methods have emerged as primary accounting methods:
• Mark to Market – The most common valuation method involves comparing the book value of an asset to the current price on the open market. Basically, this is the “fire sale” method of accounting and says your assets are only worth what you could get if you sold them right now. This method tends to have more of an “honest” feel, but is also subject to all sorts of arguments. For example, selling a loan right now requires you sell it for a discount on the total value you would receive if you held onto the loan.
• Mark to Model – The most common alternative to MtM (Mark to Market) is Mark to Model. This method assumes that the price is not readily available (e.g. the asset is Level 2 or 3). Therefore, the company makes a model that predicts asset value behavior into the future. For example, a tranche of an MBS can be priced based on assumptions about interest rate, foreclosure rate, prepayment rate, default rate and recovery rate. None of these values can be known ahead of time, so the company creates a model of what the value might be like over the life of the asset.
Mark to Model gets a big portion of the blame for the current crisis. At the time most of the “toxic assets” were created, the models showed huge profits. The executives, traders, middle office and back office people were all paid enormous bonuses based on the high profit model. As the assumptions changed, the models collapsed, resulting in massive devaluations. This has lead most pundits to describe Mark to Model in a variety of derogatory terms such as Wishcasting, Mark to Madness, Mark to Myth, and Mark to Fantasy.

[Chemical Ollie]

Words, words, words, NAKED, hu-hu, hu-hu, words, words, ASS-et, hu-hu, hu-hu, words, words, words, ASS-ume, words, words...

Hey Beavis, what is this Bull****, there is no timeline, just some stupid porn story?

[Asher]

People wonder why Europe lost its grip on the world...

[Arrian]

Thanks for that stuff, Asher.

-Arrian

[Chemical Ollie]

Originally posted by Asher
People wonder why Europe lost its grip on the world...

Canada still has a European Queen

[Asher]

Originally posted by Chemical Ollie


Canada still has a European Queen

Starchild is an exception.

[Nikolai]

So where's your name on that list Asher?

[Asher]

Originally posted by Nikolai
So where's your name on that list Asher?

I was on the 12-man team that wrote the software for Lehman to trade CDS and MBS trades.

[Nikolai]

I won't pretend I understand what that really means.:P Does it mean you made a computer program/frame work that managed/kept overview of how the values was secured in the firm? If I understood it right from your post CDS and MBS is two different ways to insure the values, right? Or am I completely in the wrong direction?

[Jon Miller]

Originally posted by Asher

I was on the 12-man team that wrote the software for Lehman to trade CDS and MBS trades.

Weren't you trying to get that done right before they started having business problems?

JM

[Asher]

Originally posted by Jon Miller


Weren't you trying to get that done right before they started having business problems?

JM

The system's been around for 10 years. It's just constantly in flux.

[Asher]

Originally posted by Nikolai
I won't pretend I understand what that really means.:P Does it mean you made a computer program/frame work that managed/kept overview of how the values was secured in the firm? If I understood it right from your post CDS and MBS is two different ways to insure the values, right? Or am I completely in the wrong direction?

There's literally dozens of instruments Lehman and other banks use, CDS and MBS trades are just two. I was working on the application that the traders used inside Lehman (it was also used by other banks, like Deutsche Bank, under contract). They had a frontend they used to display all of the information associated with the deal (which is a lot, our database was many terabytes in size). The meat of the program was in the server-side processing. The volume of traffic our servers saw was incredible, and then it had to interact with dozens of risk systems (compute clusters that analyze risk factors in each trade, and accept/reject the trades). Then it had to flow from desk to desk for approvals inside Lehman. Finally the trades had to get pushed outside Lehamn to the counterparties and government clearing houses.

It was very complicated and complex program with many other systems it communicated with.

Essentially, any time a Lehman trader needed to make a trade, they loaded up our program and actioned the trade through the program. They're all "paperless" trades. It was called a "trade capture and workflow system".

[-Jrabbit]

Excellent and informative posts, Asher.

I'm avoiding work and actually read the first three osts. I feel semi-informed now.