Reader Contribution: A Worm’s-Eye View from Wall Street
by The Editors
The Conversation
December 16th, 2008
[The following is an unusually detailed and polished reader contribution emailed to the editors from a Wall Street professional, who has asked to remain anonymous. We believe the essay offers a credible, informed "in-the-trenches" perspective on "what happened." -- The Editors]
A Worm’s-Eye View from Wall Street
by Anonymous
—
It is a privilege to read the findings of the wise as to the grand causes of the recent financial extinction event. I wouldn’t dare disagree with any of these eminences’ theories. But, to me, the theories seem far removed from the world of the individuals whose short-term business problems and practices I believe led to the asteroid impact.
Brad DeLong writes: “Our current financial crisis remains largely a mystery…” From my worm’s-eye view, it seems less enigmatic. We’ve had an acid decay of trust, which is not easily restored.
Like all great disasters, it appears from my small window to have been a systemic event, one that required sharp failures by businesses at vital steps of their processes as confidence drained away.
What was once innovative, useful, and trustworthy fell to cunning corner-cutting. The increasing pace of business chipped away at previous good practices, people compromised process for reasons that seemed acceptable at the time, and business cultures based on relationships of trust changed under the pressure of new market forces.
We in the financial industry cannot help but draw on knowledge gained through our own experience of recent events, including knowledge of the deeds of great men, the consequences of which we have seen with our own eyes.
Gurus and Mahayanas
The first man here is Lewis Ranieri, often called the father of the mortgage-backed security (MBS). He helped develop these in the late 70s and 80s, eventually launching a large business for Salomon. The problem the early MBSs solved was simply that savings-and-loans (remember them?) were in the business of financing 30-year mortgages with deposits from regular people — normal short-term household and small-business deposits. But the interest rates in the United States at that time made this a suffocating situation.
There was more capital out there in the world, which normal people might use. How to get it? Ranieri pooled mortgages and packaged them into securities. These securities were then sliced into “tranches,” so they could become far shorter-term assets attractive to investors anywhere. Ranieri hired Ph.D.s to use mathematical models for this purpose. Ranieri once estimated this innovation cut mortgage rates by 2% at the time.
The success of this model soon led to the securitization of other forms of debt, like credit card balances and car loans. This process funneled global capital into the United States in a new and useful way.
As applied to all kinds of fixed-income assets, these instruments become collateralized debt obligations (CDOs), which were invented in the late 80s. CDO creators structure them to minimize taxes and, because they should pool debt advantageously, they can offer returns higher than the original underlying instruments.
CDOs became more prominent after 2001, when David Li invented a new mathematical model to allow them to be priced more quickly. This was an important step in moving forward the time-consuming process of origination.
So the MBS/CDO business gets going in the late 80s. Fast-forward through the recession, the S&L crisis, prosperity, the Asian flu, the Russian crisis, LTCM, the birth of the dot-com bubble. . . and we arrive in the late 90s.
The second man here is William Demchak, head of the team that popularized derivatives. At that time he was at J. P. Morgan, acting in response to the 1997 Asian flu, a crisis that revealed Morgan’s exposure to Asian companies. His response to this weakness was to persuade Morgan to adopt derivatives, including the insurance-like Credit Default Swap (CDS), a move widely credited to the efforts of Blythe Masters. [1] [2] Masters was lionized on Wall Street for her acumen and towered as a powerful female executive at what became JPMorgan Chase.
The Demchak team took Morgan’s exposure, securitized it (removing it from visibility on the balance sheet), applied CDS, and sold it. An important but little-noted aspect of these innovations — whose relevance may be seen later — is that pieces of this package can be substituted for others as long as they each have the same rating from the ratings agency.
The task of creating a pool of 1,000 to 25,000 mortgages and slicing it so that each piece, or tranche, gets assigned the desired rating by the ratings agencies’ models, and so that any losses would attach at the right points, became an integral part of the structuring process. This makes the rating agency a key player in origination, not an independent third-party verifier.
Indeed, it appears JP Morgan’s first deal, known as Bistro, involved Standard & Poors helping to craft the package from the start, in what at the time probably seemed like an effort at efficient due diligence, but which later became a problematic business practice. [3]
“You start with a rating and build a deal around a rating,” Moody’s mortgage executive Brian Clarkson has said, according to the Financial Times. “We’re in the service business,” and “I don’t apologize for that.”
What Default Risk?
Those trading in Credit Default Swaps recognized that they offered risk. But, at the, time the institutions against which these were written appeared so large, so well capitalized, that no one would have dreamed they could fail in a “sound” economy, despite the recent write-downs from mortgages.
For example, who would ever have thought that Lehman — which had been in New York since before the Civil War — could go down? The large amounts invested in CDSs show that the market believed the benefits of liquidity and portioned risk must have been greater than any actual risk of default.
At the time, the probability of a Lehman default would have seemed as close to zero as could be imagined, short of nuclear war. In 2006 its equity base was about $20 billion, with assets of over $500 billion. The large write-downs in 2007 were painful, but still Lehman reported profits of $4 billion that year.
Only after the fall of Bear Sterns could the deeper problems afflicting Lehman be seen. And, once seen, the situation still didn’t appear hopeless; a merger seemed like a viable way out.
No Need to Touch the Stuff
The CDO soon became more abstract and complex as the economy did. Mixing the CDO with CDSs in the structure results in the so-called “synthetic CDO”; this prevents people having to take the underlying instruments, as they buy an abstracted flow based on instruments that are based on the performance of the swaps involved. [4]
(There isn’t enough space here to go into another structured situation, the SIV, where institutions attempted to profit by issuing low-interest, short-term instruments and buying higher-interest long-term securities. This practice however was responsible for some notable write-downs at large banks. But it also appeared to have favorable tax and accounting features.)
And here we have the vehicles of the incipient disaster, each created in response to a particular immediate business problem, or continuing a common business paradigm. These vehicles were then extended, and perhaps even distorted, as the financial environment changed. Humans are thrifty — we attempt to re-use and repurpose engines proven to power us into the future.
While some in the mainstream media (MSM) have argued that MBS structures are inherently weak or “toxic,” as a business practice they have proven useful for nearly 30 years; the CDO has proven useful for more than 20. The credit derivative/CDS, with only about 10 years in popular use, survived the post-9/11 downturn and the bursting of the dot-com bubble. If all these vehicles were as inherently poisonous as the MSM now leads us to believe, how could they have survived so much change and turmoil in the marketplaces?
Here I return to the notion of systemic failure, or to quote a genius friend of mine: “Guns don’t kill people; people kill people.” To which I would add: “Greedy people kill smart people.” With shocking regularity.
I suggest the issue was not, as some in the MSM have argued, that the financial structures are themselves inherently opaque and beyond all control. Increasing abstraction in all areas of the economy appears to be a necessary part of our technological rise. That no one understood what was in these CDOs seems more a fault of inadequate documentation in the business process than of the instrument itself.
The real problem develops by 2001 with the attempt to “manufacture” instruments just-in-time in the marketspace in something like an assembly line, with incentives so aligned that those with the most cash at stake were most driven to corrupt the process, while those who knew about the problems didn’t have the power to fix it.
The Frankenstein practice of stitching together CDOs from other CDOs obscured the amount of counterparty risk to hurried traders, who were warming seats just long enough to collect their annual bonuses. All the while, senior management sought ever-larger returns, beyond normal levels of leverage and risk.
The situation became clearer in 2006, by which time the old culture of the banks had changed so much that, instead of acting mostly as brokers in the trading divisions, they had become active market participants, using their own funds and borrowing against their reputations.
Future Echo of the Inverse-Doppler-Shifted Quant
“Maybe the problem wasn’t too much rocket science, but too little. . . . The problem with the models is that they did not assign a high enough chance of occurrence to the scenario in which many things go wrong at the same time — the ‘perfect storm’ scenario,” a September 1998 Business Week piece about the LTCM fiasco concluded.
The MSM has been hard on the quantitative and modeling community. Even other scientists have: the November Scientific American sneers at the well-paid quants, calling them “lapsed physicists.”
Employees at the ratings agencies were themselves well aware of the deficiencies in their models. Some of the most fascinating pieces of flotsam to emerge from the October House Oversight Committee hearings after impact were documents, emails, and IMs from those working with these models.
In the testimony of one modeling executive, he relates how later, more accurate and fine-tuned versions of the model were not implemented at his agency, despite executive knowledge that the new models would improve ratings.
In an IM from April 2007 that splashed onto financial blogs, one rater writes to another, “That deal is ridiculous.” To which the reply is simply, “Model def does not capture half the risk.” “We should not be rating it,” frets the first. “We rate every deal,” is the chilling response. “I personally don’t feel comfy signing off” on it, the second admits. [5]
Rocket scientists and those in charge of rating models were not powerful players in the business chain. A 2001 IM conversation submitted as evidence to the House, tells the tale of an executive rater at S&P who asks to see the mortgages underlying the CDO. The agencies don’t collect any data of their own; instead they rely on what are called representations and warranties from the issuer.
A managing director fires electrons back. “Any request for loan level tapes is TOTALLY UNREASONABLE!!!. … Most investors don’t have it and can’t provide it. … we MUST produce a credit estimate. … It is your responsibility to provide those credit estimates and your responsibility to devise some method for doing so.” Just make it up, geekboy, chop-chop. The Financial Times reported that one billion-dollar structured deal was rammed through its rating in just 90 minutes.
Here several of the clear failure points in the business culture arise, derived from the new model of “industrial” structuring: management coziness to get deals rated no matter what, foolish penny-pinching in updating software and IT, apparent over-reliance on information from issuers without any independent verification, and lack of will or ability among employees to blow the whistle. The House hearings suggest that these practices went on for at least six years, probably right up until the meteor hit this Fall.
For all the flaws inherent in historically based models, how can it be fairly said that modelers were in control of the strategic direction of any of the big firms — of any of the banks, investment banks, or rating agencies? The House documents above appear to show that they weren’t.
Didn’t the highest executives, knowing that reward requires risk, incentivize management to find more profit, and thus to take on more risk? Check back later with Business Week. Maybe the problem wasn’t too much rocket science, but still too little.
Others Playas After the Dot-Com Bubble
The Wall Street we knew and loved was itself a structured product, one created by Congress 75 years ago, in which the investment banks were private. With origins in the 19th century, investment banks did relationship banking, helping their long-time corporate friends with financial transactions, such as issuing bonds, going to the market, and advising on mergers and acquisitions.
They were partnerships, like law firms. You had a stake at the investment bank, a future; you might expect to be there 30 years. It was not in your interest to harm the bank.
As the global financial markets changed, the great banks found they had to change, too. In particular Glass-Steagall’s repeal, meant to make institutions more competitive with foreign entities and allow them to diversify, gave the banks newer and meaner competitors as firms reconstituted in the new marketplace. Once again, adjusting great proven vehicles to new situations, the banks saw that to stay relevant, they had to recreate themselves.
One by one the banks opened themselves to the public; Goldman was the last to go, in 1999, when it IPO’d 52% of itself, holding 48% in partnership. The credit rating agencies had likewise gone from early-20th-century specialized publishing houses to full-service entities that, since the 1970s, had been charging issuers as well as investors.
By the end of the dot-com bubble, there were no private investment banks left, and Moody’s went public in 2000. The culture had overall morphed to one of total short-term taking, creating the “IBG YBG” (I be gone, you be gone) attitude, according to which no one cared about their institution, but just about their own yearly bonus. [6] Add another failure point: the inability of banks to realize their incentive platforms were turning their employees against them.
But there were lots of newer creatures, the hedge funds, invented in 1949 by financial journalist Alfred Jones — short-selling, high-risk-hungry, hedge funds. They loved risk and leverage, developing highly successfully techniques at great profit.
Large banks, under shareholder pressure, with executive compensation tied to performance, copied these techniques and became themselves highly leveraged, completely changing their manner of business, moving from brokering to more active participation. [7] In fact, it now appears the banks leveraged themselves more than even most hedge funds.
In the dot-com aftermath, rates went low. By Spring 2005 how easy it was to open a newspaper and read about the endless ocean of cash sloshing around with no good place to go but into housing: “In today’s world, the most notable … markets are housing and hedge funds,” the San Francisco Chronicle marveled.
From dot-com bubble to housing bubble to debt bubble, the champagne party began. Investment banks’ appetite for risk had never seemed greater. [8] Have there ever been three bubbles right in a row before?
I hope that I have sketched how, by this time, all the vehicles, instruments, practices, cultures, and players were in place with plenty of cash to toss into this game. The disaster could not have happened just a few years earlier.
An Eye on the Fall
As the subprime crisis developed and appeared to recede — after the implosion of Countrywide, which had spun off IndyMac, which was seized by the Feds last summer — and after Bear fell into the trap — pundits began to argue what market would go next.
In this context, it seems important to note that IndyMac is said to have gone down in a classic bank run, albeit they held plenty of Alt-A mortgages, and that SEC chief Christopher Cox claimed a run on the hedge funds was what ultimately brought Bear down.
Many market watchers have claimed or have been given credit for “calling” the present crisis sometime this summer, but a creeping trust problem seems to have been corroding the undersides of market structures since Autumn 2007, permabears aside. [9] [10]
In retrospect this seems about the time loss of confidence began to spread from one market to another. For example, popular pundit Jim Jubak of MSN Money argued on January 18th that subprime would be dwarfed by issues with CDS. The recent Nobel Prize winner Paul Krugman likewise predicted that credit crunch difficulties in “obscure” markets would spill over to normal people in the regular economy in a comment about student loans on February 13th.
The specific implications of this for Wall Street investment banks’ MBS issues seemed to emerge last Spring. One example is an April 10th Financial Times blog posting in which commenters roast Goldman, predicting not just more write-downs, but suggesting insolvency — and wider troubles that would “eradicate book value across the financial system.” The Wall Street Journal also ran an article at the time.
The late, widely admired banking expert Tanta blogged a May 22nd Reuters report on CalculatedRisk, noting that defaults on certain vintages of subprime reached 37 percent, or “4 percentage points higher than previous estimates, S&P said.” She noted other figures showing prime defaults were also rising above expectations.
Recall that S&P had been rating these kinds of deals. It seems safe to assume that the extra 4 percent S&P hadn’t expected certainly wasn’t accounted for in the models made years ago, when its employees were already squirming as they force-rated deals.
In short, everyone called this crisis. It loomed openly over the public’s head like Hiroshige’s wave all summer as the MSM studied its navel to find yet more things to say about the psychological aspects of the Clinton-Obama relationship.
Business reporters and bloggers were sounding the alarm, but the election horse race appeared to keep MSM editors from promoting this story out of the back sections. If the press is a watchdog, does it have an obligation to our democracy beyond partisan election fever?
The Death of Trust
With default rates building like this over the summer, the only question was when we would drown. “The first wave of Americans to default on their home mortgages appears to be cresting, but a second, far larger one is quickly building,” remarked the New York Times on August 4th.
By August 8th, the Times wrote that “reining in the runaway freight train otherwise known as the credit default swap market is a rising priority for regulators who oversee banks and insurers. . . . This may be a rude awakening for many players in financial markets,” the paper noted delicately. This article is also notable for the caption on the main picture, in which a state regulator says CDS valuations are too optimistic.
He “pushed” the banks to unwind some deals, the Times says, using a verb that doesn’t seem even at first reading to be on par with the scale of the problem described. Not exactly an Eliot Ness tone, from the Times‘ description. “It would be very valuable for the bond insurers if we could resolve all of these,” the regulator tells the paper, in what should go down in history with other great understatements.
And here’s the last point of business failure: the regulators seemed to move too slowly long after the dogs were barking. While the MSM might be bored by stories about screaming default rates and absurd valuations, maybe the “priority” of that “runaway freight train” should have risen much higher, much faster for the bureaucrats?
The SEC regulators sitting in Lehman, those watching over Fannie, the entire chess set somehow missed the beat. [11] [12] With the call for more regulations now in a cavernous echo, even a neutral observer might ask: Are regulators able to follow what is going on in our complex, electronic markets? Can they ever act quickly enough, even if alerted weeks in advance, no matter how many new laws are enacted? Aren’t regulators always going to be underfunded, behind innovation, and reactive?
MBSs depend on solid payment streams to work. As rising defaults would hit the tranches, it now seems obvious that this had to be over in September. In May, NYU economist Nouriel Roubini had stated in a conference that “the bulk of the CDS protection-writing is likely to be concentrated in a comparatively small number of hands.” [13]
And so it proved to be. Those hands also proved to have intricately woven their fingers into those of many other players — JPMorgan Chase had been doing CDO/CDS business with AIG from early days, for example — thus putting them all in jeopardy.
As the losses attached, positions deteriorated. Firms didn’t have the reserves to match. As institutions weakened, those with CDSs had to put up collateral, which they didn’t have. [14] Leveraged to the hilt, with inadequate reserves, exposed to counterparty risk, unable to see what was in the tranches they had purchased, default rates mounting, and without confidence to support them, the giants fell.
The exact details of the catastrophe are too fresh in everyone’s minds to need reciting. But the ultimate mistake, in an environment already so short of confidence, now seems to have been Treasury’s letting Lehman fail overnight. This made other institutions fear even the standard lending from one bank to another.
No one knew who was solvent or would be solvent, or who the Administration would let die overnight in their next fit of uncharacteristic principle. No one wanted to risk any capital by sending it to an institution that might not be there the next morning to give it back. Lack of confidence became a flat-out refusal of trust.
So the full-blown disaster was born. Like a slime-mold it seems to flow endlessly towards us in unstoppable slow motion even today, despite so many extraordinary measures. Remember Treasury Secretary Paulson’s first declared victory over the credit crisis last May? [15]
What other circumstances could explain the fact that last week investors were for a moment even buying T-bills at negative rates? [16] People would rather take a guaranteed 3-month loss on a known thing than even try to make a profit on an unknown thing, as they desperately seek someplace to put their money that feels even remotely predictable.
The world needs not only more rescue, more financial back-stopping, but primarily, new modes for trust. [17] This is the financial innovation we most require now.
Notes
[1] “Ten years ago, a “watershed” moment changed the profile of the derivatives that Mr. Cassano [of AIG] traded, according to a transcript of comments he made at an industry event last year. Derivatives specialists from J. P. Morgan, a leading bank that had many dealings with Mr. Cassano’s unit, came calling with a novel idea.
Morgan proposed the following: AIG should try writing insurance on packages of debt known as “collateralized debt obligations.” CDOs were pools of loans sliced into tranches and sold to investors based on the credit quality of the underlying securities.
“The proposal meant that the London unit was essentially agreeing to provide insurance to financial institutions holding CDOs and other debts in case they defaulted — in much the same way some homeowners are required to buy mortgage insurance to protect lenders in case the borrowers cannot pay back their loans.“ http://www.nytimes.com/2008/09/28/business/28melt.html?pagewanted=2
The “insurance” was the CDS.
[2] http://mycrains.crainsnewyork.com/40under40/profiles/2004/228. Also note that some credit these two with the invention of the CDS; others say the actual invention belonged to the old Bankers Trust.
[3] http://www.portfolio.com/views/columns/wall-street/2008/10/15/Credit-Derivatives-Role-in-Crash, see p.4.
[4] A nice description of this with the math is at http://derivativedribble.wordpress.com/2008/11/11/synthetic-cdos-demystified/.
[5] Paul Kedrosky’s crucial Infectious Greed, for example.
[6] A term perhaps popularized by Jonathan A. Knee, The Accidental Investment Banker (Oxford University Press, 2006).
[7] “This trading boom, fueled by cheap money, is fundamentally different from the ones of the past. When traders last ruled Wall Street, during the mid-’90s, few banks put much of their own balance sheets at risk; most acted mainly as brokers, arranging trades between clients. Now, virtually all banks are making huge bets with their own assets on many more fronts, and using vast sums of borrowed money to jack up the risk even more.” http://www.businessweek.com/magazine/content/06_24/b3988004.htm.
[8] “Wall Street has always been about taking risk. But never has the “R” word been such an obsession for the men and women who rule the nation’s biggest investment banks. Never have they had to reconcile so many bets made on so many fronts. The conditions have been ripe. Historically low interest rates and relatively calm markets in the last few years have allowed a new type of firm to flourish, one that acts primarily as a trader and only secondarily as a traditional investment bank.” http://www.businessweek.com/magazine/content/06_24/b3988004.htm.
[9] Paul Krugman uses the term” crisis of confidence” in November: http://krugman.blogs.nytimes.com/2007/11/23/listening-to-libor/.
[10] Peter Schiff: http://www.youtube.com/watch?v=2I0QN-FYkpw.
[11] http://www.youtube.com/watch?v=H-Jm2ILQxTE – yes @2:28, Lehman’s Fuld explains how he was running what sounds like it might(?) have been an SIV past his in-house regulators. Interesting, no? If anyone has any more info on that bit of testimony, many people would like to hear it, I’m sure.
[12] http://www.bloomberg.com/apps/news?pid=20601087&sid=aQRJa.rKf6Ag&refer=home: ”It is remarkable that during the period that Fannie Mae substantially increased its exposure to credit risk its regulator made no visible effort to enforce any limits,” Raines, 59, who was ousted in 2004 and accused of accounting manipulation, told the House Oversight and Government Reform Committee in Washington today.”
[13] See Nouriel Roubini’s comments: “Consider further: we have a $62 trillion notional credit default swaps market versus $6 trillion in cash bonds, The market has been tested only for a couple of individual defaults, not a large scale downturn. The bulk of the CDS protection-writing is likely to be concentrated in a comparatively small number of hands.” http://www.nakedcapitalism.com/2008/05/nyc-rge-monitor-panel-discussion.html
[14] http://www.nytimes.com/2008/09/28/business/28melt.html: ”By Sept. 15, the unit’s troubles forced a major downgrade in AIG’s debt rating, requiring the company to post roughly $15 billion in additional collateral — which then prompted the federal rescue.”
You see, AIG had never had to put up any collateral on these before due to its own reputation!
“AIG Financial Products was happy to book income in exchange for providing insurance. After all, Mr. Cassano and his colleagues apparently assumed, they would never have to pay any claims.
Since AIG itself was a highly rated company, it did not have to post collateral on the insurance it wrote, analysts said. That made the contracts all the more profitable.” http://www.nytimes.com/2008/09/28/business/28melt.html?pagewanted=2
[15] Yes he did – Mission accomplished! http://www.iht.com/articles/2008/05/07/business/7usecon.php but then, so did the UK: http://www.marketwatch.com/news/story/bank-england-signals-worst-credit/story.aspx?guid={0FE859BB-9A81-436F-BBA5-C026931B8682}
[16] “The annualized yield on three-month T-bills dipped slightly below zero at one point Tuesday, to negative 0.01%, according to Bloomberg News data.” http://www.latimes.com/business/la-fi-treasury10-2008dec10,0,972658.story
[17] ‘Cuz when it’s gone, it’s gone, baby, gone, see: http://knowledge.wharton.upenn.edu/papers/1321.pdf
January 12th, 2009 at 7:34 pm
[...] Anonymous, “A Worm’s-Eye View of Wall Street” [...]
January 16th, 2009 at 11:19 am
[...] Anonymous, “A Worm’s-Eye View of Wall Street” [...]
January 23rd, 2009 at 10:50 pm
[...] Anonymous, “A Worm’s-Eye View of Wall Street” [...]
January 28th, 2009 at 3:18 pm
[...] Anonymous, “A Worm’s-Eye View of Wall Street” [...]