Ratings agencies helped this game along in two ways. First, banks needed them to sign off on the bogus math of the subprime era – the math that allowed banks to turn pools of home loans belonging to people so broke they couldn't even afford down payments into securities with higher credit ratings than corporations with billions of dollars in assets. But banks also needed the ratings agencies to sign off on the safety and reliability of these off-balance-sheet SIV structures.
The first of the two SIVs in question was dreamed up by a London-based hedge fund called Cheyne Capital Management (pronounced like Dick "Cheney"), run by an ex-Morgan Stanley banker duo who hired their old firm to build and stock this vast floating Death Star of subprime loans.
Morgan Stanley had multiple motives for putting together the Cheyne deal. For one thing, it earned what the bank's lead structurer affectionately called "big fat upfront fees," which bank executives estimated would eventually add up to $25 million or $30 million. It was a lucrative business, and the top dogs wanted the deal badly. "I am very focused on . . . getting this deal done to get NY to stop freaking out" and "to make our money," said Robert Rooney, the senior Morgan Stanley executive on the deal. A spokesman for Morgan Stanley, however, told Rolling Stone, "Our sole economic interest was in the ongoing success of the SIV."
But that wasn't Morgan Stanley's only motive. Not only could the bank make the "big fat upfront fees" for structuring the deal, they could also turn around and sell scads of their own mortgage-backed securities to the SIV, which in turn would be marketed to investors like Abu Dhabi and King County. In Cheyne, 25 percent of the original assets in the deal came from Morgan Stanley – over time, $2 billion of the SIV's $9 billion to $10 billion portfolio of assets came from the bank as well.
Internal Morgan Stanley memorandums show that the bank knowingly stuffed mortgages in the SIV whose borrowers were, to say the least, highly suspect. "The real issue is that the loan requests do not make sense," complained a Morgan Stanley employee back in 2005. He noted loans had been made to a "tarot reading house" operator who claimed to make $12,000 a month, and a "knock off gold club distributor" who claimed to make $16,000 a month. "Compound these issues," he groaned, "with the fact that we are seeing what I would call a lot of this type of profile."
No matter – into the soup it went! Morgan sold mountains of this crap into Cheyne's SIV, where it was destined to be sold off to other suckers down the line. The only thing that could possibly get in the way of the scam was some pesky ratings agency.
Fortunately for the bank and the hedge fund, these subprime SIVs were a relatively new kind of investment product, so the ratings agencies had little to go on in the area of historical data to measure these products. One might think this would make the ratings agencies more conservative. In fact, caution in the face of the unknown was supposed to be a core value for these companies. As Moody's put it, "Triple-A structures should not be highly dependent on untestable assumptions."
But when it came to the Cheyne SIV, Moody's punted on caution. In an e-mail sent to executives from both Morgan Stanley and Cheyne in May 2005, David Rosa, a Moody's senior analyst, admitted that when it came to this SIV, he had nothing to go on.
"Please note that in relation to assumed spread [volatility] for the Aa and A there is no actual data backing up the current model assumptions," he wrote. In lieu of such data, he went on, "We will for now accept the proposal to use the same levels as [residential mortgage-backed securities] given that this assumption is supported by the analysis of the Aaa data . . . and Cheyne's comments on their views of this asset class."
Translation: We have no historical data, so we'll just accept your reasoning for the time being, even though you have every incentive in the world to lie about the quality of your product.
At one point, a Morgan Stanley analyst even claimed that the bank had written, in Moody's name, an entire 12-page "New Issue Report" for the Cheyne SIV – a kind of ratings summary in which Morgan Stanley appears to have given itself AAA ratings for large chunks of the deal. "I attach the Moody's NIR (that we ended up writing)," yawns Morgan Stanley fixed-income employee Rany Moubarak in a March 2006 e-mail. The attached document came proudly affixed with the "Moody's Investors Service" logo. (Both Moody's and Morgan Stanley deny that anyone other than Moody's wrote that report.)
Morgan Stanley ended up getting both Moody's and S&P to rate the deal, and that was not only common, it was basically industry practice. There were many reasons for this, but a big one was a concept called "notching," in which the agencies gave ratings penalties to any instrument that had not been rated by their own company. If a SIV contained a basket of mortgage-backed securities rated AA by Standard & Poor's, Moody's might "notch" those underlying securities down to A, or even lower. This incentivized the banks to hire as many ratings agencies as possible to rate every investment vehicle they created.
Again, despite the fact that the ratings agencies enjoyed broad quasi-official subsidies, and despite the powerful market leverage that techniques like "notching" gave them, they still routinely chose to roll over for banks. And the biggest companies were equally guilty. In the case of the Cheyne deal, Standard & Poor's was every bit as craven as Moody's.
In September 2004, an S&P analyst named Lapo Guadagnuolo sent an e-mail to Stephen McCabe, the agency's lead "quant" on the Cheyne deal, who apparently was on vacation. The e-mail chain was mostly a bunch of office gossip, where the two men e-whispered about an employee who was about to quit. But sandwiched in the office banter was an offhand line about the Cheyne deal and how full of shit it was. "Hi Steve!" Guadagnuolo wrote cheerily, adding, "How is Australia and how was Thailand????Back to [Cheyne] . . . As you know, I had difficulties explaining 'HOW' we got to those numbers since there is no science behind it . . .
"Thanks and regards . . . have you heard that [redacted] has resigned . . . and somebody else will follow suit today!!"
McCabe, blowing off the "no science behind it" comment, answered eagerly, "Who, Who, Who????" The quadruple question mark must be an S&P-ism.
A month later, McCabe seemed more concerned about the lack of science in the Cheyne deal. He complained in an e-mail to his boss, Kai Gilkes, who was the agency's senior quantitative analyst in Europe.
"From looking at the numbers it is quite obvious that we have just stuck our preverbal [sic] finger in the air!!" he fumed.
Gilkes was experiencing his own crisis of conscience by mid-2005, complaining in an oddly wistful e-mail to another S&P employee that the good old days of just giving things the ratings they deserved were disappearing. "Remember the dream of being able to defend the model with sound empirical research?" he wrote on June 17th, 2005. "If we are just going to make it up in order to rate deals, then quants are of precious little value."
Frank Parisi, Standard & Poor's chief credit officer for structured finance, was even more downtrodden, saying that the model that his company used to rate residential mortgage-backed securities in 2005 and 2006 was only marginally more accurate than "if you just simply flipped a coin."
To read the new issue of Rolling Stone online, plus the entire RS archive: Click Here
MUSIC 9 Classic Devo Videos
OLYMPICS 18 Epic Opening Ceremonies
Picks From Around the Web
blog comments powered by Disqus