Inside Morgan Stanley, there was apparently never much question whether the company's elite risk takers should be allowed to buy $16 billion in subprime mortgage bonds. Howie Hubler's proprietary trading group was of course required to supply information about its trades to both upper management and risk management, but the information the traders supplied disguised the nature of their risk. The $16 billion in subprime risk Hubler had taken on showed up in Morgan Stanley's risk reports inside a bucket marked "triple A"--which is to say, they might as well have been U.S. Treasury bonds. They showed up again in a calculation known as value at risk (VaR). The tool most commonly used by Wall Street management to figure out what their traders had just done, VaR measured only the degree to which a given stock or bond had jumped around in the past, with the recent movements receiving a greater emphasis than movements in the more distant past. Having never fluctuated much in value, triple-A-rated subprime-backed CDOs registered on Morgan Stanley's internal reports as virtually riskless. In March 2007 Hubler's traders prepared a presentation, delivered by Hubler's bosses to Morgan Stanley's board of directors, that boasted of their "great structural position" in the subprime mortgage market. No one asked the obvious question: What happens to the great structural position if subprime mortgage borrowers begin to default in greater than expected numbers?
Howie Hubler was taking a huge risk, even if he failed to communicate it or, perhaps, understand it. He'd laid a massive bet on very nearly the same CDO tranches that Cornwall Capital had bet against, composed of nearly the same subprime bonds that FrontPoint Partners and Scion Capital had bet against. For more than twenty years, the bond market's complexity had helped the Wall Street bond trader to deceive the Wall Street customer. It was now leading the bond trader to deceive himself.
At issue was how highly correlated the prices of various subprime mortgage bonds inside a CDO might be. Possible answers ranged from zero percent (their prices had nothing to do with each other) to 100 percent (their prices moved in lockstep with each other). Moody's and Standard & Poor's judged the pools of triple-B-rated bonds to have a correlation of around 30 percent, which did not mean anything like what it sounds. It does not mean, for example, that if one bond goes bad, there is a 30 percent chance that the others will go bad too. It means that if one bond goes bad, the others experience very little decline at all.
The pretense that these loans were not all essentially the same, doomed to default en masse the moment house prices stopped rising, had justified the decisions by Moody's and S&P to bestow triple-A ratings on roughly 80 percent of every CDO. (And made the entire CDO business possible.) It also justified Howie Hubler's decision to buy 16 billion dollars' worth of them. Morgan Stanley had done as much as any Wall Street firm to persuade the rating agencies to treat consumer loans as they treated corporate ones--as assets whose risks could be dramatically reduced if bundled together. The people who had done the persuading saw it as a sales job: They knew there was a difference between corporate and consumer loans that the rating agencies had failed to grapple with. The difference was that there was very little history to work with in the subprime mortgage bond market, and no history at all of a collapsing national real estate market. Morgan Stanley's elite bond traders did not spend a lot of time worrying about this. Howie Hubler trusted the ratings.
The Wall Street bond traders on the other end of the phone from Howie Hubler came away with the impression that he considered these bets entirely risk-free. He'd collect a tiny bit of interest...for nothing. He wasn't alone in this belief, of course. Hubler and a trader at Merrill Lynch argued back and forth about a possible purchase by Morgan Stanley, from Merrill Lynch, of $2 billion in triple-A CDOs. Hubler wanted Merrill Lynch to pay him 28 basis points (0.28 percent) over the risk-free rate, while Merrill Lynch only wanted to pay 24. On a $2 billion trade--a trade that would, in the end, have transferred a $2 billion loss from Merrill Lynch to Morgan Stanley--the two traders were arguing over interest payments amounting to $800,000 a year. Over that sum the deal fell apart. Hubler had the same nit-picking argument with Deutsche Bank, with a difference. Inside Deutsche Bank, Greg Lippmann was now hollering at the top of his lungs that these triple-A CDOs could one day be worth zero. Deutsche Bank's CDO machine paid Hubler the 28 basis points he craved and, in December 2006 and January 2007, cut two deals, of $2 billion each. "When we did the trades, the whole time we were both like, 'We both know there is no risk in these things,'" said the Deutsche Bank CDO executive who dealt with Hubler.
In the murky and curious period from early February to June 2007, the subprime mortgage market resembled a giant helium balloon, bound to earth by a dozen or so big Wall Street firms. Each firm held its rope; one by one, they realized that no matter how strongly they pulled, the balloon would eventually lift them off their feet. In June, one by one, they silently released their grip. By edict of CEO Jamie Dimon, J.P. Morgan had abandoned the market by the late fall of 2006. Deutsche Bank, because of Lippmann, had always held on tenuously. Goldman Sachs was next, and did not merely let go, but turned and made a big bet against the subprime market--further accelerating the balloon's fatal ascent.* When its subprime hedge funds crashed in June, Bear Stearns was forcibly severed from its line--and the balloon drifted farther from the ground.
Not long before that, in April 2007, Howie Hubler, perhaps having misgivings about the size of his gamble, had struck a deal with the guy who ran the doomed Bear Stearns hedge funds, Ralph Cioffi. On April 2, the nation's largest subprime mortgage lender, New Century, was swamped by defaults and filed for bankruptcy. Morgan Stanley would sell Cioffi $6 billion of his $16 billion in triple-A CDOs. The price had fallen a bit--Cioffi demanded a yield of 40 basis points (0.40 percent) over the risk-free rate. Hubler conferred with Morgan Stanley's president, Zoe Cruz; together they decided that they'd rather keep the subprime risk than realize a loss that amounted to a few tens of millions of dollars. It was a decision that wound up costing Morgan Stanley nearly $6 billion, and yet Morgan Stanley's CEO, John Mack, never got involved. "Mack never came and talked to Howie," says one of Hubler's closest associates. "The entire time, Howie never had a single sit-down with Mack."*
By May 2007, however, there was a growing dispute between Howie Hubler and Morgan Stanley. Amazingly, it had nothing to do with the wisdom of owning $16 billion in complex securities whose value ultimately turned on the ability of a Las Vegas stripper with five investment properties, or a Mexican strawberry picker with a single $750,000 home, to make rapidly rising interest payments. The dispute was over Morgan Stanley's failure to deliver on its promise to spin Hubler's proprietary trading group off into its own money management firm, of which he would own 50 percent. Outraged by Morgan Stanley's foot-dragging, Howie Hubler threatened to quit. To keep him, Morgan Stanley promised to pay him, and his traders, an even bigger chunk of GPCG's profits. In 2006, Hubler had been paid $25 million; in 2007, it was understood, he would make far more.
A month after Hubler and his traders improved the terms of trade between themselves and their employer, Morgan Stanley finally asked the uncomfortable question: What happened to their massive subprime mortgage market bet if lower-middle-class Americans defaulted in greater than expected numbers? How did the bet perform, for instance, using the assumption of losses generated by the most pessimistic Wall Street analyst? Up to that point, Hubler's bet had been "stress tested" for scenarios in which subprime pools experienced losses of 6 percent, the highest losses from recent history. Now Hubler's traders were asked to imagine what would become of their bet if losses reached 10 percent. The demand came directly from Morgan Stanley's chief risk officer, Tom Daula, and Hubler and his traders were angered and disturbed that he would issue it. "It was more than a little weird," says one of them. "There was a lot of angst about it. It was sort of viewed as, These folks don't know what they're talking about. If losses go to ten percent there will be, like, a million homeless people." (Losses in the pools Hubler's group had bet on would eventually reach 40 per
cent.) As a senior Morgan Stanley executive outside Hubler's group put it, "They didn't want to show you the results. They kept saying, That state of the world can't happen."
It took Hubler's traders ten days to produce the result they really didn't want to show anyone: Losses of 10 percent turned their complicated bet in subprime mortgages from a projected profit of $1 billion into a projected loss of $2.7 billion. As one senior Morgan Stanley executive put it, "The risk officers came back from the stress test looking very upset." Hubler and his traders tried to calm him down. Relax, they said, those kinds of losses will never happen.
The risk department had trouble relaxing, however. To them it seemed as if Hubler and his traders didn't fully understand their own gamble. Hubler kept saying he was betting against the subprime bond market. But if so, why did he lose billions if it collapsed? As one senior Morgan Stanley risk manager put it, "It's one thing to bet on red or black and know that you are betting on red or black. It's another to bet on a form of red and not to know it."
In early July, Morgan Stanley received its first wake-up call. It came from Greg Lippmann and his bosses at Deutsche Bank, who, in a conference call, told Howie Hubler and his bosses that the $4 billion in credit default swaps Hubler had sold Deutsche Bank's CDO desk six months earlier had moved in Deutsche Bank's favor. Could Morgan Stanley please wire $1.2 billion to Deutsche Bank by the end of the day? Or, as Lippmann actually put it--according to someone who heard the exchange--Dude, you owe us one point two billion.
Triple-A-rated subprime CDOs, of which there were now hundreds of billions of dollars' worth buried inside various Wall Street firms, and which were assumed to be riskless, were now, according to Greg Lippmann, only worth 70 cents on the dollar. Howie Hubler had the same reaction. What do you mean seventy? Our model says they are worth ninety-five, said one of the Morgan Stanley people on the phone call.
Our model says they are worth seventy, replied one of the Deutsche Bank people.
Well, our model says they are worth ninety-five, repeated the Morgan Stanley person, and then went on about how the correlation among the thousands of triple-B-rated bonds in his CDOs was very low, and so a few bonds going bad didn't imply they were all worthless.
At which point Greg Lippmann just said, Dude, fuck your model. I'll make you a market. They are seventy-seventy-seven. You have three choices. You can sell them back to me at seventy. You can buy some more at seventy-seven. Or you can give me my fucking one point two billion dollars.
Morgan Stanley didn't want to buy any more subprime mortgage bonds. Howie Hubler didn't want to buy any more subprime-backed bonds: He'd released his grip on the rope that tethered him to the rising balloon. Yet he didn't want to take a loss, and insisted that, despite his unwillingness to buy more at 77, his triple-A CDOs were still worth 95 cents on the dollar. He simply handed the matter to his superiors, who conferred with their equivalents at Deutsche Bank, and finally agreed to wire over $600 million. The alternative, for Deutsche Bank, was to submit the matter to a panel of three Wall Street banks, randomly selected, to determine what these triple-A CDOs were actually worth. It was a measure of the confusion and delusion on Wall Street that Deutsche Bank didn't care to run that risk.
At any rate, from Deutsche Bank's point of view, the collateral wasn't that big a deal. "When Greg made that call," said a senior Deutsche Bank executive, "it was like last on the list of the things we needed to do to keep our business running. Morgan Stanley had seventy billion dollars in capital. We knew the money was there." There was even some argument inside Deutsche Bank as to whether Lippmann's price was accurate. "It was such a big number," said a person involved in these discussions, "that a lot of people said it couldn't possibly be right. Morgan Stanley couldn't possibly owe us one point two billion dollars."
They did, however. It was the beginning of a slide that would end just a few months later, in a conference call between Morgan Stanley's CEO and Wall Street's analysts. The defaults mounted, the bonds universally crashed, and the CDOs composed of the bonds followed. Several times on the way down, Deutsche Bank offered Morgan Stanley the chance to exit its trade. The first time Greg Lippmann called him, Howie Hubler might have exited his $4 billion trade with Deutsche Bank at a loss of $1.2 billion; the next time Lippmann called, the price of getting out had risen to $1.5 billion. Each time, Howie Hubler, or one of his traders, argued about the price, and declined to exit. "We fought with those cocksuckers all the way down," says one Deutsche Bank trader. And, all the way down, the debt collectors at Deutsche Bank sensed the bond traders at Morgan Stanley misunderstood their own trade. They weren't lying; they genuinely failed to understand the nature of the subprime CDO. The correlation among triple-B-rated subprime bonds was not 30 percent; it was 100 percent. When one collapsed, they all collapsed, because they were all driven by the same broader economic forces. In the end, it made little sense for a CDO to fall from 100 to 95 to 77 to 70 and down to 7. The subprime bonds beneath them were either all bad or all good. The CDOs were worth either zero or 100.
At a price of 7, Greg Lippmann allowed Morgan Stanley to exit a trade it had entered into at roughly 100 cents on the dollar. On the first $4 billion of Howie Hubler's $16 billion folly, the loss came to roughly $3.7 billion. By then Lippmann was no longer speaking to Howie Hubler, because Howie Hubler was no longer employed at Morgan Stanley. "Howie was on this vacation thing for a few weeks," says one member of his group, "and then he never came back." He'd been allowed to resign in October 2007, with many millions of dollars the firm had promised him at the end of 2006, to prevent him from quitting. The total losses he left behind him were reported to the Morgan Stanley board as a bit more than $9 billion: the single largest trading loss in the history of Wall Street. Other firms would lose more, much more; but those losses were typically associated with the generation of subprime mortgage loans. Citigroup and Merrill Lynch and others sat on huge piles of the things when the market crashed, but these were the by-product of their CDO machines. They owned subprime mortgage-backed CDOs less for their own sake than for the fees that their deals would generate once they had sold them. Howie Hubler's loss was the result of a simple bet. Hubler and his traders thought they were smart guys put on earth to exploit the market's stupid inefficiencies. Instead, they simply contributed more inefficiency.
Retiring to New Jersey, with an unlisted number, Howie Hubler took with him the comforting sense that he was not the biggest fool at the table. He might have let go of the balloon rope too late to save Morgan Stanley, but, as he fell to earth, he could look up at the balloon drifting higher in the sky and see Wall Street bodies still dangling from it. In early July, just days before Greg Lippmann had called him to ask for $1.2 billion, Hubler had found a pair of buyers for his triple-A-rated CDOs. The first was the Mizuho Financial Group, a trading arm of Japan's second biggest bank. As a people, the Japanese had been bewildered by these new American financial creations, and steered clear of them. Mizuho Financial Group, for some reason that would remain known only to itself, set itself up as a clever trader of U.S. subprime bonds, and took $1 billion in subprime-backed CDOs off Morgan Stanley's hands.
The other, bigger, buyer was UBS--which took $2 billion in Howie Hubler's triple-A CDOs, along with a couple of hundred million dollars' worth of his short position in triple-B-rated bonds. That is, in July, moments before the market crashed, UBS looked at Howie Hubler's trade and said, "We want some of that, too." Thus Howie Hubler's personal purchase of $16 billion in triple-A-rated CDOs dwindled to something like $13 billion. A few months later, seeking to explain to its shareholders the $37.4 billion it had lost in the U.S. subprime markets, UBS would publish a semi-frank report, in which it revealed that a small group of U.S. bond traders employed by UBS had lobbied hard right up until the end for the bank to buy even more of other Wall Street firms' subprime mortgage bonds. "If people had known about the trade, it would have been open revolt," said one UBS bond trader close to the action. "It was a very controversial trade
in UBS. It was kept very, very secret. There were a lot of people, had they known the trade was happening, would have screamed eight ways from Sunday. We took the correlation trade off Howie's hands when everyone knew the correlation was one." (Which is to say, 100 percent.) He further explained that the traders at UBS who executed the trade were motivated mainly by their own models--which, at the moment of the trade, suggested they had turned a profit of $30 million.
On December 19, 2007, Morgan Stanley held a call for investors. The company wanted to explain how a trading loss of $9.2 billion--give or take a few billion--had more than overwhelmed the profits generated by its fifty thousand or so employees. "The results we announced today are embarrassing for me; for our firm," began John Mack. "This was a result of an error in judgment incurred on one desk in our Fixed Income area, and also a failure to manage that risk appropriately.... Virtually all write downs this quarter were the result of trading about [sic] a single desk on our mortgage business." The CEO explained that Morgan Stanley had certain "hedges" against its subprime mortgage risk and that "the hedges didn't perform adequately in extraordinary market condition of late October and November." But market conditions in October and November were not extraordinary; in October and November, for the first time, the market began accurately to price subprime mortgage risk. What was extraordinary is what had happened leading up to October and November.
After saying he wanted "to be absolutely clear [that] as head of this firm, I take responsibility for performance," Mack took questions from the bank analysts of other Wall Street firms. It took this group a while to get to the source of embarrassment, but eventually they did. Four analysts elected not to probe Mack too closely about what was almost certainly the single greatest proprietary trading loss in Wall Street history, and then William Tanona, from Goldman Sachs, spoke:
The Big Short: Inside the Doomsday Machine Page 22
