The Big Short: Inside the Doomsday Machine

by Michael Lewis

  Yet this did not happen. Instead, between February and June of 2007, big Wall Street firms, led by Merrill Lynch and Citigroup, created and sold $50 billion in new CDOs. "We're totally baffled," said Charlie. "Because everyone and everything just goes back to normal, even though it obviously wasn't normal. We knew the collateral for the CDOs had collapsed. And yet everything went on, as if nothing had changed."

  It was as if an entire financial market had tried to change its mind--and then realized that it could not afford to change its mind. Wall Street firms--most notably Bear Stearns and Lehman Brothers--continued to publish bond market research reaffirming the strength of the market. In late April, Bear Stearns held a CDO conference, into which Charlie sneaked. On the original agenda was a presentation entitled "How to Short a CDO." It had been removed from the final conference--so, too, had been the slides that accompanied the talk that had been posted on the Bear Stearns Web site. Moody's and S&P flinched, too, but in a telling manner. In late May, the two big rating agencies announced that they were reconsidering their subprime bond ratings models. Charlie and Jamie hired a lawyer to call Moody's and ask them, if they were going to rate subprime bonds by different criteria going forward, might they also reconsider the two trillion dollars' worth or so of bonds they had already rated, badly. Moody's didn't think that was a good idea. "We were like, 'You don't have to re-rate all of them. Just the ones we're short,'" said Charlie. "They were like, 'Hmmmmmm...no.'"

  To Charlie and Ben and Jamie it seemed perfectly clear that Wall Street was propping up the price of these CDOs so that they might either dump losses on unsuspecting customers or make a last few billion dollars from a corrupt market. In either case, they were squeezing and selling the juice from oranges that were undeniably rotten. By late March 2007, "We were pretty sure one of two things was true," said Charlie. "Either the game was totally rigged, or we had gone totally fucking crazy. The fraud was so obvious that it seemed to us it had implications for democracy. We actually got scared." They both knew reporters who worked at the New York Times and the Wall Street Journal--but the reporters they knew had no interest in their story. A friend at the Journal hooked them up with the enforcement division of the SEC, but the enforcement division of the SEC had no interest either. In its lower Manhattan office, the SEC met with them and listened, but politely. "It was almost like a therapy session," said Jamie. "Because we sat down and said, 'We've just had the most crazy experience.'" As they spoke, they sensed the audience's incomprehension. "We probably had like this wild-eyed we've-been-up-for-three-days-straight look in our eyes," said Charlie. "But they didn't know anything about CDOs, or asset-backed securities. We took them through our trade but I'm pretty sure they didn't understand it." The SEC never followed up.

  Cornwall had a problem more immediate than the collapse of society as we know it: the collapse of Bear Stearns. On June 14, 2007, Bear Stearns Asset Management, a CDO firm, like Wing Chau's, but run by former Bear Stearns employees who had the implicit backing of the mother ship, declared that it had lost money on bets on subprime mortgage securities and that it was being forced to dump 3.8 billion dollars' worth of these bets before closing the fund. Up until this moment, Cornwall Capital had been unable to see why Bear Stearns, and no one else, had been so eager to sell them insurance on CDOs. "Bear was able to show us liquidity in the CDOs that I couldn't understand," said Ben. "They had a standing buyer on the other side. I don't know that our trades went directly into their funds, but I don't know where else they would have gone."

  And therein lay a new problem: Bear Stearns had sold Cornwall 70 percent of its credit default swaps. Because Bear Stearns was big and important, and Cornwall Capital was a garage band hedge fund, Bear Stearns hadn't been required to post collateral to Cornwall. Cornwall was now totally exposed to the possibility that Bear Stearns would be unable to pay off its gambling debts. Cornwall Capital couldn't help but notice that Bear Stearns was not so much shaping the subprime mortgage bond business as being reshaped by it. "They'd turned themselves from a low-risk brokerage operation into a subprime mortgage engine," said Jamie. If the subprime mortgage market crashed, Bear Stearns was going to crash with it.

  Back in March, Cornwall had bought $105 million in credit default swaps on Bear Stearns--that is, they'd made a bet on the collapse of Bear Stearns--from the British bank HSBC. If Bear Stearns failed, HSBC would owe them $105 million. Of course this only shifted their risk to HSBC. HSBC was the third largest bank in the world, and one of those places it was hard to think about going down. On February 8, 2007, however, HSBC rocked the market with the announcement that it was taking a big, surprising loss on its portfolio of subprime mortgage loans. It had entered the U.S. subprime lending business in 2003, when it had bought America's biggest consumer lending operation, Household Finance. The same Household Finance that had pushed Steve Eisman over the narrow border between Wall Street skeptic and Wall Street cynic.

  From the social point of view the slow and possibly fraudulent unraveling of a multi-trillion-dollar U.S. bond market was a catastrophe. From the hedge fund trading point of view it was the opportunity of a lifetime. Steve Eisman had started out running a $60 million equity fund but was now short around 600 million dollars' worth of various subprime-related securities, and he wanted to short more. "Sometimes his ideas cannot be manifested in a trade," said Vinny. "This time they could." Eisman was enchained, however, by FrontPoint Partners and, by extension, Morgan Stanley. As FrontPoint's head trader, Danny Moses found himself caught in the middle, between Eisman and FrontPoint's risk management people, who didn't seem to completely understand what they were doing. "They'd call me and say, 'Can you get Steve to take some of this off?' I'd go to Steve and Steve would say, 'Just tell them to fuck off.' And I'd say, 'Fuck off.'" But risk management hounded them, and cramped Eisman's style. "If risk had said to us, 'We're very comfortable with this and you can do ten times this amount,'" said Danny, "Steve would have done ten times the amount." Greg Lippmann was now blasting Vinny and Danny with all sorts of negative information about the housing market, and, for the first time, Vinny and Danny began to hide the information from Eisman. "We were worried he'd come out of his office and shout, 'Do a trillion!'" said Danny.

  In the spring of 2007, the subprime mortgage bond market, incredibly, had strengthened a bit. "The impact on the broader economy and the financial markets of the problems in the subprime markets seems likely to be contained," U.S. Federal Reserve chairman Ben Bernanke was quoted as saying in the newspapers on March 7. "Credit quality always gets better in March and April," said Eisman. "And the reason it always gets better in March and April is that people get their tax refunds. You would think people in the securitization world would know this. And they sort of did. But they let the credit spreads tighten. We just thought that was moronic. What are you, fucking stupid?" Amazingly, the stock market continued to soar, and the television over the FrontPoint trading desks emitted a ceaselessly bullish signal. "We turned off CNBC," said Danny Moses. "It became very frustrating that they weren't in touch with reality anymore. If something negative happened, they'd spin it positive. If something positive happened, they'd blow it out of proportion. It alters your mind. You can't be clouded with shit like that."

  Upon their return from Las Vegas, they set out to pester the rating agencies, and the Wall Street people who gamed their models, for more information. "We were trying to figure out what, if anything, would make the ratings agencies downgrade," said Danny. In the process, they picked up more disturbing tidbits. They'd often wondered, for instance, why the rating agencies weren't more critical of bonds underpinned by floating-rate subprime mortgages. Subprime borrowers tended to be one broken refrigerator away from default. Few, if any, should be running the risk of their interest payment spiking up. As most of these loans were structured, however, the homeowner would pay a fixed teaser rate of, say, 8 percent for the first two years, and then, at the start of the third year, the interest rate would skyrocket to, say, 12 perc
ent, and thereafter it would float at permanently high levels. It was easy to understand why originators like Option One and New Century preferred to make these sorts of loans: After two years the borrowers either defaulted or, if their home price had risen, refinanced. To them the default was a matter of indifference, as they kept none of the risk of the loan; the refinance was merely a chance to charge the borrower new fees. Bouncing between the rating agencies and people he knew in the subprime bond packaging business, Eisman learned that the rating agencies simply assumed that the borrower would be just as likely to make his payments when the interest rate on the loan was 12 percent as when it was 8 percent--which meant more cash flow for the bondholders. Bonds backed by floating-rate mortgages received higher ratings than bonds backed by fixed-rate ones--which was why the percentage of subprime mortgages with floating rates had risen, in the past five years, from 40 to 80.

  A lot of these loans were now going bad, but subprime bonds weren't moving--because Moody's and S&P, disturbingly, still hadn't changed their official opinions of them. As an equity investor, FrontPoint Partners was covered by Wall Street stockbrokers. Eisman asked stock market salesmen at Goldman Sachs and Morgan Stanley and the others to bring over the bond people for a visit. "We always asked the same question," says Eisman. "'Where are the ratings agencies in all this?' And I'd always get the same reaction. It was a physical reaction because they didn't want to say it. It was a smirk." Digging deeper, he called S&P and asked what happened to default rates if real estate prices fell. The man at S&P couldn't say: Their model for home prices had no ability to accept a negative number. "They were just assuming home prices would keep going up," says Eisman.*

  Eventually he'd hop onto the subway with Vinny and ride down to Wall Street to meet with a woman at S&P named Ernestine Warner. Warner worked as an analyst in the surveillance department. The surveillance department was meant to monitor subprime bonds and downgrade them if the loans that underpinned them went bad. The loans were going bad but the bonds weren't being downgraded--and so once again Eisman wondered if S&P knew something he did not. "When we shorted the bonds, all we had was the pool-level data," he said. The pool-level data gave you the general characteristics--the average FICO scores, the average loan-to-value ratios, the average number of no-doc loans, and so forth--but no view of the individual loans. The pool-level data told you, for example, that 25 percent of the home loans in some pool were insured, but not which loans--the ones likely to go bad or the ones less likely to. It was impossible to determine how badly the Wall Street firms had gamed the system. "We of course thought that the ratings agencies had more data than we had," said Eisman. "They didn't."

  Ernestine Warner was working with the same rough information available to traders like Eisman. This was insane: The arbiter of the value of the bonds lacked access to relevant information about the bonds. "When we asked her why," said Vinny, "she said, 'The issuers won't give it to us.' That's when I lost it. 'You need to demand to get it!' She looked at us like, We can't do that. We were like, 'Who is in charge here? You're the grown-up. You're the cop! Tell them to fucking give it to you!!!'" Eisman concluded that "S&P was worried that if they demanded the data from Wall Street, Wall Street would just go to Moody's for their ratings."*

  As an investor, Eisman was allowed to listen in on the quarterly conference calls held by Moody's, but not invited to pose questions. The people at Moody's were sympathetic to his need for more genuine interaction, however; and the CEO, Ray McDaniel, even invited Eisman and his team to his office for a visit, a gesture that forever endeared him to Eisman. "When are shorts welcome anywhere?" asked Eisman. "When you're short, the whole world is against you. The only time a company met me with complete knowledge that we were short was Moody's." After their trip to Las Vegas, Eisman and his team were so certain the world had been turned upside down that they just assumed Raymond McDaniel must know it, too. "But we're sitting there," recalls Vinny, "and he says to us, like he actually means it, 'I truly believe that our ratings will prove accurate.'" And Steve shoots up in his chair and asks, 'What did you just say?'--as if the guy had just uttered the most preposterous statement in the history of finance. He repeated it. And Eisman just laughed at him. "With all due respect, sir," said Vinny deferentially, as they left, "you're delusional." This wasn't Fitch or even S&P. This was Moody's. The aristocrats of the rating business, 20 percent owned by Warren Buffett. And its CEO was being told he was either a fool or a crook, by Vincent Daniel, from Queens.

  By early June the subprime mortgage bond market had resumed what would become an uninterrupted decline, and the FrontPoint positions began to move--first by thousands and then by millions of dollars a day. "I know I'm making money," Eisman would often ask. "So who is losing money?" They already were short the stocks of mortgage originators and the home builders. Now they added to their short positions in the stocks of the rating agencies. "They were making ten times more rating CDOs than they were rating GM bonds," said Eisman, "and it was all going to end."

  Inevitably, their attention turned to the beating heart of capitalism, the big Wall Street investment banks. "Our original thesis was that the securitization machine was Wall Street's big profit center and it was going to die," said Eisman. "And when that happened, their revenues would dry up." One of the reasons Wall Street had cooked up this new industry called structured finance was that its old-fashioned business was every day less profitable. The profits in stockbroking, along with those in the more conventional sorts of bond broking, had been squashed by Internet competition. The minute the market stopped buying subprime mortgage bonds and CDOs backed by subprime mortgage bonds, the investment banks were in trouble. Right up until the middle of 2007, Eisman had not suspected that the firms were so foolish as to invest in their own creations. He could see that their leverage had increased dramatically, in just the past few years. He could of course see that they were holding more and more risky assets with borrowed money. What he could not see was the nature of their assets. Triple-A-rated corporate bonds, or triple-A-rated subprime CDOs? "You couldn't know for sure," he said. "There was no disclosure. You didn't know what they had on their balance sheet. You naturally assumed that they got rid of this shit as soon as they created it."

  A combination of new facts, and actual human contact with the people who ran the big firms and the rating agencies, had stirred his suspicion. The first new fact had been HSBC's announcement, in February 2007, that it was losing a lot of money on its subprime loans, and a second announcement, in March, that it was dumping its subprime portfolio. "HSBC were supposed to be the good guys," said Vinny. "They were supposed to have cleaned up Household. We thought, Holy crap, there are so many people worse than that." The second new fact was in Merrill Lynch's second-quarter results. In July 2007, Merrill Lynch announced yet another sensationally profitable quarter, but admitted it had suffered a decline in revenues from mortgage trading due to losses in subprime bonds. What sounded to most investors like a trivial piece of information was to Eisman the big news: Merrill Lynch owned a meaningful amount of subprime mortgage securities. Merrill's CFO, Jeff Edwards, told Bloomberg News that the market need not worry about this, as "active risk management" had allowed Merrill Lynch to reduce its exposure to the lower-rated subprime bonds. "I don't want to get too deep into exactly how we positioned ourselves at any one point in time," Edwards said, but went deep enough to say that the market was paying too much attention to whatever Merrill happened to be doing with subprime mortgage bonds. Or, as Edwards elliptically put it, "There's a disproportionate focus on a particular asset class in a particular country."

  Eisman didn't think so--and two weeks later persuaded a UBS analyst named Glenn Schorr to escort him to a small meeting between Edwards and Merrill Lynch's biggest shareholders. The Merrill CFO began by explaining that this little subprime mortgage problem Merrill Lynch seemed to have was firmly under the control of Merrill Lynch's models. "We're not that far into the meeting," said someone who was there. "Jeff i
s still giving his prepared remarks and Steve just bursts out, 'Well, your models are wrong!' This very awkward silence comes over the room. Do you laugh? Do you try to think up some question so everyone can move on? Steve was sitting at the end of the table and he starts to put his papers in order really conspicuously--as if to say, 'If it wasn't rude, I'd walk out now.'"

  Eisman, for his part, considered the event a polite exchange of views, after which he lost interest. "There was nothing more to say. I just figured, You know what? This guy doesn't get it."

  On the surface, these big Wall Street firms appeared robust; below the surface, Eisman was beginning to think, their problems might not be confined to a potential loss of revenue. If they really didn't believe the subprime mortgage market was a problem for them, the subprime mortgage market might be the end of them. He and his team now set about searching for hidden subprime risk: Who was hiding what? "We called it The Great Treasure Hunt," he said. They didn't know for sure if these firms were in some way on the other side of the bets he'd been making against subprime bonds, but the more he looked, the more sure he became that they didn't know either. He'd go to meetings with Wall Street CEOs and ask them the most basic questions about their balance sheets. "They didn't know," he said. "They didn't know their own balance sheets." Once, he got himself invited to a meeting with the CEO of Bank of America, Ken Lewis. "I was sitting there listening to him. I had an epiphany. I said to myself, 'Oh my God, he's dumb!' A lightbulb went off. The guy running one of the biggest banks in the world is dumb!" They shorted Bank of America, along with UBS, Citigroup, Lehman Brothers, and a few others. They weren't allowed to short Morgan Stanley because they were owned by Morgan Stanley, but if they could have, they would have. Not long after they established their shorts against the big Wall Street banks, they had a visit from a prominent analyst who covered the firms, Brad Hintz, at Sanford C. Bernstein & Co. Hintz asked Eisman what he was up to.