Money and Power

by William D. Cohan

  ——

  ON NOVEMBER 1, 2010—nearly two and a half years later—Janet Broeckel, a Goldman managing director and associate general counsel, wrote to her counterpart at the Financial Crisis Inquiry Commission, the supposedly bipartisan commission looking into the causes of the financial crisis that began in 2007, to refute the suggestion that Goldman’s decision in 2007 to lower its marks exacerbated the problems at the two Bear Stearns hedge funds. In Broeckel’s letter, she challenged “the assertions” that Goldman’s lower marks “were in some way responsible for the ultimate failure of the BSAM funds” by causing one of the funds to lower its NAV by 12.5 percentage points. “Rather,” she wrote, “the revision in the Fund’s NAV, and the ultimate failure of the BSAM funds, was attributable to the market-wide decline in prices of mortgage-backed securities during this period, and the highly leveraged Fund’s concentrated holdings in those types of assets.” She wrote that by providing the “new information” at the request of the FCIC, the commission “will conclude that Goldman Sachs’ April 2007 month-end marks could not have caused the revision of the Fund’s NAV or the ultimate failure of the BSAM funds.”

  The gist of Broeckel and Goldman’s argument to the FCIC was that the hedge funds were already “in distress” by March 2007; that revisions to its marks in April could have, at most, caused a maximum of $26.3 million of downward revisions—and likely less—and, in any event, “could not have resulted in a 12.5 [percentage point] reduction in the NAV”; that BSAM did not, anyway, “mark its positions consistent with Goldman Sachs’ marks”; and that in a June 7, 2007, conference call with Goldman, Cioffi told Goldman’s employees that “three undisclosed dealers (not including Goldman Sachs) had significantly re-marked their April 2007 month-end marks”—marked “some positions down big,” Cioffi supposedly told Goldman—“and that this caused the BSAM funds to restate their April NAV.”

  Curiously, the three firms aren’t named, and there remains no credible evidence that any other firm on Wall Street was even remotely as aggressive—or accurate, by the way—during this time period as was Goldman with the marking of its CDO portfolio, which makes Goldman’s denials all the more perplexing. Under normal circumstances, Goldman would be crowing about its mark-to-market prowess and how it—alone—was being honest about the value of its portfolio (as can be seen in any number of its contemporaneous e-mails), but the postcrash politics of the situation forced Goldman to try to argue against its own considerable skills in favor of deflecting blame.

  Another of Goldman’s arguments was that it “had no incentive” to cause the failure of the funds because Goldman was a short-term lender to them. At the end of April 2007, Broeckel wrote, Goldman had extended to the funds $453 million in “repo loans,” secured by the funds’ mortgage-backed securities and so “simply put, it was against the financial interest of Goldman Sachs to cause the failure of the Fund and it did not do so.” She noted that had the fund not been able to repay its repo loans, Goldman would have seized the collateral—the mortgage-backed securities—and then sold it into the market and likely “would have suffered significant losses as the securities declined in value throughout the time period.” What Broeckel did not mention was that not only was this fact irrelevant, since Goldman was a secured lender, but that the funds did default on their repo loans, that Goldman did seize the collateral and threatened to sell it into the market, and that Bear Stearns then came to the rescue of the repo lenders, including Goldman and many others, by taking them out at 100 cents on the dollar—a decision that ultimately led to the collapse of Bear Stearns in March 2008.

  Broeckel’s letter did not mention—or include—Craig Broderick’s fateful May 11 e-mail about Goldman’s decision to lower its marks and convey that information to the market. Instead, the Goldman documents purport to show that Goldman’s marks barely changed at all during March, April, and May 2007 on the squirrelly securities that the Bear hedge funds had in their portfolios, despite what was going on at Goldman. For instance, an ABACUS 2006 HGS1 tranche that was valued at 65 cents in March was valued at 65 cents in May. Another tranche of the same security that was valued at 55 cents in April was valued at 55 cents in May. It’s as if the firm would rather appear to be just like every other firm rather than take a victory lap for seeing—and acting upon—what others didn’t and couldn’t.

  The actual timing itself—clearly it was May—was moot because the moment the bid-asked spread on these securities widened to the point where there needed to be such an extensive debate about their value was the beginning of the end for Bear Stearns. For its part, the Financial Crisis Inquiry Commission concluded: “Broderick was right about the impact of Goldman’s marks on clients and counterparties.” The FCIC’s report continued, contrary to what Broeckel, the Goldman attorney, hoped: “As the crisis unfolded, Goldman marked mortgage-related securities at prices that were significantly lower than those of other companies. Goldman knew that those lower marks might hurt those other companies—including some clients—because they could require marking down those assets and similar assets. In addition, Goldman’s marks would get picked up by competitors in dealer surveys. As a result, Goldman’s marks could contribute to other companies recording ‘mark-to-market’ losses: that is, the reported value of their assets could fall and their earnings would decline.”

  ——

  AFTER ITS SPIKE downward in February 2007 to around 60, the ABX index recovered back to the high 70s by mid-April 2007, before plunging again. The cost for Goldman to continue to hedge its risk against the collapse of the subprime index increased along with the general worry in the market. But Goldman gave Swenson and Birnbaum approval to keep hedging against the mortgage market, despite the rising cost of the insurance. On the morning of April 5, Deeb Salem wrote to Swenson with the idea of selling $200 million of protection against the ABX index. Seven minutes later, Sweeny replied to Salem: “Make that $500mm.” A week or so later, though, the market had moved again against Birnbaum’s short positions on the ABX index and showed a loss.

  Birnbaum and his colleagues kept riding this roller coaster for the next few months. Some days the shorts looked brilliant. Some days the shorts inflicted serious pain. Some days the message was ambiguous. For instance, on May 17, after some bad news about the performance of one CDO that Goldman had some long positions in, Deeb Salem wrote to Swenson that the “bad news” was the firm lost $2.5 million on the write-down of the long positions but the “good news” was that the firm had bought insurance on one of the same securities. “[W]e make $5mm,” Salem wrote.

  Birnbaum knew he would be proven right if he was just permitted to stick with the trade. He also knew he was right about the falling value of the residual CDO positions the firm was desperately trying to sell. And this meant more screaming from counterparties on the other side of these marks. “Every month brought more and more markdowns,” he explained. “But the original motivation for all this was very simple. It was: We owned some of this stuff, and there’s a discipline—if you mark the stuff at seventy you’re going to be much more likely to accept selling it at ninety-five than if you don’t. And we’re like, ‘Get this shit out of here because this stuff is worth nothing. And there’s going to be a window of opportunity to sell this stuff. Take the window. Take your losses and move on.’ ” This was the message that Birnbaum, Swenson, Salem, and Primer kept pounding away at with the senior executives at Goldman. Sometimes being right, though, does not matter. “From where I was sitting I just saw this thing as sort of like a comedic stalemate where the market was in denial for a long time about things still being worth par,” Birnbaum said. “We felt that it was worth a lot less. Somebody was gonna have to blink.”

  ——

  THAT MOMENT CAME at the beginning of June when the two Bear Stearns hedge funds were forced to revise their monthly performance numbers downward in the wake of the delivery of Goldman’s new marks. A June 7 letter to investors not only announced April’s 18.97 percent decline in th
e Enhanced Leverage Fund—just three weeks after a May 15 letter said the loss was 6.5 percent—but also announced the news that redemptions, of some $250 million on a $642 million fund, would be suspended because the “investment manager believes the company will not have sufficient liquid assets to pay investors.” On a conference call with investors the next day to discuss the fund’s poor performance, Cioffi and Tannin refused to answer investors’ questions. “They didn’t want to say anything,” one investor said. Inevitably, the increasingly angry investors in the fund put the word out about what was happening. On June 12, buried deep in the paper, the Wall Street Journal reported that the Enhanced Leverage Fund had fallen 23 percent in the first four months of the year and that redemptions from the fund had been blocked. “While the fund is down significantly, it is hard to tell what the actual losses will be because a few good trades could bring it back into the clear,” the Journal wrote. “Still, given the fund’s heavy exposure to this deteriorating corner of the mortgage market, in which many people are struggling to pay down their home loans, the news isn’t good.” The paper suggested that while the fund’s losses would be a “blow” to Cioffi and Tannin, the “paper losses will have a limited impact on Bear” because the firm only had $45 million invested in the fund.

  Ironically, that same day, BSAM was offering for sale $3.86 billion of the highest-rated mortgage-related securities in Cioffi’s hedge funds as a way to raise cash. That sale did not go well. A week later, Cioffi and BSAM ended up negotiating a number of bilateral agreements with the hedge funds’ lenders, including Goldman Sachs, whereby Goldman would take back its collateral and then attempt to sell it in the market. As part of the deal Bear Stearns reached with many of the lenders, Goldman would be made whole with either cash or securities. Among the collateral that Goldman took back was $300 million of the $400 million Timberwolf securities Cioffi had bought in March from Goldman at par. In an internal June 22 memorandum, Goldman’s mortgage trading desk spread the news that it now had to sell a $200 million slug of Timberwolf, at 98.5 cents on the dollar, and a $100 million slug at 95 cents on the dollar. Goldman was eager to sell the Timberwolf securities and advertised them as “Senior CDO axes” and as a “Super AAA Offering.” The trading desk described the Timberwolf securities as being senior in the capital structure to between 40 percent and 50 percent of other “Aaa/AAA bonds.” After receiving the memo, Montag wrote to Sparks asking for a “complete rundown of everything we bought from [BSAM] and what[’]s left?” Sparks offered to get Montag “a complete summary with details” but confirmed that the “main thing left is [$]300mm [T]imberwolfs” as well as “some small [RMBS]”—residential mortgage-backed securities—“positions.” Everything else had been sold. Within minutes, Montag replied, “[B]oy that [T]imberwo[l]f was one shitty deal”—giving Senator Levin the cudgel he used to bash Goldman for eleven hours at the end of April 2010. A week later, the Timberwolf securities still had not sold.

  The two Bear Stearns hedge funds were officially liquidated on July 30. Investors in the funds lost around $1.5 billion. Since Bear had become the short-term lender to the funds on June 22—replacing Goldman, among others—when the funds were liquidated, Bear Stearns seized $1.3 billion of underlying collateral, which it eventually wrote down in the fourth quarter of 2007, leading to the first quarterly loss in the firm’s eighty-five-year history. Bear Stearns collapsed in March 2008 and was scooped up by JPMorgan Chase for $10 a share, after having reached an all-time high of $172.69 in January 2007. To get the JPMorgan deal with Bear done, American taxpayers agreed to absorb losses on $29 billion of “toxic securities” that JPMorgan did not want. As of September 30, 2010, those securities were worth $27 billion, according to the New York Fed.

  ——

  FOR BIRNBAUM, THE trouble at the Bear Stearns hedge funds was a sweet symphony. “Once the Cioffi news came there was really a couple of us on the desk who were like, ‘Okay, that’s the Come-to-Jesus moment. It’s gonna be a big mark-to-market event here. We’re gonna go for it again—meaning short the market again.’ We had covered a lot of our shorts because the VAR police found us—they caught up to us and we did that in March. We covered, meaning we effectively gave a lot of our positions to hedge-fund managers, gave them our shorts, and they benefited. Many homes in Italy were purchased off of those shorts.” Despite being forced to sell off its profitable short bets in March, Birnbaum’s desk still killed it during the second quarter of 2007. According to internal Goldman documents, his group made $457 million in profit for the quarter, up from $288 million in profit in the first quarter. Indeed, the structured products trading group seemed to be carrying the overall mortgage department, which managed to lose $174 million in the second quarter—despite Birnbaum’s profitable contribution—in large part due to the write-downs on Goldman’s long positions in mortgages in order to sell them off into the market.

  In early July, in part because the group’s VAR had been reduced dramatically in the second quarter, Birnbaum moved quickly to bet the ABX would fall in the wake of the Cioffi hedge-fund debacle. He did not have to seek the approval of the senior executives to make the bet. “We’re given a set of quantitative parameters, a box in which we’re supposed to play from a risk standpoint,” he explained. “And as long as we stay within those parameters, you can do what you want. We were well within our box. So we started shorting the market again.” He sold the ABX index like crazy. And by July 12, the bet was already paying off, a fact that can be seen even though an e-mail started by Birnbaum that day has been heavily redacted by the U.S. Senate subcommittee that released it. He sent around the results of the trading of the ABX index to [redacted] at noon that day. Five minutes later, he got a reply: “That’s the way to make my markets, [J]oshua, that’s my boy!” Birnbaum responded: “Seen massive flows recently. Many accounts ‘throwing in the towel.’ Anybody who tried to call the bottom left in bodybags.” Then came the reply: “We hit a bilsky”—a billion dollars—“in PNL today.… I’m not [J]ohn [P]aulson though,” causing Birnbaum to reply: “He’s definitely the man in this space, up [$]2–3 bil[lion] on this trade. We were giving him a run for his money for a while but now are a definitive #2.”

  By July 20, the profits Birnbaum was racking up caught the attention of Blankfein. He asked Viniar and Cohn to explain why the numbers were so big. After some back and forth in a fairly incomprehensible internal jargon about hedges here and there, Cohn explained to Blankfein that basically the firm had a “net short” on the mortgage market that was paying off. “Bet all the dads at camp are talking about the same stuff,” Blankfein responded to Cohn. On July 24, the daily P&L showed Birnbaum’s secondary trading group up $72 million for the day, and overall the firm had made $9 million in pretax earnings. Blankfein e-mailed Viniar and Cohn. “I’ve seen worse,” he wrote, with some understatement. Viniar replied, “Mergers, overnight [A]sia and especially short mortgages saved the day.”

  The next day was even better. Birnbaum’s trading bet was up $373 million. Goldman then used the cover of Birnbaum’s extraordinary profits to write down further the value of the CDOs the firm retained, and could not sell, as well as other mortgage-related residuals. Together, the write-offs in these long positions cost the firm $322 million, but because of Birnbaum’s short position, the mortgage group was still able to show a profit for the day of $48.7 million. “Tells you what might be happening to people who don’t have the big short,” Viniar wrote to Cohn that day. On July 26, Montag wrote to Blankfein and Cohn that “mortgages goi[n]g to show up [$]135 [million] or so today it seems.” Blankfein responded: “Is that right?” And then Montag increased the estimate to $170 million “hopefully.” Blankfein wondered, “I assume we are properly marking down our longs?” To which Montag responded that the firm had “marked things down [$]100 [million] yesterday. Could have done [$]15 [million] or more today but don’t know.” He added the thought that there was “not a lot left” to mark down. No doubt pleased with the firm’s good fortune, Bl
ankfein replied, “If the shorts went up today, shouldn’t the longs have dropped (unless they’re already at zero … [?]).”

  Birnbaum understood why the firm would net its mortgage losses against his gains, but it still rankled. “The firm was a great beneficiary of the idea that ‘it’s very easy to have a clear head and to look at something in an unbiased way when marking things correctly isn’t going to hurt you,’ ” he said. “Right? Because our desk was as short as it was and we were making money on the way down, we effectively had all this dry powder in terms of P&L to play with.” Another trader echoed this sentiment: “If we’re making $50 million in one day on our desk, it was not unusual to say, ‘Okay, we got $50 million to play with. These CDOs, let’s mark them down further.’ And it would drive us crazy because, you know, we’d be high-fiving, going, ‘We made 100 million bucks today.’ I mean it was unheard of in one day to make $100 million. And then when the P&Ls would actually go through, you’d be like, ‘Oh, the department’s up $20 million.’ ”

  This set Goldman apart in the market and caused a fair amount of resentment at other firms that resisted marking their books to the real market, because doing so would mean having to absorb some serious losses. “Other firms didn’t have that luxury,” Birnbaum explained. “Everybody else couldn’t afford to mark this stuff ’cause they had so much of this shit, they couldn’t mark it.… They were in denial. If they took the full extent of our marks, these guys would have failed earlier. Everyone was trying to buy time. Just what any trader who’s losing money would try to do.”