The Financial Crisis five Years Later: How Goldman Sachs Made Money Mid-Crisis


April 27, 2010: Goldman Sachs CEO Lloyd Blankfein appears before the Senate

Of Lloyd Blankfein’s 3 hours and 28 minutes before the U.S. Senate’s permanent subcommittee on investigations on the afternoon of April 27, 2010, the most memorable moment came when Democratic Senator Carl Levin of Michigan, for the umpteenth time, held up an e-mail that had been written nearly three years earlier by two of Goldman Sachs’s (GS) most senior traders. The e-mail described a Goldman-underwritten collateralized-debt obligation, or CDO, as “one sh-‍-‍-y deal.” It was the end of a long day, and as Levin bore down on Blankfein, he wanted to know if it was ethical for Goldman to sell a security that its traders thought was bad while Goldman, as a principal, bet against those very same securities in order to make a profit.

It was not the chief executive officer’s finest response. He winced. He perseverated. He parsed. He looked uncomfortable. Finally, he lamely defended Goldman’s behavior. “In the context of market-making, that is not a conflict. What the clients are buying, or customers are buying, is—they are buying an exposure. The thing that we are selling to them is supposed to give them the risk they want.” Levin at first looked confused, then exasperated, as Blankfein insisted that what Goldman thought of a deal was immaterial: Clients “are not coming to us to represent what our views are.”

As public relations, Blankfein’s performance was uneven, and it could have been a lot worse. If only Levin had dug out of the 900 pages of Goldman documents released that same day another three-year-old e-mail, one written by Craig Broderick, Goldman’s chief risk officer1 and one of Blankfein’s top lieutenants. Broderick’s memo, sent on May 11, 2007, revealed in terse, unemotional prose how Goldman planned to keep the firm in the black while many other banks went down the tubes. He noted that Goldman’s decision to mark down the prices on its portfolio of derivatives such as CDOs and synthetic CDOs “will potentially have a big [profit and loss] impact on us, but also to our clients. … We need to survey our clients and take a shot at determining the most vulnerable clients, knock on implications, etc. This is getting lots of 30th floor attention right now.”

What Broderick did not say in his e-mail was that Goldman was net-short the mortgage market by May 2007 and—alone among Wall Street firms—stood to benefit by marking down the billions of hard-to-value and thinly traded mortgage- and mortgage-related securities on its balance sheet. Goldman would prove to be correct about the value of these securities—they were worth far less than most of Wall Street was saying they were. But its decision to mark them down aggressively, as Broderick indicated, set off a chain of events among its competitors that would exacerbate their demise a year later and lead to Goldman cashing in—legally, if not morally—at their expense.

Although the old Wall Street adage, “Success has many fathers; failure is an orphan,” applies to Goldman Sachs’s outrageously profitable navigation of the 2008 crisis, there’s little doubt that Broderick, now 54, was and remains the man largely responsible for keeping the firm from blowing itself up. With the approval of top management—including Blankfein, then only six months on the job, and longtime Chief Financial Officer David Viniar—Goldman followed a strategy that Broderick and others began to design in the second half of 2006, when they made a huge bet (the firm prefers the term “hedge”) that the mortgage market would collapse. As a result, Goldman had its most profitable year to date in 2007, earning $17.6 billion pretax, including $4 billion from the hedge. In 2009, with most of its competition anesthetized, Goldman did even better, earning almost $20 billion pretax.

“I’ve predicted 10 of the last 3 downturns”

The man behind Goldman’s prescience is no Lex Luthor. He’s an athletic, clean-cut, self-deprecating graduate of the College of William and Mary, where he met and married his sweetheart, Camille. The Brodericks graduated in the class of 1981: He studied economics; she studied biology. They have two teenage sons, Mark and Sean. He denies having any special predictive powers. “Being a risk person, I’ve predicted 10 of the last 3 downturns,” he says. He’s also unapologetic about doing what he says needed to be done to prevent Goldman from suffering the fate of its competitors.

As he watched the markets in 2006, Broderick says he couldn’t believe “how buoyant they felt, how low volatility was, how tight credit spreads were, how aggressive credit terms were, and how much leverage was being built into the system in different ways.” While “it was easy to draw the conclusion that we were in an exceptionally—and probably an unjustifiably—buoyant time,” he says, “it was a whole lot harder to determine when you were at risk of having this situation reverse, and how severely it was likely to reverse. I was surprised personally at how long it continued, but then I was also very surprised at how vicious, and how broad-based, the overall correction process was. I don’t make any pretense about having a real sense as to just how violent the correction would be.”


Photograph by Andrew Harrer/BloombergBroderick

There was no single moment when it dawned on Broderick that it all might crumble. Rather, he says, it was a slow recognition that market players were taking on more and more risk without properly compensating for it. By December 2006 a group of Goldman traders and executives, including Broderick, decided to “get closer to home,” in Viniar’s words, during a meeting in Viniar’s 30th floor conference room at 85 Broad Street, Goldman’s former headquarters. They’d do this by creating a complex series of trades that would vastly reduce Goldman’s long exposure to the mortgage market and position the firm, in the course of six months, to be net-short, thus benefiting financially if the mortgage market collapsed. The wager worked magnificently. On July 25, 2007, for instance, Goldman’s bet resulted in a one-day profit of $51 million. “Tells you what might be happening to people who don’t have the big short,” Viniar wrote in an e-mail that day to Blankfein and Gary Cohn, Goldman’s president.

The firm had other things going for it as the crisis deepened. There was its long-standing “mark-to-market” discipline, practiced by few other firms, in which traders valued their trades precisely where they could be sold in the market, rather than at some amorphous value based on a financial model or someone’s idea of where a security should trade (so-called marking “to myth”). While the mark-to-market convention forced Goldman to take losses on various complex securities before other banks, it also gave the firm an early, and accurate, sense of trouble. Broderick says Goldman’s investment in systems and technology, totaling billions of dollars, allowed the firm to rapidly calculate its financial exposure to various markets and financial instruments. The proprietary, real-time securities database, known internally as SecDB, tracks all the trades that Goldman makes and constantly assesses the bank’s exposure. No other firm came close to having that level of real-time information about its risk.

Unlike Citigroup (C), Goldman also refrained from off-balance-sheet investment vehicles. Relying minimally on short-term secured financing, the firm had built up a reserve of $50 billion to $60 billion in liquid assets—much of it accumulated through costly long-term borrowings—that it expected to help weather an inevitable run on the banks. (These days the reserve is closer to $160 billion to $180 billion, well above its capital requirements.) “It turned out that was absolutely essential to our ability to not only survive the crisis itself but to be well positioned to provide the market-making functions and other services that our clients absolutely needed,” Broderick says.

Whether Goldman could have gone the way of Lehman Brothers or Merrill Lynch (BAC) remains the subject of much debate. Goldman maintains that it did not need, or want, the $10 billion bailout that Hank Paulson pushed on it and other firms in October 2008. But the fact remains that when the Federal Reserve allowed Goldman and Morgan Stanley (MS)—but not Lehman Brothers—to become bank holding companies on Sept. 21, 2008, Goldman was able, three days later, to raise $10 billion in equity, $5 billion from the public and another $5 billion from investor Warren Buffett. That would probably not have happened without the Fed’s expedited decision and support. (A week later, Morgan Stanley saved itself from bankruptcy when it negotiated a $9 billion equity investment from Mitsubishi UFJ Financial Group (MTU).)

Broderick did not sleep well during 2008. “I’d kick myself over decisions made previously and wish we’d done different ones,” he says. “I’d be remiss as a risk manager to be satisfied with the combined industry’s risk management practices over this period.”

Unlike Blankfein and seven of his Goldman colleagues, Broderick was never called to testify before Levin, nor was he asked to appear before the House of Representatives during its probe of financial executives. He did go before the Financial Crisis Inquiry Commission, testifying that Goldman was served well during the crisis by its long-standing focus on putting risk management on a par with the revenue-generating parts of the firm. “The nature of our role as a financial intermediary means that we take this risk willingly, but only subject to basic principles which define our overall approach to prudent risk management,” he said.

Some interpreted Broderick’s e-mail on May 11, 2007, as Goldman driving down the price for many illiquid and hard-to-value mortgage securities at a time when the firm had implemented its “big short”—thereby sticking it to those who were exposed. A range of Goldman’s counterparties, from Bear Stearns to American International Group (AIG), have argued subsequently that in 2007 Goldman began manipulating the price of these illiquid securities, knowing full well that it alone was in a position to benefit because of its short position. Competitors contend that Goldman’s marked-down prices exacerbated their financial problems by forcing them to lower the value of these securities on their books, vastly reducing their equity and calling into question their financial viability. It’s a complex conspiracy theory, fed oxygen by those who assume Goldman is evil.

Asked about the May 11 e-mail, Broderick says he doesn’t recall it specifically. “We always aspire not to kid ourselves or our clients as to what the right price for any given security or product or transaction is in the market,” he says. “That sometimes is painful to do. It’s painful for us. It’s painful for our clients. But ultimately, incremental, timely adjustments resulting in incremental, timely collateral calls or recognition of profit or loss, or other signaling mechanisms and actions that result from it, are absolutely in the best interest of the market overall. And the contrary, which is to pretend like the market hasn’t moved and not take the loss or not charge your clients for the loss in terms of collateral, is ultimately far more destructive.”

Broderick has little patience for the conspiracy theory, although he’s well aware of it. “We mark on a basis that is overseen by our controllers group,” he says. “We have no ax”—Goldman-speak for having a financial incentive to sell trading positions—“in the process. They are determined to get the right prices measured by where you can actually trade. And if that’s where you can trade right now, then that’s what the price is.”

With Goldman’s zealous devotion to monitoring risk—Blankfein likes to say he spends “98 percent of his time worrying about things with a 2 percent probability”—Broderick ought to be less worried. But he can’t be. “There’s always more risks,” he says.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s