Birnbaum immersed himself in trading mortgages. “I traded pretty much every seat in that business,” he said. “It gave me a ton of exposure with accounts and sales forces around the world, with flow trading. It’s a unique skill set all its own. I remember one of the guys on the desk early in my career said of trading, ‘The best analogy I can make is to quote from Chuck Yeager: the more sorties you do, the better you are, and you can’t underestimate the value of experience.’ Well, the more trades you’ve done, the more times you’ve been put in the line of fire, you’re just going to be that much better than the new guy that comes in.” Birnbaum also began to get exposure to the most senior executives of the firm, especially to Jon Corzine. Corzine would come around and chat with the mortgage traders, including Birnbaum, on a regular basis. “He’d just reach out and chat with the mortgage traders for long periods of time,” he said of Corzine, “and I thought that was pretty cool that the head of the firm thought our area was that important.”
Goldman soon promoted Birnbaum to a full-fledged associate after the end of his two-year post-college stint as an analyst. In 1998, Birnbaum became “intrigued,” as he said, with the growing intersection between the mortgage market and what was known as the “swaps market,” which traditionally had been for those investors who had a fixed-rate debt instrument and for whatever reasons wanted a floating-rate debt instrument. In the swaps market—for a fee—an investor could swap the fixed for the floating and everybody would be happy. “But you could kind of swap anything,” he explained. “One guy wants one thing, another guy wants another thing.” In mortgages, the perennial issue had always been what the prepayment rate on mortgages would turn out to be as people inevitably sought to refinance their home loans. “So there was an idea: ‘Hey, why don’t we create some swaps that are based on the realized prepayment experience of a certain type of mortgage?’ And that was the first time that the swaps concept was applied to mortgages, and ultimately, fast-forward the clock all the way to the CDS [credit-default swaps] market, that was based on credit and that was what ultimately brought down AIG and other places like that. But the first time that swaps appeared in mortgages was based on betting on prepayments. I thought that was a very interesting business.”
At first, Birnbaum did deals between those counterparties who were willing to bet prepayments would be low and those willing to bet prepayments would be high. That was a nice business. But things got much more interesting and lucrative when Birnbaum started to come up with some innovative, interesting new structures. One of these he dubbed a “synthetic CMO,” based upon certain investors wanting one part of a CMO but not another. Some investors found some parts of the CMO cheaper than other parts based upon the collateral associated with it. Birnbaum exploited these differences. “Our concept was why don’t we isolate a portion of the CMO deal that we view as being expensive and why don’t we just facilitate the underwriting to that and take that risk instead of selling all the parts,” he said. Whatever Birnbaum had come up with precisely, Goldman made a lot of money from it, and Birnbaum’s career began to take off.
He was soon promoted to be head of Goldman’s swaps business; as that did increasingly well, he was promoted further to be head of Goldman’s mortgage derivatives business. Four years into his tenure at Goldman, he was promoted to vice president, like every other person who had been at the firm for four years. “The VP designation is typically robotic,” he said. “You don’t get it earlier if you’re killing it or later if you’re underperforming. You just get it kind of in four years.” Goldman was trying—it seemed to him—to distinguish his pay from that of his peer group. “They were doing their job and paid me just enough to keep me around,” he said.
In 2001, he came up with another new innovation: something he called “CMM,” for constant maturity mortgage, a kind of interest-rate product tied to mortgage rates, instead of LIBOR (for London Interbank Offered Rate). A CMM was “an attempt to simplify the trading of mortgage price risk by transforming the most liquid portion of the mortgage market into a rate-based market” and can “be used to hedge for mortgage products that are sensitive to changes in mortgage rates.” Regardless of whether mere mortals understood the new product (or not), Birnbaum’s point about it was “that it was another innovation that did well and made good money doing it,” of around $100 million a year or so. At one point, in July 2002, Birnbaum, along with Goldman colleague Ashwin Rao, published “A Simple Algorithm to Compute Short-Dated CMM Forwards,” which of course was anything but simple. Still, the short paper, filled with such complex ideas as “PC (x) = PC(E[x]) + DurC (E[x]) * ∆x + ½ * ConvC(E[x]) * (∆x)2,” or, roughly translated, as Birnbaum’s calculation that mortgages prices were “purely quadratic functions of the level of swap rates,” must have impressed Goldman’s clients.
In the thirteen years Birnbaum had been at Goldman, much had changed at the firm and on Wall Street. Inside 85 Broad Street the firm had been transformed from one where investment bankers held sway to one where traders, and trading, were the dominant force in the building. The 1999 IPO accelerated this trend because the money raised gave the firm even more capital with which to trade. What’s more, it was other people’s money with the extra benefit that the rewards—in the billions of dollars annually—for taking risks with it went mostly to the Goldman employees and the losses, should anything go wrong, belonged to the firm’s shareholders and creditors.
During Birnbaum’s career at Goldman, Wall Street had changed in other ways as well. Whereas in 1992, LBO firms, or leveraged-buyout firms, were few and far between—once you got beyond the well-known firms such as KKR and Forstmann Little—by 2005, buyouts firms, now known as private-equity firms, were everywhere. Goldman, which had a $1 billion fund in 1992, was on its fifth fund by 2005, with $8.5 billion to invest. (Goldman is now on its sixth fund, with $20.3 billion to invest.) This was just one of many such funds at Goldman, which also included a technology fund, an infrastructure fund, a loan fund, a real-estate fund, a mezzanine fund, and a little-known fund of partners’ money, which for instance had made an absolute killing (that few people knew about) investing in Jinro Ltd., a Korean liquor manufacturer that had been in receivership. “They got enormous in this thing and it made multiple, multiple, multiples!” recalled one impressed hedge-fund manager.
More important, from Birnbaum’s perspective, were the surprising number of former Goldman bankers and traders who had left the firm to start their own hedge funds and were making tens of millions of dollars in annual compensation—in some cases, hundreds of millions of dollars and even billions of dollars in compensation. Among them were people such as Thomas Steyer, Daniel Och, Richard Perry, Jonathan Savitz, Eric Mindich, Edward Mule, David Tepper, David Einhorn, Edward Lampert, and Mark McGoldrick (who used to be in charge of Goldman’s Special Situations Group). Birnbaum, though a few years younger than most of these men, became increasingly tempted by the idea of the mountains of money he thought he could make on his own or working as a principal at another hedge fund.
Goldman certainly recognized Birnbaum’s talent and was doing what it could to keep him at the firm. “I was still there,” he said. “I was still listening. They weren’t saying the wrong things. But it was more driven by just an intellectual desire to try to maximize my potential.” One of the more intriguing opportunities that Birnbaum spied in the market by the end of 2005 was the increasing use of credit-default swaps, or CDS—a form of insurance that could be bought on whether a debt would in fact be paid—in the mortgage or any other debt market. Increasingly, insurance companies, such as AIG, or other Wall Street firms were willing to sell protection on whether the mortgages that went into mortgage-backed securities would in fact be paid. To get the insurance, buyers had to pay premiums to the issuers, as they did to obtain any other form of insurance. Instead of buying life insurance, or fire insurance on your house, or auto insurance on your car, buying a credit-default swap allowed investors to make bets on whether people ended up payi
ng their mortgages.
There were a number of reasons that the markets for mortgages and for credit-default swaps started to intersect in the summer of 2005. First, there was the sheer magnitude of the market for mortgage-backed securities. By the first quarter of 2004, the market for mortgage-backed securities was $6.9 trillion, some 40 percent larger than the market for U.S. corporate debt, which was $5 trillion, and the market for U.S. Treasury debt, which was $4.9 trillion. The mortgage market had more than doubled in ten years. As a result, on a sheer volume basis alone there would likely be increased demand for protection against default, especially since some prognosticators, such as David Rosenberg, the chief North American economist at Merrill Lynch, had started publishing research, in August 2004, questioning whether the real-estate market was heading for trouble. Under the headline “If Not a Bubble Then an Oversized Sud,” Rosenberg wrote, “We assess the likelihood that the housing sector has entered into a ‘bubble’ phase. There are numerous shades of gray, but when we examine the classic characteristics of a ‘bubble’ (extended valuation, over-ownership, excessive leverage, a surge in supply, complacency (denial?), and speculative behavior), it seems to fit the bill. At the very least, housing is overextended, and even the Fed has acknowledged as much. The next question is what pricks the ‘bubble’ if in fact there is one?”
In February 2005, Paul Volcker, the former Federal Reserve chairman, picked up that thread in a speech at Stanford University, where he indicated he was increasingly worried about the bubble he thought might be forming in the real-estate market. “There has been a lot of good news in the past couple of years,” he said that day, but “I have to tell you my old central banking blood still flows. Under the placid surface, at least the way I see it, there are really disturbing trends: huge imbalances, disequilibria, risks—call them what you will. Altogether the circumstances seem to me as dangerous and intractable as any I can remember, and I can remember quite a lot. What really concerns me is that there seems to be so little willingness or capacity to do much about it.” Volcker’s observations came some four months after Alan Greenspan, the Federal Reserve chairman at the time, gave a speech in Washington where he allowed that although “pockets of severe stress within the household sector … remain a concern,” the likelihood of “housing price bubbles” appeared small. By January 2006, even Goldman itself was claiming that “the oft predicted, overly anticipated subprime blow up MAY occur” in 2006 but then dampened that by suggesting that “a blow up” was less likely than a “fizzle out.” Still Goldman’s economists believed U.S. housing prices to be “overvalued by 15%+.”
The other reason for the increasing use of CDS in the mortgage market by the end of 2005 was more technical. In June 2005, the Wall Street firms got together and “standardized” the CDS contracts that were used to insure the risks in the mortgage market.
The combination of the standard contract for credit insurance, plus the explosion of mortgage issuance—especially to borrowers of lower and lower credit quality—and the outlines of concern as a result, led to a noticeable increase in the purchase of CDS on mortgages. “Particularly in the fall of 2005, you really had some of the early hedge-fund trades, significant hedge-fund trades … but this was when [hedge-fund manager] John Paulson first crept up and started doing some trades,” Birnbaum recalled. According to Goldman, there were $150 billion of credit-default swaps outstanding on “structured product” at the end of 2005, up exponentially from $2 billion the year before. The use of CDS to protect against mortgage-securities’ defaults “[g]rew faster than even we predicted,” Goldman wrote in its marketing documents.
At the same time, Birnbaum had also noticed that the cost of the premiums to buy the insurance on the mortgage payments had spiked upward, tripled in price really, from costing 1 percent of the aggregate amount being insured to 3 percent of the aggregate amount being insured. Like any market, the cost of the insurance is driven by supply and demand—the greater the demand to insure a certain debt, the greater the likelihood of a higher cost associated with it. In this market at that time, since it was so thinly traded, a sudden increase in demand can have a particularly noticeable effect on the price of the insurance. “It was just like this huge, ‘What the heck’s going on?’ ” he said. “The Street got blindsided.” Birnbaum was not trading the CDS on mortgages at that time, but he was fairly sure that John Paulson—and perhaps another hedge-fund manager—was starting to build his portfolio of CDS. They were buying credit protection against the default of individual tranches of mortgage-backed securities. “All I know is there was a very thin market, so probably this much inquiry for protection would probably create a huge change,” he said. Curious about this new dynamic in this obscure market, Birnbaum decided to buy some credit-default swaps on his trading desk—just to dabble in the market, really—and then sold them relatively quickly, making enough money to impress his colleagues. “Oh, nice job doing that,” one of them told him, although it was just one trade. “It was an alert to a lot of folks who had never looked at this market that ‘hey, there’s something interesting going on here,’ ” he said.
At the same time that a standardized CDS contract for mortgages came into being and John Paulson was starting to buy increasing amounts of CDS against mortgages, Wall Street also got together and created an index composed of securities backed by home loans issued to borrowers with weak credit that, for the first time, allowed investors to bet on the performance of the subprime mortgage market. This new index—the ABX.HE, or ABX for short—was the brainchild of people like Rajiv Kamilla, at Goldman Sachs, then in his early thirties and a former nuclear physicist. The idea, Birnbaum said, was to create “an index that was a portfolio that everyone could agree upon that was representative of the subprime market, and then instead of trading each single name, you would trade this portfolio and it could be more liquid and maybe you could trade it as a hedge instrument, maybe as a speculative instrument.”
An obscure partnership—with the odd name of CDS IndexCo LLC—owned by sixteen investment banks, including Goldman, created the index in January 2006 and for the first time, investors had a way to bet on the performance of the subprime mortgage market. (Brad Levy, a Goldman partner, served as CDS Index’s chairman). Another firm—Markit.com, based in London, and of which Goldman also owned a piece—collected and published on a daily basis the data from Wall Street firms related to the securities in the ABX index, effectively administering the index. (Markit.com was, in 2009, the subject of an investigation by the Justice Department, which, according to Forbes, “wants to know if the firms”—like Goldman and JPMorgan—“benefited from the way prices of credit indexes were posted on Markit.com. These prices, taken from the average of quotes from the dealers themselves, could only be accessed in the afternoon via the site. Otherwise, the buying and selling of credit derivatives has been completely invisible to the investing public.”)
Before the creation of the ABX index in January 2006, if the mortgage-backed securities market sold off, no one really knew for sure by how much. But with the creation of the ABX, there was now a published index that people could observe and, more important, could short to hedge whatever risks they perceived existed in the mortgage market.
The more he heard about the ABX, the more interested Birnbaum became in the possibility of trading it. This, he thought, might be his next opportunity and a way for him to stay at Goldman. (In a January 2006 presentation to a mortgage client, Goldman’s bankers described the January 19 launch of the ABX index as “THE market event” of the first half of the year.) When his managers at Goldman, Mike Swenson and Dan Sparks, agreed to let Birnbaum trade the index, he decided he would stay and do it. “The thinking was that I had all that flow experience,” he said, “but I also knew the guts of the mortgage cash flows very well. This was something where knowing both would be really useful to you.” Birnbaum recalled that Swenson, known around the office as “Swenny,” was particularly supportive of the idea. A former hockey play
er at Williams College with four children, a preppy demeanor, and a wry sense of humor, Swenson had been at Goldman since 2000. “He’s one of these guys who has a great nose for kind of knowing when people are going to be good at something,” Birnbaum said. Birnbaum gave Swenson “a lot of credit” for bringing to that desk “the right guys” with “complementary skills” and “not worrying about turf wars or ego or politics.” The idea was to put the best team together, one that would outsmart and outhustle the competition.
In early January 2006, Birnbaum moved to the desk where the ABX index would be traded. It was about four or five rows of desks away from where he had been sitting. The move was a little unusual, but all parties to it embraced the idea as a good one. The first ABX trade occurred on January 19. “At the time nobody had any sense that ABX or the credit-default swap market and mortgages would be anything close to what it ultimately was,” he said. “But it was still a very interesting product.… I think that Goldman—and myself personally—we had a very bullish view on what the index would do in terms of trading flows and just what the business prospects were in terms of having an index trading business, more so than I think any other bank.”
Money and Power Page 67
