All the Devils Are Here

by Bethany McLean; Joe Nocera

  PWG: President’s Working Group on Financial Markets. Consists of the secretary of the Treasury and the chairmen of the Securities and Exchange Commission, the Federal Reserve, and the Commodities Futures Trading Commission.

  REMIC: Real Estate Mortgage Investment Conduit. The second of two laws passed in the 1980s to aid the new mortgage-backed securities market by enabling such securities to be created without the risk of dire tax consequences.

  RMBS: Residential mortgage-backed securities. Securities backed by residential mortgages, rather than commercial mortgages.

  RTC: Resolution Trust Corporation. Government agency created to clean up the S&L crisis.

  SEC: Securities and Exchange Commission. Regulates securities firms, mutual funds, and other entities that trade stocks on behalf of investors.

  SMMEA: Secondary Mortgage Market Enhancement Act. The first of two laws passed in the 1980s to aid the new mortgage-backed securities market

  SIV: Structured investment vehicle. Thinly capitalized entities set up by banks and others to invest in securities. By the height of the boom, many ended up owning billions in CDOs and other mortgage-backed securities.

  VaR: Value at Risk. Key measure of risk developed by J.P. Morgan in the early 1990s.

  Prologue

  Stan O’Neal wanted to see him. How strange. It was September 2007. The two men hadn’t talked in years, certainly not since O’Neal had become CEO of Merrill Lynch in 2002. Back then, John Breit had been one of the company’s most powerful risk managers. A former physicist, Breit had been the head of market risk. He reported directly to Merrill’s chief financial officer and had access to the board of directors. He specialized in evaluating complex derivatives trades. Everybody knew that John Breit was one of the best risk managers on Wall Street.

  But slowly, over the years, Breit had been stripped of his authority—and, more important, his ability to manage Merrill Lynch’s risk. First O’Neal had tapped one of his closest allies to head up risk management, but the man didn’t seem to know anything about risk. Then many of the risk managers were removed from the trading floor. Within the span of one year, Breit had lost his access to the directors and was told to report to a newly promoted risk chief, who, alone, would deal with O’Neal’s ally. Breit quit in protest, but returned a few months later when Merrill’s head of trading pleaded with him to come back to manage risk for some of the trading desks.

  In July 2006, however, a core group of Merrill traders had been abruptly fired. Most of the replacements refused to speak to Breit, or provide him the information he needed to do his job. They got abusive when he asked about risky trades. Eventually, he was exiled to a small office on a different floor, far away from the trading desks.

  Did Stan O’Neal know any of this history? Breit had no way of knowing. What he did know, however, was that Merrill Lynch was in an awful lot of trouble—and that the company was still in denial about it. He had begun to hear rumblings that something wasn’t right on the mortgage desk, especially its trading of complex securities backed by subprime mortgages—that is, mortgages made to people wuth substandard credit. For years, Wall Street had been churning out these securities. Many of them had triple-A ratings, meaning they were considered almost as safe as Treasury bonds. No firm had done more of these deals than Merrill Lynch.

  Calling in a favor from a friend in the finance department, Breit got ahold of a spreadsheet that listed the underlying collateral for one security on Merrill’s books, something called a synthetic collateralized debt obligation squared, or sythentic CDO squared. As soon as he looked at it, Breit realized that the collateral—bits and pieces of mortgage loans that had been made by subprime companies—was awful. Many of the mortgages either had already defaulted or would soon default, which meant the security itself was going to tumble in value. The triple-A rating was in jeopardy. Merrill was likely to lose tens of millions of dollars on just this one synthetic CDO squared.

  Breit started calling in more favors. How much of this stuff did Merrill Lynch have on its books? How bad was the rest of the collateral? And when in the world had all this happened? Pretty soon he had the answers. They were worse than he could possibly have imagined. Merrill Lynch had a staggering $55 billion worth of these securities on its books. They were all backed by subprime mortgages made to a population of Americans who, in all likelihood, would never be able to pay those loans back. More than $40 billion of that exposure had been added in the previous year, after he had been banished from the trading floor. The reckless behavior this implied was just incredible.

  A few months earlier, two Bear Stearns hedge funds—funds that contained the exact same kind of subprime securities as the ones on Merrill’s books—had collapsed. Inside Merrill, there was a growing nervousness, but the leaders of the mortgage desk kept insisting that its losses would be contained—they were going to be less than $100 million, they said. The top brass, including O’Neal, accepted their judgment. Breit knew better. The losses were going to be huge—there was no getting around it. He began to tell everybody he bumped into at Merrill Lynch that the company was going to have to write down billions upon billions of dollars in its subprime-backed securities. When the head of the fixed-income desk found out what Breit was saying, he called Breit and screamed at him.

  Stan O’Neal had also heard that Breit had a higher estimate for Merrill Lynch’s potential losses. That is why he summoned Breit to his office.

  “I hear you have a model,” O’Neal said.

  “Not a model,” Breit replied. “Just a back-of-the-envelope calculation.” The third quarter would end in a few weeks, and Merrill would have to report the write-downs in its earnings release. How bad did he think it would be? O’Neal asked. “Six billion,” said Breit. But he added, “It could be a lot worse.” Breit had focused only on a small portion of Merrill’s exposure, he explained; he hadn’t been able to examine the entire portfolio.

  Breit would never forget how O’Neal looked at that moment. He looked like he had just been kicked in the stomach and was about to throw up. Over and over again, he kept asking Breit how it could have happened. Hadn’t Merrill Lynch bought credit default swaps to protect itself against defaults? Why hadn’t the risk been reflected in the risk models? Why hadn’t the risk managers caught the problem and stopped the trades? Why hadn’t Breit done anything to stop it? Listening to him, Breit realized that O’Neal seemed to have no idea that Merrill’s risk management function had been sidelined.

  The meeting finally came to an end; Breit shook O’Neal’s hand and wished him luck. “I hope we talk again,” he said.

  “I don’t know,” replied O’Neal. “I’m not sure how much longer I’ll be around.”

  O’Neal went back to his desk to contemplate the disaster he now knew was unavoidable—not just for Merrill Lynch but for all of Wall Street. John Breit walked back to his office with the strange realization that he—a midlevel employee utterly out of the loop—had just informed one of the most powerful men on Wall Street that the party was over.

  1

  The Three Amigos

  The seeds of financial disaster were sown more than thirty years ago when three smart, ambitious men, working sometimes in concert—allies in a cause they all believed in—and sometimes in opposition—competitors trying to gain advantage over each other—created a shiny new financial vehicle called the mortgage-backed security. In the simplest of terms, it allowed Wall Street to scoop up loans made to people who were buying homes, bundle them together by the thousands, and then resell the bundle, in bits and pieces, to investors. Lewis Ranieri, the messianic bond trader who ran the Salomon Brothers mortgage desk and whose role in the creation of this new product would be immortalized in the best-selling book Liar’s Poker, was one. Larry Fink, his archrival at First Boston, who would later go on to found BlackRock, one of the world’s largest asset management firms, and who served as a key adviser to the government during the financial crisis, was another. David Maxwell, the chief executiv
e of the Federal National Mortgage Association, a quasi-governmental corporation known as Fannie Mae, was the third. With varying degrees of fervor they all thought they were doing something not just innovative but important. When they testified before Congress—as they did often in those days—they stressed not (heaven forbid!) the money their firms were going to reap from mortgage-backed securities, but rather all the ways these newfangled bonds were making the American Dream of owning one’s own home possible. Ranieri, in particular, used to wax rhapsodically about the benefits of mortgage-backed securities for homeowners, claiming, correctly, that the investor demand for the mortgage bonds that he and the others were creating was increasing the level of homeownership in the country.

  These men were no saints, and they all knew there were fortunes at stake. But the idea that mortgage-backed securities would also lead inexorably to the rise of the subprime industry, that they would create hidden, systemic risks the likes of which the financial world had never before seen, that they would undo the connection between borrowers and lenders in ways that were truly dangerous—that wasn’t even in their frame of reference. Or, as Ranieri told Fortune magazine after it was all over: “I wasn’t out to invent the biggest floating craps game of all time, but that’s what happened.”

  It was the late 1970s. The baby boom generation was growing up. Boomers were going to want their own homes, just like their parents. But given their vast numbers—there were 76 million births between 1949 and 1964—many economists worried that there wouldn’t be enough capital to fund all their mortgages. This worry was exacerbated by the fact that the main provider of mortgages, the savings and loan, or thrift, industry, was in terrible straits. The thrifts financed their loans by offering depositors savings accounts, which paid an interest rate set by law at 5¾ percent. Yet because the late 1970s was also a time of high inflation and double-digit interest rates, customers were moving their money out of S&Ls and into new vehicles like money market funds, which paid much higher interest. “The thrifts were becoming destabilized,” Ranieri would later recall. “The funding mechanism was broken.”

  Besides, the mortgage market was highly inefficient. In certain areas of the country, at certain times, there might be a shortage of funds. In other places and other times, there might be a surplus. There was no mechanism for tapping into a broader pool of funds. As Dick Pratt, the former chairman of the Federal Home Loan Bank Board, once told Congress, “It’s the largest capital market in the world, virtually, and it is one which was sheltered from the normal processes of the capital markets.” In theory at least, putting capital to its most efficient use was what Wall Street did.

  The story as it would later be told is that Ranieri and Fink succeeded by inventing the process of securitization—a process that would become so commonplace on Wall Street that in time it would be used to bundle not just mortgages but auto loans, credit card loans, commercial loans, you name it. Ranieri named the process “securitization” because, as he described it at the time, it was a “technology that in essence enables us to convert a mortgage into a bond”—that is, a security. Fink developed a key technique called tranching, which allowed the securitizer to carve up a mortgage bond into pieces (tranches), according to the different risks it entailed, so that it could be sold to investors who had an appetite for those particular risks. The cash flows from the mortgages were meted out accordingly.

  The truth is, though, that the creation of mortgage-backed securities was never something Wall Street did entirely on its own. As clever and driven as Fink and Ranieri were, they would never have succeeded if the government hadn’t paved the way, changing laws, for instance, that stood in the way of this new market. More important, they couldn’t have done it without the involvement of Fannie Mae and its sibling, Freddie Mac, the Federal Home Loan Mortgage Corporation. The complicated interplay that evolved between Wall Street and these two strange companies—a story of alliances and feuds, of dependency and resentments—gave rise to a mortgage-backed securities market that was far more dysfunctional than anyone realized at the time. And out of that dysfunction grew the beginnings of the crisis of 2008.

  Almost since the phrase “The American Dream” was coined in the early 1930s, it has been synonymous with homeownership. In a way that isn’t true in most other countries, homeownership is something that the vast majority of Americans aspire to. It suggests upward mobility, opportunity, a stake in something that matters. Historically, owning a home hasn’t just been about taking possession of an appreciating asset, or even having a roof over one’s head. It has also been a statement about values.

  Not surprisingly, government policy has long encouraged homeownership. The home mortgage interest deduction is a classic example. So is the thirty-year fixed mortgage, which is standard in only one other country (Denmark) and is designed to allow middle-class families to afford monthly mortgage payments. For decades, federal law gave the S&L industry a small interest rate advantage over the banking industry—the housing differential, this advantage was called. All of these policies had unswerving bipartisan support. Criticizing them was political heresy.

  Fannie Mae and Freddie Mac were also important agents of government homeownership policy. They, too, were insulated from criticism. Fannie Mae, the older of the two, was born during the Great Depression. Its original role was to buy up mortgages that the Veterans Administration and the Federal Housing Administration were guaranteeing, thus freeing up capital to allow for more government-insured loans to be made.

  In 1968, Fannie was split into two companies. One, nicknamed Ginnie Mae, continued buying up government-insured loans and remained firmly a part of the government. Fannie, however, was allowed to do several new things: it was allowed to buy conventional mortgages (ones that had not been insured by the government), and it was allowed to issue securities backed by mortgages it had guaranteed. In the process, Fannie became a very odd creature. Half government enterprise, it had a vaguely defined social mandate from Congress to make housing more available to low- and middle-income Americans. Half private enterprise, it had shareholders, a board of directors, and the structure of a typical corporation.

  At about the same time, Congress created Freddie Mac to buy up mortgages from the thrift industry. Again, the idea was that these purchases would free up capital, allowing the S&Ls to make more mortgages. Until 1989, when Freddie Mac joined Fannie Mae as a publicly traded company, Freddie was actually owned by the thrift industry and was overseen by the Federal Home Loan Bank Board, which regulated the S&Ls. People in Washington called Fannie and Freddie the GSEs, which stood for government-sponsored enterprises.

  Here’s a surprising fact: it was the government, not Wall Street, that first securitized modern mortgages. Ginnie Mae came first, selling securities beginning in 1970 that consisted of FHA and VA loans, and guaranteeing the payment of principal and interest. A year later, Freddie Mac issued the first mortgage-backed securities using conventional mortgages, also with principal and interest guaranteed. In doing so, it was taking on the risk that the borrower might default, while transferring the interest rate risk from the S&Ls to a third party: investors. Soon, Freddie was using Wall Street to market its securities. Volume grew slowly. It was not a huge success.

  Though a thirty-year fixed mortgage may seem simple to a borrower, mortgages come full of complex risks for investors. Thirty years, after all, is a long time. In the space of three decades, not only is it likely that interest rates will change, but—who knows?—the borrowers might fall on hard times and default. In addition, mortgages come with something called prepayment risk. Because borrowers have the right to prepay their mortgages, investors can’t be sure that the cash flow from the mortgage will stay at the level they were expecting. The prepayment risk diminishes the value of the bond. Ginnie and Freddie’s securities removed the default risk, but did nothing about any of these other risks. They simply distributed the cash flows from the pool of mortgages on a pro rata basis. Whatever happened after that, w
ell, that was the investors’ problem.

  When Wall Street got into the act, it focused on devising securities that would appeal to a much broader group of investors and create far more demand than a Ginnie or Freddie bond. Part of the answer came from tranching, carving up the bond according to different kinds of risks. Investors found this appealing because different tranches could be jiggered to meet the particular needs of different investors. For instance, you could create what came to be known as stripped securities. One strip paid only interest; another only principal. If interest rates declined and everyone refinanced, the interest-only strips could be worthless. But if rates rose, investors would make a nice profit.

  Sure enough, parceling out risk in this fashion gave mortgage-backed securities enormous appeal to a wide variety of investors. From a standing start in the late 1970s, bonds created from mortgages on single-family homes grew to more than $350 billion by 1981, according to a report by the Securities and Exchange Commission. (By the end of 2001, that number had risen to $3.3 trillion.)

  Tranching was also good for Wall Street, because the firms underwriting the mortgage-backed bonds could sell the various pieces for more money than the sum of the whole. And bankers could extract rich fees. Plus, of course, Wall Street could make money from trading the new securities. By 1983, according to Business Week, Ranieri’s mortgage finance group at Salomon Brothers accounted for close to half of Salomon’s $415 million in profits. Along with junk bonds, mortgage-backed bonds became a defining feature of the 1980s financial markets.

  Tranching, however, was not the only necessary ingredient. A second important factor was the involvement of the credit rating agencies: Moody’s, Standard & Poor’s, and, later, Fitch Ratings. Ranieri pushed hard to get the rating agencies involved, because he realized that investors were never going to be comfortable with—or, to be blunt, willing to work hard enough to understand—the intricacies of the hundreds or thousands of mortgages inside each security. “People didn’t even know what the average length of a mortgage was,” Ranieri would later recall. “You needed to impose structures that were relatively simple for investors to understand, so that they didn’t have to become mortgage experts.” Investors understood what ratings meant, and Congress and the regulators placed such trust in the rating agencies that they had designated them as Nationally Recognized Statistical Ratings Organizations, or NRSROs. Among other things, the law allowed investors who weren’t supposed to take much risk—like pension funds—to invest in certain securities if they had a high enough rating.