13 Bankers: The Wall Street Takeover and the Next Financial Meltdown (14 page)

This process began in the 1970s, when Michael Milken, a trader at Drexel Burnham Lambert, had the insight that “junk bonds”—bonds that were rated below “investment grade” by the credit rating agencies
*
—were generally underpriced, either because investors had an irrational aversion to them or because they lacked a liquid market in which to trade them. He capitalized on that market inefficiency, building an operation that dominated the trading and sales of junk bonds. By creating a large, liquid market for junk bonds—which grew from $6 billion in 1970 to $210 billion in 1989
⁶⁵
—he made it easier for companies to raise money and opened up vast new opportunities for investment banks to generate profits by underwriting, selling, and trading these formerly neglected bonds. By making it easy to raise large amounts of money quickly, junk bonds also made possible the leveraged buyout craze of the 1980s, in which acquirers would pay for acquisitions by issuing large amounts of new debt. Those acquisitions, in turn, generated huge fees for the investment banks that advised the companies engaged in those transactions and also underwrote and sold the necessary debt. They also left companies struggling with huge debt burdens, often requiring painful restructuring and sometimes leading to bankruptcy.

In the 1980s, the Securities and Exchange Commission and the U.S. Attorney’s Office for the Southern District of New York (led by Rudy Giuliani) launched investigations of Milken and Drexel Burnham Lambert for insider trading, securities manipulation, and fraud. The investigations eventually led to convictions for both Milken and his employer for securities and reporting violations.
⁶⁶
But junk bonds—rebranded as “high-yield” bonds—remained a popular form of financing, with over $600 billion in new bonds issued by U.S. corporations from 2003 through 2007.
⁶⁷
Equally important, investor demand for higher-yield, higher-risk bonds remained strong—driving the recent boom in mortgage-backed securities, especially as returns on Treasury bonds fell to historic lows in the past decade.

The private mortgage-backed securities invented by Lewis Ranieri at Salomon Brothers, with an assist from the federal government, had even larger implications for the future than Milken’s creation of the junk bond market. Without securitization, banks generally either held on to the mortgage loans that they made or sold them on the secondary market to Fannie Mae or Freddie Mac—government-sponsored enterprises that provided liquidity to the housing market. By buying mortgages, guaranteeing their principal payments, and turning them into mortgage-backed securities, Fannie and Freddie provided funding for banks to make more mortgages and absorbed some of the risk of the market. Because whole mortgages were difficult to trade (since every mortgage is unique), the number of transactions generated by each mortgage was small.

Securitization created many new ways for banks to profit. For the banks making the initial mortgages, securitization created a new market for their loans, making it easier for them to recover their cash and lend it out again to another borrower, boosting volume. Investment banks had three new ways to make money. They could take fees out of each securitization that they created; they could earn fees selling the new mortgage-backed securities to investors; and they could earn fees (or trading profits) by trading these securities. In each case, the revenues available depended on the volume of mortgage-backed securities. The total volume of private mortgage-backed securities (excluding those issued by Ginnie Mae, Fannie Mae, and Freddie Mac) grew from $11 billion in 1984 to over $200 billion in 1994 to close to $3 trillion in 2007.
⁶⁸
In addition, once mortgage securitization had caught on, investment banks looked for any other financial assets they could turn into securities—credit card debt, student loans, anything that paid a more or less steady stream of cash flows—further expanding the market.

While mortgage securitization was good for the banks that originated the mortgages and turned them into securities, it was a mixed blessing for everyone else. On the one hand, by attracting additional investors to the mortgage market, it expanded the pool of money available to people who wanted to buy houses. On the other hand, it made possible the “originate to distribute” business model, in which lenders could make profits lending money to people who could not pay it back, making mortgage defaults and foreclosures more likely. Under this model, the risk of default is passed downstream to the investors buying the mortgage-backed securities, who therefore become responsible for policing the quality of the underlying loans; instead of having to make sure that borrowers were likely to pay them back over the next thirty years, lenders could be paid back immediately when they sold off the mortgages for securitization. When it turned out that investors had no idea what risks they were taking on, the result was the collapse of a housing bubble larger than any in recorded American history.

In addition to mortgage-backed securities, Salomon Brothers also pioneered arbitrage trading, which took a variety of forms. For example, traders could make certain money by finding two securities that should but did not have the same value—say, U.S. Treasury bonds maturing in August 2040 and September 2040—buying one, selling the other,
*
and waiting for their prices to converge. Or they could achieve the same goal by buying the interest payments and the final principal payment on a 30-year bond separately while selling the whole 30-year bond (including interest and principal) for a higher price. In general, such trades make money in one of two ways. Either there is a complex product, often involving embedded options, that other people in the market do not know how to value correctly, or there is a regulation that creates a market inefficiency that can be exploited. In any case, under ordinary market conditions, arbitrage opportunities are close to free money—at least until the number of competitors mimicking a particular strategy drives its profitability down to zero—a far cry from the traditional business of lending money and taking the risk that it might not be paid back.

Although Salomon pioneered quantitative arbitrage trading, the practice soon spread to other investment banks, causing a huge flow of talent into Wall Street (significantly driving up the mathematical aptitude shown on the Philippon-Reshef skill curve) and leading to vast growth in banks’ proprietary trading activities (trading on their own account, rather than executing trades for clients).
*
Its popularity also fueled the rapid growth of hedge funds (lightly regulated investment funds open only to institutions and rich individuals), which grew collectively from less than $30 billion in assets under management in 1990 to over $1.2 trillion in 2005,
⁶⁹
with estimates over $2 trillion by 2008.
⁷⁰
Arbitrage became a staple trading strategy of many hedge funds. And because hedge funds rely on investment banks to execute their trades, the growth of hedge funds provided another source of revenue for the banks.

But while it made the hedge funds and the investment banks plenty of money, arbitrage trading had two other effects. Because hedge funds are largely unregulated, large risk exposures were building up outside the view of the financial regulators. And because arbitrage spreads are typically very thin—pricing inefficiencies tend to vanish as traders take advantage of them—making significant profits required large amounts of borrowed money. This leverage was the reason why some proprietary trading operations lost large amounts of money during the 1997–1998 emerging markets crises when prices failed to converge as expected—leading even Salomon Brothers to largely disband its arbitrage team.
⁷¹

The fourth money machine of modern finance—after high-yield debt, securitization, and arbitrage trading—was the modern derivatives market. While commodity futures contracts (in which, for example, a farmer commits to sell wheat in the future for a pre-specified price) had been around for centuries,
*
the market for financial derivatives remained small until the early 1980s, largely because traders—or, more accurately, the managers responsible for keeping those traders in line—had no good way of calculating what they were worth. But in the 1970s, the Black-Scholes Model gave banks a new way to calculate the value of complicated derivatives and the hedges they used to protect themselves.
^†

Wall Street embraced these quantitative models because they made it easier to price and trade derivatives in bulk. Nassim Taleb and Pablo Triana have argued that quantitative models are
worse
at pricing derivatives than the heuristics used by traders from the pre-quantitative era.
⁷²
Nonetheless, the formulas gave banks the confidence to sell growing volumes of increasingly complicated trades to clients, since the models enabled them to keep track of how much money they were making on each one. Mathematical models also made it possible to break down complicated trades into simpler ones, a critical factor in the development of sophisticated financial products.

The modern derivatives revolution began with the invention of the interest rate swap (by Salomon Brothers) in 1981. In this transaction, Company A pays interest at a fixed rate to Company B and Company B pays interest at a floating rate (which can go up or down as economic conditions change) to Company A. Interest rate swaps allow companies to exchange fixed rate payments for floating rate payments, or vice versa—“swapping” interest rate risks between the two parties.
^‡
Similarly, currency swaps allow companies to swap currency risks by exchanging different currencies (or combinations of currencies). Interest rate swaps can also be combined with currency swaps. These two basic derivatives became popular ways for companies to manage financial or operational risks.

For example, a company might have issued $100 million in bonds on which it has to pay a floating interest rate (like an adjustable rate mortgage). Because it does not want to bear the risk that interest rates will go up, it can buy an interest rate swap from a derivatives dealer. The company will pay a fixed rate—say, 7 percent, or $7 million per year—to the dealer; in exchange, the dealer will pay the company the same floating rate that the company has to pay on its bonds. The net result is that the company now pays $7 million per year whether interest rates go up or down, and the dealer bears the interest rate risk. Currency swaps are similar; for example, a U.S. manufacturer with a multiyear contract in Thai baht might enter into a currency swap in order to lock in a constant future exchange rate and protect itself from the risk that the baht will fall. In either case, these derivatives allow companies to offload risks that they do not want to take onto the financial markets, where presumably some party will take on those risks in exchange for a high expected return. (Because the evolution of derivatives has run ahead of regulatory and accounting rules, derivatives can also serve other purposes, such as helping companies smooth their earnings over multiple periods or reduce their tax bills by deferring earnings into the future.)

By the middle of 2008, the market for over-the-counter (customized) interest rate swaps had grown to over $350 trillion in face value (the amount on which interest is calculated) and over $8 trillion in gross market value.
*
73
The derivatives dealers—both investment banks and large commercial banks—were taking a piece of every interest rate swap in fees. Even better, the dealers would typically hedge their exposures; ideally, for every swap with one client, they would conduct an opposite swap with another client, so the two trades canceled out—leaving nothing but fees from both clients.

But ordinary swaps were easy for any derivatives dealer to duplicate, and competition between banks soon drove profit margins down near zero. The solution was to invent newer, more complex versions of interest rate and currency derivatives that were hard for competitors to duplicate—and hard for clients to understand. There were “inverse floaters,” whose interest rates went in the opposite direction from market interest rates, souped up with leverage and embedded options. There were interest rate swaps that changed their terms based on currency exchange rates. Each of these trades was a customized combination of bonds, swaps, and options that was virtually impossible for a typical client to accurately value; the dealer would charge a large fee, break the transaction into its component parts, and hedge them individually at a much lower cost.

Derivatives expert Satyajit Das has outlined the abusive economics of these transactions. Describing the mechanics of a typical inverse floater, he wrote, “Chairman Greenspan might wax lyrical about the unbundling of risks but we spent most of our waking hours frantically rebundling the risks and stuffing them down the throats of any investor we could find.”
⁷⁴
Because these derivatives were zero-sum trades—one party’s loss was the other’s profit—one side could be badly burned. Orange County lost almost $2 billion on inverse floaters and similar trades that treasurer Robert Citron clearly did not understand; real-economy companies such as Procter & Gamble and Gibson Greetings similarly lost tens or hundreds of millions of dollars.
⁷⁵
But these transactions generated large fees for the dealers; Merrill Lynch alone made $100 million on deals with Orange County.
⁷⁶