Infectious Greed (21 page)

Robert Citron was about as far from the high-tech Arbitrage Group at Salomon as a person could get. He was a University of Southern California dropout who had been to New York only four times in his life. He kept investment records on index cards and a wall calendar, and used his wristwatch calculator more than any financial software. Instead of developing computer pricing models, he consulted psychics and astrologers for advice about interest rates.

Citron had spent his entire career in Orange County's treasury department. After several decades, he held the elected position of treasurer—with no term limits. Throughout the 1980s and early 1990s, Citron had outperformed every other county investment manager, sometimes by several percentage points. While some counties were earning just two or three percent, Citron was earning almost 10 percent. The voters loved Citron and reelected him, over and over.

By the early 1990s, Citron was one of the largest investors in the country, managing $7.4 billion of Orange County taxpayers' money. But instead of investing the money in plain-vanilla Treasury bonds, he bought structured notes—the same instruments with embedded formulas that Allen Wheat and First Boston had sold by the billions. Citron bought a few of these notes from First Boston, but most of his purchases were from other banks, especially Merrill Lynch. By this time, structured notes had spread to every major investment bank, including Merrill.

Merrill's structured notes looked just like First Boston's: they ranged in maturities from 3 to 5 years (as Orange County's guidelines—which Citron had helped to write—required), and were issued by highly rated entities, such as the Federal Home Loan Bank (as Orange County's guidelines required). On paper, they looked like very safe, AAA-rated investments.

But the structured notes contained formulas that essentially were a big bet on interest rates remaining low. For example, one $100 million note paid a coupon of 10 percent minus LIBOR, the short-term interest-rate index. So if LIBOR were 3 percent, Orange County would receive a 7 percent coupon. As interest rates rose, Orange County's coupon went down, and vice versa. By buying these
inverse floaters,
Citron effectively was borrowing at short-term rates and investing at longer-term rates. Most of the notes were based on U.S. rates, but Citron also bet that European rates would remain low; he even purchased $1.8 billion of structured notes tied to Swiss LIBOR.
³

Citron quickly came to believe that $7.4 billion was not a big-enough
bet, so he borrowed as much as he could—about $13 billion—from various banks, including Merrill Lynch. He invested that money in structured notes, too. In all, by early 1994, Citron had made a $20 billion bet on low short-term rates.

In many ways, Robert Citron was simply a big, public version of Jim Johnsen of Gibson Greetings, the company that had unknowingly paid more than $10 million in fees for the interest-rate bets it bought from Bankers Trust. Like Johnsen, Citron wanted to gamble on interest rates, but was constrained by investment guidelines that permitted only highly rated, short-maturity bonds. Like Johnsen, Citron did not understand how to evaluate the derivatives embedded in structured notes, and he paid more for them than he should have, especially because he frequently didn't bother to shop around. That made Citron an extremely valuable client to Merrill Lynch. From 1990 to 1993, Merrill Lynch earned profits of $3.1 billion, more than it had earned during its 18-year history as a public company, and a big chunk of the profits—about $100 million—came from Orange County.
⁴
Merrill earned $62.4 million from Orange County in 1993 and 1994 alone.

Thus placed in its proper context, the story of Orange County's losses was simple, and all too familiar. Citron used structured derivatives to bet on low interest rates, just as Gibson Greetings had, and he lost his bets when the Federal Reserve raised rates on February 4. The only difference between Orange County and Gibson Greetings was the size of the bet: millions of dollars for Jim Johnsen, billions for Citron.

Who was to blame for the losses? Obviously, Citron was at the top of the list. He made all the decisions and hid all the risks. On January 17, 1995, he told a special committee of the California State Senate, “I was an inexperienced investor. In retrospect, it is clear that I followed the wrong course. I will carry that burden the rest of my life.”

Many public accounts also blamed Merrill Lynch. Michael Stamenson, the top salesman in Merrill's San Francisco office, didn't help Merrill's media relations when he publicly praised Citron's investment acumen. No one thought Stamenson genuinely believed, as he testified, that “Mr. Citron is a highly sophisticated, experienced, and knowledgeable investor. I learned a lot from him.”

But while Stamenson may have engineered Citron's gambling, the structured notes Merrill sold to Orange County were no different from tens of billions of dollars of similar instruments sold by numerous banks to numerous investors. To the extent there were problems with structured
notes, they weren't specific to Merrill; they were endemic to the financial system. If Merrill was to blame for Orange County, then all of Wall Street was to blame for billions of dollars of losses in 1994.

In fact, Merrill seemed less culpable than other banks, especially Bankers Trust. Merrill had warned Citron on at least eight occasions, from 1992 to 1994, to reconsider his risky interest-rate bets. But Citron ignored all warnings, including those from prestigious Goldman Sachs, which disapproved of Citron's risks and refused to sell him any structured notes. When Goldman criticized Citron, he responded with an angry letter, saying, “You don't understand the type of investment strategies that we are using. I would suggest that you not seek doing business with Orange County.”

The central problem for Merrill and other banks was not necessarily selling structured notes (as troubling as it might seem for Merrill to be selling securities with complex embedded formulas to an antediluvian client like Citron). Instead, the problem was that Merrill had played a dual role of derivatives salesman and bond underwriter. In this second role, Merrill and other banks had arranged new Orange County bond issues and sold them to the public. Unfortunately for Merrill, the public disclosures related to those bond issues did not mention Citron's risky interest-rate bets. That meant that even if Merrill had dealt properly with Citron in selling structured notes, it arguably had violated the securities laws by failing to disclose risks associated with Orange County's bonds.

The other blameworthy parties were the credit-rating agencies. Just as Standard & Poor's and Moody's had been critical to the development of new financial instruments at First Boston and Salomon Brothers, the agencies had played a central role in the collapse of Orange County.

First, they had given AAA ratings to the structured notes Orange County bought, even though the market risks of those notes were much greater than those of more typical AAA-rated investments. That enabled Robert Citron to fit his large interest-rate bets within the technical boundaries of Orange County's investment guidelines, even though the structured notes he bought were riskier than any highly rated bond in existence when the guidelines were written.

Second, both Standard and Poor's and Moody's gave Orange County
itself
their highest ratings through December 1994, when the county filed for bankruptcy. These high ratings gave confidence not only to Orange County residents, but also to investors in the county's bonds.
Investors in many conservative bond mutual funds—including those at Franklin Advisors, Putnam Management, Alliance Capital, Dean Witter, and many others—had purchased Orange County's bonds, precisely because of the high credit ratings.
⁵

The rating agencies collected substantial fees for rating Orange County's bonds (S&P made more than $100,000 from Orange County in 1994 alone).
⁶
They collected even greater fees for rating structured notes. These fees raised questions about whether the agencies had been objective in assessing Orange County's risks. More than six months before Orange County's bankruptcy, the agencies had learned about Citron's losses on structured notes, but they kept this information secret, and didn't adjust their ratings in response.

For example, according to notes taken by a rating-agency analyst, on a telephone conference call with the rating agencies on May 9, 1994—seven months before the agencies downgraded Orange County's debt—Matthew Raabe, Citron's assistant treasurer, candidly discussed the county's risks, noting “DISASTER” and “danger!” and concluding that if interest rates rose one percent, Orange County's collateral would be “gone.”
⁷
At another meeting, officials from the rating agencies learned about large numbers of “inverse floaters.”
⁸
In addition, Raabe explained that Orange County was not “marking to market” its portfolio, meaning that it wasn't recording changes in value over time, even as interest rates increased.
⁹
In other words, if Orange County had paid $20 billion for structured notes in 1993, they were still recorded as being worth $20 billion in 1994, even though the Fed's rate hike had decimated their value.

Even if the agencies hadn't been privy to this information, they knew of numerous public warning signs about Citron, including Citron's own remarks from the previous year. In July 1993, when Citron was asked how he knew that interest rates would not rise, he replied, “I am one of the largest investors in America. I know these things.” In a September 1993 report to the County Board of Supervisors, Citron wrote, “Certainly there is nothing on the horizon that would indicate that we will have rising interest rates for a minimum of three years.” These statements were red flags showing how important the level of interest rates was to Citron's investment strategy, and how naïve Citron was. In spring 1994, John Moorlach, an accountant from Costa Mesa, California, launched an aggressive campaign to unseat Citron. Moorlach warned voters that
the county was taking on huge risks and predicted a billion-dollar loss. But Citron won reelection, the losses remained hidden, and the rating agencies stayed mum.

Finally, on December 1, 1994—long after it was clear Orange County had lost more than a billion dollars—Standard & Poor's finally said it
might
lower Orange County's credit rating.
¹⁰
The agencies finally downgraded Orange County during the second week of December 1994, as the county was preparing to file for bankruptcy. The downgrades were much too late. Robert Froelich, director of bond research at Van Kampen Merritt, said, “If rating agencies can't keep track of one of the largest counties in the U.S., what is the value of their ratings on other counties?”
¹¹
Or, he might have said, other companies?

As Orange County fell into bankruptcy, the major derivatives dealers swooped down to feed on its carcass. First, they made sure their loans would be repaid. In early December, First Boston sold $2.6 billion of collateral the county had posted for its loans.
¹²
Chapter 11, the relevant section of the bankruptcy laws for companies and individuals, would have prohibited such a sale, but First Boston's lawyers argued it was permitted under Chapter 9, the little-known and little-used section for municipalities. Other banks liked this argument, and sold collateral
after
the filing, something most experts agreed was prohibited.

Freed from the loans they had made to Orange County, the banks began feasting on the county's structured notes. In 1994, dealers bought back about $20 billion of structured notes, roughly one-fifth of the total outstanding in the entire market.
¹³
They repackaged the notes and sold off their risks, using swaps and securitizations, the recent innovations by First Boston and other Wall Street firms. They commonly earned fees of about a half percent from these deals, which meant profits of another $100 million.

Notably absent from these dealings was Salomon Brothers. By this time, its Arbitrage Group had disbanded, and the firm was slow to enter the Orange County fray.
¹⁴
Now under the control of Warren Buffett, Salomon was taking fewer risks and having a terrible year; it was well on its way to losing $400 million in 1994.

Given the spread of financial innovation, the collapse of Orange County seemed, with hindsight, to have been entirely predictable. It was dangerous to have someone like Robert Citron managing billions of dollars, especially when Wall Street firms were inventing new financial instruments to help fund managers place aggressive bets without
regard for investment restrictions. Moreover, many of these new instruments were largely unregulated, so there were no ground rules to prevent fund managers—even municipal treasurers—from putting other people's money at risk. There weren't even disclosure requirements related to the new instruments. Given these circumstances, Orange County's collapse was not only understandable, it was practically unavoidable.

Orange County emerged from bankruptcy intact, but it wasn't the only municipality to buy complex derivatives. Municipal officials throughout the country had wanted to gamble on interest rates during the early 1990s, and because they couldn't evaluate derivatives properly, they paid high fees, just like Orange County and Gibson Greetings. The same factors that led Michael Stamenson to pitch structured notes to Robert Citron also led other Wall Street salesmen to pitch complex derivatives deals to various city and county treasurers throughout the United States. As one example, brokers called Joseph Flowers, the controller of Escambia County—in Florida's panhandle—four or five times a day, until he finally put 15 percent of Escambia's $45 million portfolio into structured derivatives.
¹⁵