Askin bought many of the same securities Worth Bruntjen of Piper bought, including hundreds of millions of dollars of inverse IOs, sold by Kidder Peabody, which General Electric owned. By 1994, Askin had become such a big player that the market for exotic mortgage derivatives simply could not function without the fund. Bruntjen was dependent on Askin’s trading of complex CMOs; without it, the market would dry up, and the price of the securities would plummet.
After the rate hike, Bruntjen’s fund had lost a great deal of money, but Askin was in much worse condition. The fund was stuck with long-maturity securities that paid virtually nothing, and Askin had borrowed money to leverage its positions. As the markets went down, Askin needed to sell CMOs to repay some of the loans. But when the firm began selling, the sales created a downward spiral in the CMO market, with prices dropping so low that many dealers refused to trade at all.24
Askin also was a victim of the same kinds of valuation problems that had plagued Bankers Trust and Salomon. When the Fed raised rates on February 4, the Granite fund used its own computer models to determine the value of its mortgages, instead of looking to the market for values. As the markets declined, Granite’s computer models said that the market was wrong, and that the mortgages were still valuable. Granite did not mark to market its securities based on reliable, outside data. Instead, it valued its portfolio based on the computer’s outputs.
Askin survived briefly because of the fund’s cozy relationship with brokers. The brokers made millions from Askin’s purchases, so they gave the fund favorable lending terms, and even allowed Askin to take possession of securities he had not yet paid for. As described in the previous chapter, the value of a mortgage derivative depended greatly on assumptions about how quickly homeowners were prepaying their mortgages; with this flexibility, brokers could give Askin favorable mark-to-market numbers, to help justify relying on the fund’s computer outputs, and thereby smooth the fund’s income and, potentially, hide losses. In describing the special consideration David Askin received, one investor called him “the Hillary Clinton of the bond market.”25 (This was a reference to Hillary Clinton’s foray into cattle-futures trading, where she netted $100,000, allegedly because her broker had allocated only winning trades to her account.)
The special treatment helped at first. When Askin reported its February 1994 results to investors, there appeared to be only a two-percent loss. But a few weeks later, when Askin obtained some less friendly marks from other dealers, the loss was 20 percent. In late March, Askin told investors the loss was 35 percent.26
At this point, it appeared that Askin might not be able to repay its loans. Askin’s brokers—fair-weather friends—abruptly abandoned the fund and began issuing margin calls, demanding that Askin repay the money it had borrowed. When Askin couldn’t repay the loans, brokers sold Askin’s CMOs—just as Orange County’s brokers had sold its structured notes. The CMO markets crashed, destroying not only Askin, but Worth Bruntjen and numerous other investors. By April 7, all $600 million of Askin’s fund was gone.
Throughout this time, Askin was paying huge fees for the CMOs it purchased. Either Askin was overpaying its brokers and hoping for favorable treatment, or it was no more sophisticated than Worth Bruntjen. The first possibility raised troubling questions about conflicts of interest. But the second possibility—that the Askin fund’s emperor had no clothes—was more likely. It was difficult for any fund to impersonate Salomon’s Arbitrage Group, and Askin apparently had failed.
Taped conversations among salesmen at Kidder Peabody, then the leading CMO dealer, confirmed this second view. After the Fed’s rate hike, Kidder’s mortgage salesmen were hurting, because trading in complex mortgages had declined. The salesmen passed the time by mocking Askin, just as Bankers Trust salesmen had mocked Procter & Gamble. Kidder had sold Askin hundreds of millions of dollars of CMOs during the weeks before and after the Fed’s rate hike, just as Bankers Trust had sold swaps to P&G during February 1994, and Kidder had made even more money than Bankers Trust from the sales.
Here is a record of one conversation alleged to have taken place between William O’Connor, the Kidder salesman who covered Askin, and his assistant, Jay Pappas, on March 25, 1994, just seven weeks after the Fed’s rate hike, in the heat of the uncertainty about whether Askin would survive:PAPPAS: Just pull the plug if Askin doesn’t pay.
O’CONNOR: You don’t seem to understand. We can’t pull the plug. We are in bed with these guys. OK? Last thing in the world we want to hear is that Askin can’t meet a margin call and we got to liquidate the nuclear waste they bought.
PAPPAS: A lot of it’s nuclear waste?
O’CONNOR: It’s all nuclear waste, come on. You get paid a plus [1/64 of one percent] for selling a Ginnie 7 [a plain-vanilla mortgage security]. You get paid a plus for selling a long bond. What do you think these are? We get paid a point [one percent]. Nuclear waste.
PAPPAS: You got a point and a half, two points on some of them.
O’CONNOR: Yes.
PAPPAS: Did you know that when you sold them?
O’CONNOR: Of course I did.27
To translate: Wall Street was making just as much money from sales to Askin as it was making from sales of similar instruments to municipalities and mutual funds. Interestingly, the brokers fed on Askin, as it lay dying, just as they had picked at Orange County after its bankruptcy. Howard Rubin—the trader who had lost $377 million for Merrill Lynch in 1987 and then resurfaced, unpunished, at Bear Stearns—was one of the vultures, along with traders from Kidder Peabody. It appeared that Kidder Peabody and Bear Stearns were “cooperating” to buy Askin’s bonds at fire-sale prices, just as the structured note dealers seemed to have colluded in buying Orange County’s bonds. William O’Connor of Kidder reportedly said that “basically we told Howie Rubin, we’ll bid all your bonds, you got to bid all ours, we just need legal for court later on.”28 The dealers apparently had plenty of “legal”—meaning they believed they could defend each other’s low valuations as lawful and appropriate—and they were never punished for these actions. Howard Rubin and Bear Stearns allegedly made $20 million in a few days, buying and then reselling Askin’s CMOs.29 Askin filed for bankruptcy a few days later. The fund had been able to survive pretending to be Salomon’s Arbitrage Group for only about a year.
Askin’s troubles were terrible news for Worth Bruntjen, who was struggling to evaluate the damage done to his fund. By March 1994, Piper was unable even to determine end-of-day prices of the assets in Bruntjen’s fund on a daily basis, as required by law. Bruntjen and his staff had switched to a weekly pricing procedure—comically dubbed “Pricing Thursdays”—in which they would attempt to evaluate the prices of all of their instruments by calling various financial institutions and plugging data they often didn’t understand into computer models.30 During early April 1994, the pricing procedure took most of the day.31
Like Askin, Piper did not disclose the risks associated with these complex instruments, or the difficulties of pricing them. At Piper, the problems were not even disclosed to employees outside Bruntjen’s circle, or to many of the salesmen who had sold so many millions of dollars’ worth of Bruntjen’s fund to their accounts. Piper employees later would express shock when they learned that Bruntjen had not been such an expert in mortgage derivatives. Many people had assumed that Bruntjen’s background was similar to those of the traders and Ph.D.s at Bankers Trust, First Boston, and Salomon Brothers. But the reality was that Bruntjen was far from a “rocket scientist”—he did not have a college degree, and he certainly did not have the training necessary to understand complex CMOs.
The supposedly ultra-safe mutual funds Worth Bruntjen managed at Piper were the worst-performing bond funds in the entire market in 1994. Bruntjen was down 28 percent for the year, an inexplicable loss for a short-term government-bond fund that advertised preservation of principal. John Rekenthaler, editor of Morningstar—which had given Piper its highest ratings—said that although Piper had claimed
Bruntjen’s fund was short-term in nature, “the fund had to be a hell of a lot longer.”32 Yet these longer-maturity risks, embedded in highly rated CMOs, had been invisible to fund investors.
In September 1994, CEO Tad Piper maintained that the instruments Bruntjen owned were still undervalued, and said he believed the fund would come back. He admitted that the fund “got caught in a market that we thought we understood,”33 but insisted that “we have great confidence in Worth.”34 He did not explain how a conservative government-bond fund could lose 28 percent in less than a year, or how it previously could have averaged returns of more than 13 percent for several years. In any event, Tad Piper would not be taking Worth Bruntjen on any more ski vacations. Bruntjen would continue to work for Piper only until just after the lawsuits against his firm were resolved. Bruntjen settled the SEC’s case against him by agreeing to be barred from the industry for five years. He paid a $100,000 fine.35
Like Orange County and the various municipalities, Piper was far from the only mutual fund at the gambling tables. In the early 1990s, managers of many of the largest mutual funds in the world had placed similar, hidden bets, using various types of derivatives. The bets included not only structured notes and mortgage derivatives, but novel variations on the currency derivatives Andy Krieger had pioneered. For example, the Alliance North American Government Income Fund—then one of the largest and best-performing bond funds in the world—had placed large bets on the Mexican peso, which had remained stable relative to the U.S. dollar for several years. For years, the bets had paid off, and Alliance’s fund was the best performer in its category. (The aftermath of the peso devaluation of late 1994 is covered in Chapter 8.)
The major money-market funds—which had bet on U.S. interest rates, just as Orange County and Piper had—did not fare as well. Money-market funds were supposed to be ultra-safe, basically a substitute for cash or a checking account. Yet several banks were forced to prop up the money-market funds they managed, in order to avoid losses, including such major funds as BankAmerica ($68 million), First Boston ($40 million), Merrill Lynch ($20 million), and PaineWebber ($268 million).36 As if Kidder Peabody’s bad year selling mortgages hadn’t been bad enough for Jack Welch, the chairman of General Electric—Kidder’s parent company—Welch also had to spend $7 million to cover derivative-related losses in five of Kidder’s money-market funds.37 Atlantic Richfield Company’s ARCO managed a $400 million money-market fund, and lost $22 million.38 United Services Advisors lost $93 million,39 Fleet Financial Group lost $5 million. And so on. In September 1994, a money-market fund called Community Assets Management Inc., in Denver, was liquidated, and there was no parent company to make up for losses. Community Assets became the first-ever money-market fund to lose money for its shareholders.
On June 1, 1994, Edward C. Johnson III, chairman of the Fidelity family of mutual funds, sent a letter to all of Fidelity’s bond investors, explaining the funds’ use of derivatives and blaming the recent losses on the interest-rate hike and the role of leveraged hedge funds, which he said “were caught by surprise by the magnitude of the rise in interest rates and the decline in bond prices and were forced to sell to meet margin calls.”40 Fund giant Vanguard distributed a similar notice in a bulletin called “Plain Talk About Derivatives.”
The losses at mutual funds affected millions of investors. Moreover, now that commercial banks were increasingly engaging in the securities business, it was difficult for investors to separate a risky investment in derivatives from a simple deposit at their local bank. Consider NationsBank, for example. In 1993, NationsBank received permission to begin operating a securities firm, called NationsSecurities, which immediately began selling Term Trusts to NationsBank clients.41 Term Trusts were 10-year-maturity bond funds that paid an extra one percent or more above the 10-year U.S. Treasury Yield. To a bank customer with money in a checking account or a certificate of deposit, this was a huge increase in yield.
Of course, the increase didn’t come without risk. Unlike a bank deposit, return of principal was very much at risk. The Term Trusts achieved a high yield by investing up to 40 percent of their assets in inverse floaters, which—as Orange County made clear—were not without risk.
NationsSecurities made the Term Trusts its first focus product, creating monetary incentives for its representatives to sell them. In the first months of the program, through February 1994, NationsSecurities sold more than $300 million of the Term Trusts, generating more than $16 million in fees. Sales representatives pitched the Term Trusts to investors in traditional Certificates of Deposit, telling NationsBank customers who never had invested in anything other than CDs that Term Trusts were as safe as CDs, but better because they paid more. NationsSecurities paid NationsBank employees referral fees for sending these customers.
The sales representatives were given detailed prospectuses with fine print describing the various risks associated with Term Trusts, but apparently they either ignored them or did not read them very carefully. Instead, the sales representatives told clients, as they had been told during presentations, that the Term Trusts were backed by the U.S. government and would not lose principal. On several occasions, a sales manager held up a picture of a Term Trust brochure with a picture of the U.S. Capitol Building on the cover, and said, “If the Capitol is standing in 10 years, these people will get their money back.” The representatives fed these pitches to their customers during “call nights,” when they emphasized the Term Trusts’ safety, predictability, and return. Some Texas representatives told customers that “until now you had to be Texas A&M or have the wealth of Ross Perot, to get access to this quality of management.” Some representatives even told customers that because the Term Trusts would return all of the principal payment in 10 years, there was no fee (in reality, the sales commission was a whopping 5.5 percent). These representatives didn’t discuss the details about all the inverse floaters the Term Trusts held.
NationsSecurities blurred the line it was supposed to maintain between itself and NationsBank. Even the names were almost interchangeable. NationsSecurities sent a mass mailing about the Term Trusts to more than a million NationsBank customers, in envelopes bearing the NationsBank logo and colors. Representatives were taught to call the Term Trusts “accounts at the bank”—rather than securities—and to say they were calling “from the bank.”
Not surprisingly, the Term Trusts did not perform very well. Recall that inverse floaters decline in value when short-term interest rates rise. After the Fed’s rate hike, the Term Trusts declined in value by more than a third. So much for protection of principal. (NationsSecurities later was censured by the SEC and fined $4 million.)42
The spread of financial innovation to more traditional and conservative financial intermediaries—particularly, money-market funds and commercial banks—created new dangers. How could individual investors discern the risks of these new menus of investment possibilities? The Securities and Exchange Commission had taken the position that stock and bond mutual funds should be able to invest in whatever they wanted, so long as they disclosed the risks. But what constituted adequate disclosure? And even with adequate disclosure, should an elderly banking client who had never bought anything other than a CD really be purchasing inverse floaters under any circumstances? The questions were not merely theoretical. NationsBank was one of several large banks entering the securities business to sell new financial products to millions of individual banking customers.
In addition, paying bonuses based on performance skewed the incentives of fund managers and bank salespeople. If they performed well, they made a lot of money for themselves. If they performed poorly, they lost someone else’s money. Federal law generally barred investment advisers from receiving incentive compensation for sales to small clients (people with less than $1 million of net worth or $500,000 in investments). But there was an exception for mutual funds, and financial institutions had shown, in recent years, a great ability to drive trucks through even small exceptions. Mut
ual funds had to compete with Wall Street for talent, and if they couldn’t pay big bonuses, they wouldn’t be able to attract the best managers. The more Wall Street traders made, the higher the fund managers’ compensation would have to be.
Still another problem was the lack of sophistication among fund managers. The data showed that most active managers underperformed market indexes. The Arbitrage Group at Salomon Brothers had been able to spot inefficiencies and outperform their peers, year after year. But not many fund managers were like John Meriwether’s traders. Certainly Robert Citron and Worth Bruntjen didn’t measure up. And the greatest damage had come from the smartest manager, David Askin.
Fund managers all seemed to be buying the same financial instruments, like gamblers all betting on red at a roulette wheel. Fortune magazine conducted a detailed survey and found only four money managers who had outperformed the major bond-market indices in 1994.43 It was remarkable, but nearly every professional money manager during the early 1990s had been speculating that interest rates would remain low.
The Federal Reserve had fueled this speculation by committing to keep interest rates low. In 1989, short-term rates had been higher than long-term rates, which were around eight percent. By 1992, short-term rates were three percent, while long-term rates were still near eight percent. As long as this interest-rate environment held, it was easy to make money, with just a little bit of (relatively painless) math.
Suppose you use $1,000 of savings to buy long-term bonds. If long-term rates are eight percent, you make $80 in a year, if rates don’t change. Not bad.
Now suppose that you also borrow $10,000 and invest the entire $11,000 amount in long-term bonds. You still make $80 on your $1,000 of savings. But now you also make $800 on your $10,000 of borrowings—$880 in all. If short-term rates were only three percent, you would only have to repay $300 of interest on a 1-year loan, so you would keep $580—a return of more than 50 percent on your money. And, in that example, your leverage—the ratio of your borrowings to your savings—was only 11-to-1. Financial institutions typically had leverage in the range of 25-to-1. Throughout the early 1990s, financial institutions could do this simple carry trade—borrowing short-term and investing long-term—and make a fortune. (“Carry” refers to the difference between short-term and long-term rates.)
Infectious Greed Page 17
