Infectious Greed

by Frank Partnoy

  Sanford deliberately tried to create anxiety among his managers. The head of human resources at Bankers Trust described Sanford as an “obsessive-compulsive neurotic.” But in the aftermath of Andy Krieger’s losses, Sanford was briefly circumspect. He had learned the importance of a more-stable source of revenue that did not require the bank to put so much capital at risk. Ira Stepanian, the CEO of Bank of Boston—one of Sanford’s competitors and friends—had expressed concern about Bankers Trust’s aggressive practices: “Any company that makes a lot of money from trading can also lose that much. It all depends on having the right controls in place, and I’m sure that Charlie Sanford has thought a lot about that.”19

  Sanford had thought a lot about controls. He downplayed Bankers Trust’s cowboy image, saying “we don’t bet the bank.”20 He admitted Bankers Trust had made some bad decisions, as in a $100 million unsecured loan to Donald Trump (“We were brain dead when we made that loan”21), but he continued to defend Krieger’s trading (“We didn’t take too big a position for Bankers Trust, but we may have taken too big a position for that market.”)22 He focused on the bank’s risk-management systems, which were state-of-the-art, but obviously needed improving.

  Sanford had initiated the idea of measuring the profitability of a particular business by its Risk-Adjusted Return On Capital (he used the acronym RAROC, pronounced RAY-rock), rather than simply looking at returns.23 RAROC was a revolutionary concept, which many large corporations soon adopted.

  The idea behind RAROC was that a company should reward employees based not only on how much money they made, but also on how risky their business was. A particular business—currency-options trading, for example—might generate huge returns; but, if it was too risky, the bank might be better off allocating its resources somewhere else. Employees could take on risks, but they needed to be smart about it. Bankers Trust would force employees to be smart by creating incentives for them to generate consistent returns relative to their risks.

  RAROC required a kind of thought experiment. Bankers Trust would look at the history of a particular market and hypothetically allocate enough capital to cover expected losses in that market for one year at, say, the 99 percent level.24 That meant employees in risky businesses needed more of this hypothetical capital, so their returns appeared smaller relative to that capital. In contrast, less risky businesses looked more profitable, because they needed less of this hypothetical capital.

  Different divisions within Bankers Trust competed for capital, and were judged—on a risk-adjusted basis—against each other. Andy Krieger, for example, made a lot of money, but the business of currency-options trading was risky, so his RAROC was lower. That was one reason why his bonus had been only $3 million, even though his profits were so high. Equity derivatives in Japan had been an attractive business because it didn’t put much capital at risk, and those salesmen were very well paid.

  In this way, Sanford’s RAROC system encouraged employees to focus both on how much money they were making for Bankers Trust, and on how much of the firm’s precious capital they were risking. Much more than other banks, Bankers Trust gave its employees a road map for how they could make big bonuses. As a former member of Bankers Trust’s management committee put it, Sanford “set the tone in a simple way—compensation.”25

  As the pressures mounted, Bankers Trust personnel at all levels began to push the edges of the law. Since the 1930s, a law called the Glass-Steagall Act had separated investment and commercial banking, in response to public outcry about banks’ alleged abuses in the stock-market mania before the Great Crash of 1929. The solution had been to split the two functions of banking. Commercial banks were left with the traditional business of making loans and taking deposits. Investment banks would trade and underwrite securities.

  Not surprisingly, the law was unpopular among commercial bankers. As early as 1982, Sanford was saying, “We’re hoping to see some break in Glass-Steagall.”26 Although Congress would not repeal Glass-Steagall for more than a decade, Charlie Sanford began running circles around this law, finding loopholes and lobbying regulators to make exceptions for Bankers Trust.

  For example, Bankers Trust had lobbied the administration of President Ronald Reagan to deregulate the process of securities offerings so that banks could participate. Regulators began allowing so-called shelf registrations, in which companies—instead of processing from scratch all of the documentation necessary to create new securities each time they wanted to raise money from investors—created the documentation upfront and put it “on the shelf,” to be used at a later date. That way, companies could issue new stocks or bonds much more quickly and at lower cost. Only securities firms—which Bankers Trust was not—could actually underwrite the securities, putting them on the shelf. But Sanford had argued that, once that was done, anyone, including Bankers Trust, could use the shelf registration to distribute new securities. Officials at the Securities and Exchange Commission winked and nodded, and Bankers Trust got its first foothold in the securities business.

  Bankers Trust began distributing securities—mostly bonds—to institutional investors through deals known as private placements. In a private placement, the borrower did not need to register under the securities laws, and did not need to comply with generally accepted accounting principles. The bank aggressively pursued these deals, doing about $1 billion of them in 1983, and billions more in later years,27 even though the Securities and Exchange Commission did not formally allow companies to raise funds in this way. (In April 1990, the SEC approved the now widely used Rule 144A, which allows companies to do private placements with so-called Qualified Institutional Buyers—basically, large companies and very rich people.)

  Bankers Trust entered other new businesses, too, arranging hundreds of interest-rate and currency swaps, and dealing in government bonds and short-term corporate debt. It was a lender in many leveraged buyouts—the high-profile merger deals developed in the 1980s—in which corporate raiders purchased stocks with borrowed money.28 Sanford also began selling and trading billions of dollars of the bank’s own loans, a profitable practice that other banks soon followed.29

  By the mid-1980s, Bankers Trust was starting to look a lot like an investment bank—something Glass-Steagall supposedly had made illegal. Regulators didn’t like Sanford’s reluctance to hold loans, and one banking regulator reportedly suggested Bankers Trust should give up its bank charter.30 But the bank insisted that it wasn’t really an investment bank; it was simply acting as an agent in a few deals, nothing for the regulators to fret about. As one Bankers Trust veteran put it, “We are obviously law-abiding citizens, but we mean to press on all legitimate fronts.”31

  The bank’s experience in the market for interest rate and currency swaps was critical to its plans for the early 1990s. (Recall that swaps are contracts where parties agree to exchange payments based on interest rates or currencies.) The competition for swaps in the mid-1980s had been fierce. By 1984, interest-rate swaps were a $70 billion business, considered huge at the time.32 But profit margins were dwindling, and various regulators—including the Federal Deposit Insurance Corp., the Federal Reserve Board, and the Comptroller of the Currency—were starting to ask questions about the risks of swaps.

  There were a few ugly stories about firms using swaps to manipulate their accounting results. One bank contemplated internal swaps—swaps with itself—whereby it would set aside reserves depending on how much profit it wanted to declare in a particular quarter. In 1984, Morgan Stanley began experimenting with delaying the starting dates of swaps, which made them more difficult to evaluate.33 Japanese banks in particular used swaps to inflate profits and hide losses.

  In February 1985, the Financial Accounting Standards Board (FASB)—the private group that established most accounting rules—asked whether banks should begin including swaps on their balance sheets, the financial statements that recorded their assets and liabilities.34 Banks already accounted for loans as assets, because the right to recei
ve loan payments from borrowers was valuable. Banks also accounted for deposits as liabilities, because the obligation to pay depositors was costly.

  However, since the early 1980s, banks had not included swaps as assets or liabilities, arguing that swaps were different from loans and deposits, and belonged off their balance sheets. In a swap, the parties agreed to exchange payments based on some reference amount, say, $100 million, but—unlike borrowers who repaid the principal amount of their loans, and depositors who could withdraw the entire amount of their deposit—the parties to a swap never actually paid each other the underlying reference amount. Instead, parties merely used the reference amount to calculate the amount they owed each other at different times. (For example, a swap contract might require one party to pay 10 percent every year. If the reference amount were $100 million, that party would pay $10 million per year, but would never pay the underlying $100 million reference amount.)

  Thus, the argument went, a swap was not really an asset or a liability, even though it carried both the right to receive payments and the obligation to make payments. Moreover, banks argued, it wasn’t fair to require them to list swaps as assets and liabilities on their balance sheets, because they often used swaps to reduce their risks, and if banks were forced to record the fluctuation in the value of swaps over time, investors would perceive that the banks were riskier than they really were.

  The banks’ argument was deeply flawed. The right to receive money on a swap was a valuable asset, and the obligation to pay money on a swap was a costly liability. The fact that swaps, unlike loans and deposits, didn’t require repayment of the reference amount meant only that swaps would have different values from loans and deposits, but it didn’t justify removing them from the balance sheets altogether. Moreover, if swaps, indeed, were fluctuating in value, as banks claimed, that fact was important to the banks’ shareholders. If banks wanted to explain in a footnote of their financial statements why shareholders shouldn’t worry about these fluctuations (e.g., because the swap actually was reducing some other risk), nothing prevented the bank from doing so.

  But bankers knew that the fluctuations in their swaps would worry their shareholders, and they were determined to keep swaps off their balance sheets. FASB’s inquiry about banks’ treating swaps as off-balance sheet—a term that would become widespread during the 1990s—mobilized and unified the banks, which until that point had been competing aggressively for business, and not cooperating much on regulatory issues. All banks strongly opposed disclosing more information about their swaps, and so they threw down their swords and banded together at several high-level meetings, temporarily suspending their fierce competition. Within weeks after FASB’s February 1985 inquiry, the ten largest swaps dealers had formed the International Swap Dealers Association.

  Bankers Trust was one of the founding members of ISDA. According to Jonathan Berg, a vice president at Bankers Trust at the time, the banks formed ISDA to “organize before any problems arise.”35 The dealers planned to establish standardized documentation and practices, to lobby against new regulations, and to determine how big the swaps market really was. Estimates were in the $100 billion range, but it was embarrassing that no one knew such a basic fact, especially now that regulators were asking some hard questions.

  ISDA’s first press release said the group’s goal was to “advance general market practices and to discuss issues of relevance to the financial community.”36 But everyone involved understood that the primary role would be to lobby against regulation of swaps, or—as one commentator put it—to “explore the accounting and regulatory implications.”37 ISDA would prove to be the most powerful and effective lobbying force in the recent history of financial markets.

  The ten original dealer members of ISDA were Bankers Trust, Citicorp, First Boston, Goldman Sachs, Kleinwort Benson, Merrill Lynch, Morgan Guaranty Trust, Morgan Stanley, Salomon Brothers, and Shearson Lehman Brothers; together, they made up 80 percent of the swaps market.38 With the exception of a few mergers and name changes, the list of top dealers would look essentially the same for nearly two decades.

  In 1989, the Federal Reserve permitted some commercial banks—including Bankers Trust—to form investment-banking subsidiaries to engage in the securities business, but only if total revenues from investment banking were less than 10 percent of the bank’s total revenues. Bankers Trust formed a securities subsidiary, not too creatively named BT Securities Corporation. Its competitors followed, and Sanford predicted that universal banking would soon come to the United States, meaning that banks would offer securities, banking, and insurance services under one roof.39 The banks, having just tasted the profits from equity derivatives, began circling the securities waters like sharks that had just tasted blood.

  The Fed helped banks in another way: by lowering short-term interest rates, which had been above eight percent. Fed Chairman Alan Greenspan wanted to encourage investors and institutions to borrow more to invest in various aspects of the economy, including long-term bonds and stocks. This policy created the perfect environment for investors who wanted to borrow short term, betting interest rates would stay low.

  However, the Fed was not all-powerful, and it could not make the economy turn on a dime. The year 1990 was a horrific one on Wall Street. Every financial asset collapsed: junk bonds, commercial real estate, investments related to savings and loans, even Japanese stocks. On February 13, 1990, Drexel Burnham Lambert—Michael Milken’s bank—which had epitomized the 1980s financial culture, filed for bankruptcy. That marked the end of an era, and it seemed unlikely that banks would ever again generate 1980s-like profits. Even traditionally profitable businesses were hurting: stock commissions paid to Wall Street during 1990 were a relatively paltry $8.9 billion.

  Wall Street, by becoming so ruthlessly efficient, was destroying itself. New markets became competitive more quickly than ever, with profit margins narrowing, within months, as other financial institutions entered the business. Bankers Trust was in a quandary: every time it discovered a new source of profit, it was victimized by this inevitable competition. Yes, Bankers Trust’s employees had quantitative and risk-management skills that were superior to those of any bank. But how could it use them to make a consistent profit?

  The answer came from an unlikely source. Merton Miller, a Nobel-laureate economist at the University of Chicago, was an enthusiastic supporter of swaps and other derivatives. He agreed with many dealers that swaps were important financial innovations that enabled parties to allocate risks more efficiently.

  But Miller also believed these instruments served a different, perhaps more dubious, purpose. In 1986, Miller argued that a major impulse for financial innovation was a desire to avoid regulation.40 Financial-market participants faced very different regulatory environments, and they often were governed by rules they would prefer not to follow. By using derivatives, they could avoid rules that seemed senseless to them. Miller was a free-market economist and thought these rules were senseless, too, so he applauded companies’ efforts to avoid regulation, efforts that came to be called “regulatory arbitrage.”41

  The accounting rules for swaps were a perfect example. Swaps were simply investment contracts, and they were economically equivalent to many other financial instruments, including exchange-traded futures and options, all of which were regulated. Yet swaps were unregulated and immune from most securities-law disclosure requirements. Merton Miller’s insight implied that companies would do swaps not necessarily because swaps allocated risk more efficiently, but rather because they were unregulated. They could do swaps in the dark, without the powerful sunlight that securities regulation shined on other financial instruments.

  And here was the crucial point: to the extent companies and their financial officers could use custom-tailored swaps to avoid regulation or to hide risks, Bankers Trust’s profits from selling swaps to those companies might not disappear so quickly. Corporate treasurers hoping to benefit from such swaps would pay a premium—it wa
sn’t their money, after all—if the swaps were structured in a way that created more opportunity for profit, but hid the risks from their bosses. Just as the Japanese insurance executives paid large fees for equity derivatives, even fully informed financial officers would pay a hefty premium if they could avoid disclosing the risks of swaps to shareholders and regulators. Unlike the plain-vanilla instruments from earlier years, these swaps could have all sorts of hidden bells and whistles, and shareholders and regulators might not ever even know about them.

  Substantial fees from these complex swaps would persist so long as companies valued the difference in legal treatment between swaps and their regulated brethren. And if other banks were reluctant to sell such swaps to companies in the shadow of the law, there wouldn’t be as much competition for Bankers Trust.

  Finally, a business that might last for more than a year! Bankers Trust began gearing up to sell custom-tailored swaps.

  Gibson Greetings, Inc., was a Cincinnati, Ohio, company that made and sold greeting cards. Gibson had a traditional commercial-banking relationship with Bankers Trust, dating back to 1983.42 To finance its business, Gibson borrowed money from various sources, including a standard revolving-credit agreement arranged by Bankers Trust.

  In May 1991, Gibson borrowed $50 million in a straightforward debt deal arranged by a different bank, at an interest rate of 9.33 percent.43 As interest rates fell during fall 1991, Jim Johnsen, the company’s treasurer, began to explore the idea of using interest-rate swaps to reduce the interest rate on its debt.44 In an interest-rate swap with a bank, Gibson could agree to receive a fixed rate (offsetting the fixed rate on its debt deal), and pay a much lower floating rate. In other words, Gibson could reduce its effective interest rate in exchange for the risk that rates might rise, just as a homeowner could switch from a fixed-rate mortgage to an adjustable one. Such plain-vanilla swaps were well established by 1991, and margins were low. A basic $50 million interest-rate swap would cost around $50,000.