When Genius Failed: The Rise and Fall of Long-Term Capital Management

by Roger Lowenstein

The first modern swap was engineered in 1981. IBM had bonds denominated in Swiss francs and German marks and wanted to convert this debt to dollars. David Swensen, a Yale Ph.D. newly arrived at Salomon, suggested that perhaps some other borrower could be persuaded to issue debt that, aside from being denominated in dollars, was identical to IBM’s. One obvious choice was the World Bank, which had an appetite for holding debt in a variety of currencies. As an inducement to borrow, Salomon gave the bank a slightly lower-than-market interest rate. Then the two borrowers switched—IBM winding up with dollar debt, the World Bank with the foreign stuff—and voilà! the world of swaps was born. The business grew—at first slowly, then exponentially. Soon, banks were swapping obligations for currencies, interest rate payments, equities, any future stream of cash that could be traded for another. In 1990, there were $2 trillion worth of interest rate swaps (just one type of derivative) outstanding. By 1997, the total had soared to $22 trillion.5 One offshoot—largely unintended—of this tremendous growth was that banks’ financial statements became increasingly obscure. Derivatives weren’t disclosed in any way that was meaningful to outsiders. And as the volume of deals exploded, the banks’ balance sheets revealed less and less of their total obligations. By the mid-1990s, the financial statements of even many midsized banks were wrapped in an impenetrable haze.