An Empirical Analysis of Interest Rate Swap Spreads

DONALD J. SMITH is associate professor of finance and economics at the School of Management of Boston University. nterest rate swaps are, without a doubt, one of the major financial innovations of the 1980s. The swap market is immense; total notional principal in U.S. l dollar swaps outstanding is now measured in the trillions of dollars. In a 1993 survey by the trade publication Treasury and Risk Management, 83% of corporate respondents indicated that they expected to use interest rate swaps in the coming year, while only 17% expected to use listed interest rate futures and options. In a typical swap agreement, two counterparties exchange streams of fixedand floating-rate interest payments, such that underlying fixed-rate debt can be transformed into floating-rate debt, and vice versa. Essentially, then, an interest rate swap is a series of forward contracts on some reference interest rate, such as LIBOR (the London Interbank Offered Rate). Beidleman [1991] has provided an extensive analysis of the development of these agreements. There has been little empirical research on interest rate swaps in the academic literature.' An obstacle to this line of inquiry surely has been the general dearth of data, due in large part to the fact that swaps are transacted in private, over-the-counter markets. While the International Swap Dealers Association, an industry trade group, provides survey data on an aggregate basis, there has been no public source for swap price and transaction quotations. In this study we use an extensive data set of weekly swap rate quotations compiled between 1985 and 1991 at Salomon Brothers Inc. Our objective is to analyze and empirically explain volatility in the swap spread, which is the main pricing variable in an interest rate swap. More precisely, the swap spread is defined as the difference between