Draft 6 RIDING THE YIELD CURVE:AVARIETY OF STRATEGIES SEPTEMBER 2005 I n its simplest form, the rational expecta- tions hypothesis of the term structure of interest rates (REHTS) posits that in a world with risk-neutral investors, the n-period long rate is a weighted average of the future spot rates and thus any 1-period forward rate is an unbiased predictor of the corre- sponding future 1-period spot rate. Conse- quently, the expectations hypothesis implies that with the possible exception of a term pre- mium, the holding period returns (HPRs) of a class of fixed-income instruments are iden- tical, independent of the instruments’ original maturity. 1 Under this assumption, for example, the returns from purchasing a 3-month gov- ernment security and holding it until maturity and the returns from purchasing a 12-month government security and holding it for 3 months are identical. The strategy of pur- chasing a longer-dated security and selling it before maturity is referred to as “riding the yield curve.” If the REHTS holds, then, for any given holding period, riding strategies should not yield excess returns compared with holding a short-dated security until maturity. Any evi- dence of persisting excess returns from such trading strategies would indicate the existence of risk premia associated with the term struc- ture. The body of literature on different tests of the expectations hypothesis is very large, and overall the results remain inconclusive. 2 Although the majority of tests of the expectations hypothesis are hinged on testing for the predictive power of forward rates in terms of future sport rates, there is a small strand of literature that examines the persis- tence of excess returns from riding strategies across different holding horizons with different maturity instruments. In their seminal paper, Dyl and Joehnk [1981] examine different riding strategies for U.S. T-bill issues from 1970 to 1975 and find that there are significant, albeit small, excess HPRs to be made from riding the yield curve. They use a simple filter rule based on break-even yield changes in order to quantify the ex ante riskiness of riding the yield curve. Based on this filter, their results indicate that the returns increase with both the holding horizon and the maturity of the instrument. Grieves and Marcus [1992] are able to produce similar results by looking at a much longer time series of monthly zero-coupon T- bill rates from 1949 to 1988. They apply the same filter rule as Dyl and Joehnk to identify, ex ante, under what type of yield curve envi- ronment excess returns from rolling can be anticipated. Although their results confirm that longer-maturity rides outperform the simple buy-and-hold strategy of the short-term instrument, they conclude that, on a risk- adjusted basis, longer rides perform slightly worse because of increased interest rate risk. Overall, they find evidence against the pure form of the expectations hypothesis since it appears that profitable trading strategies have gone unexploited. Using daily closing prices Riding the Yield Curve: A Variety of Strategies DAVID S. BIERI AND LUDWIG B. CHINCARINI DAVID S. BIERI is adviser to the general manager at the Bank for International Settlements in Basel, Switzerland. david.bieri@bis.org LUDWIG B. CHINCARINI is an adjunct professor and financial consultant at Georgetown University in Washington, DC. chincarinil@hotmail.com