Using a number of long-term maturities and monthly data, 1989-1997, we prov
ide a number of tests of the expectations hypothesis (EH) of the term struc
ture. The main insight in this paper is the use of the excess holding perio
d return to provide a proxy for a possible time-varying term premium. Nearl
y all previous studies using the VAR methodology have used only the spread
and the change in (short) rates and they have ignored the excess holding pe
riod return. We find that we cannot reject the EH, but we do reject the pre
sence of time-varying risk premia. Copyright (C) 2001 John Wiley & Sons, Lt
d.