VII. Debt Financing 24. Valuing Debt
implies that such naïve strategies won’t work. If short rates are lower than long
rates, then investors must be expecting interest rates to rise. When the term struc-
ture is upward-sloping, you are likely to make money by borrowing short only if
investors are overestimating future increases in interest rates.
Even on a casual glance the expectations theory does not seem to be the com-
plete explanation of term structure. For example, if we look back over the period
1926–2000, we find that the return on long-term U.S. Treasury bonds was on aver-
age 1.9 percent higher than the return on short-term Treasury bills. Perhaps short-
term interest rates did not go up as much as investors expected, but it seems more
likely that investors wanted a higher expected return for holding long bonds and
that on the average they got it. If so, the expectations theory is wrong.
The expectations theory has few strict adherents, but most economists believe
that expectations about future interest rates have an important effect on term struc-
ture. For example, the expectations theory implies that if the forward rate of inter-
est is 1 percent above the spot rate of interest, then your best estimate is that the
spot rate of interest will rise by 1 percent. In a study of the U.S. Treasury bill mar-
ket between 1959 and 1982, Eugene Fama found that a forward premium does on
average precede a rise in the spot rate but the rise is less than the expectations the-
ory would predict.
20
The Liquidity-Preference Theory
What does the expectations theory leave out? The most obvious answer is “risk.”
If you are confident about the future level of interest rates, you will simply choose
the strategy that offers the highest return. But, if you are not sure of your forecast,
you may well opt for the less risky strategy even if it offers a lower expected return.
Remember that the prices of long-duration bonds are more volatile than those
of short-term bonds. For some investors this extra volatility may not be a concern.
For example, pension funds and life insurance companies with long-term liabili-
ties may prefer to lock in future returns by investing in long-term bonds. However,
the volatility of long-term bonds does create extra risk for investors who do not
have such long-term fixed obligations.
Here we have the basis for the liquidity-preference theory of the term struc-
ture.
21
If investors incur extra risk from holding long-term bonds, they will de-
mand the compensation of a higher expected return. In this case the forward rate
must be higher than the expected spot rate. This difference between the forward
rate and the expected spot rate is usually called the liquidity premium. If the
liquidity-preference theory is right, the term structure should be upward-sloping
more often than not. Of course, if future spot rates are expected to fall, the term
structure could be downward-sloping and still reward investors for lending long.
But the liquidity-preference theory would predict a less dramatic downward
slope than the expectations theory.
680 PART VII
Debt Financing
20
See E. F. Fama, “The Information in the Term Structure,” Journal of Financial Economics 13 (December
1984), pp. 509–528. Evidence from the Treasury bond market that the forward premium has some power
to predict changes in spot rates is provided in J. Y. Campbell, A. W. Lo, and A. C. MacKinlay, The Econo-
metrics of Financial Markets, Princeton University Press, Princeton, NJ, 1997, pp. 421–422.
21
The liquidity-preference hypothesis is usually attributed to Hicks. See J. R. Hicks, Value and Capital:
An Inquiry into Some Fundamental Principles of Economic Theory, 2nd ed., Oxford University Press, Ox-
ford, 1946. For a theoretical development, see R. Roll, The Behavior of Interest Rates: An Application of the
Efficient-Market Model to U.S. Treasury Bills, Basic Books, Inc., New York, 1970.