interest rate. What happens to the size of the time spread
C H A P T E R
2 2
Futures
Markets
791
Forward versus Futures Pricing
Until now we have paid little attention to the differing time profile of returns of futures and
forward contracts. Instead, we have taken the sum of daily mark-to-market proceeds to the
long position as P
T
2 F
0
and assumed for convenience that the entire profit accrues on the
delivery date. Our parity theorems apply only to forward pricing because they assume that
contract proceeds are in fact realized only on delivery. In contrast, the actual timing of cash
flows conceivably might affect the futures price.
Futures prices will deviate from parity when marking to market gives a systematic
advantage to either the long or short position. If marking to market tends to favor the long
position, for example, the futures price should exceed the forward price, because the long
position will be willing to pay a premium for the advantage of marking to market.
When will marking to market favor either a long or short trader? A trader will benefit
if daily settlements are received (and can be invested) when the interest rate is high and
are paid (and can be financed) when the interest rate is low. Because long positions will
benefit if futures prices tend to rise when interest rates are high, they will be willing to
accept a higher futures price. Therefore, a positive correlation between interest rates and
changes in futures prices implies that the “fair” futures price will exceed the forward price.
Conversely, a negative correlation means that marking to market favors the short position
and implies that the equilibrium futures price should be below the forward price.
For most contracts, the covariance between futures prices and interest rates is so low
that the difference between futures and forward prices will be negligible. However, con-
tracts on long-term fixed-income securities are an important exception to this rule. In this
case, because prices have a high correlation with interest rates, the covariance can be large
enough to generate a meaningful spread between forward and future prices.
So far we have considered the relationship between futures prices and the current spot
price. What about the relationship between the futures price and the expected value of
the spot price? In other words, how well does the futures price forecast the ultimate spot
price? Three traditional theories have been put forth: the expectations hypothesis, normal
backwardation, and contango. Today’s consensus is that all of these traditional hypotheses
are subsumed by modern portfolio theory. Figure 22.7 shows the expected path of futures
under the three traditional hypotheses.
Do'stlaringiz bilan baham: