MARKETS AND LENGTHS OF HEDGES STUDIED
Three fed cattle markets in the South and
Southern Plains were selected for use in the study.
These were Kentucky, Georgia, and the Southern
Plains region of Texas and Oklahoma. Omaha was
selected as the contract delivery point reference
market. Prices used for the four markets were weekly
average prices as reported by the USDA Market News
Reporting Service. Futures contract prices used were
weighted weekly averages of daily prices. Minor
reporting differences occur in the cash price series.
Prices for Omaha and Kentucky are reported from
terminal market sales while Georgia prices are
reported on an at-plant direct basis, and Southern
Plains prices are reported F.O.B. feedlot with a 4
percent shrink. While these reporting differences have
some minor effect on the level of price differentials
among markets, no effect on variances occurs. No
scalar change in reporting basis is involved as would
be the case in Uve-Weight versus carcass-weight
comparisons.
A total of 21 successive Uve cattle futures
contracts and their hedging results were observed
from January 1969 to June 1972. This period
encompassed a structural change in the Uve cattle
futures contract in that, beginning with the August
1971 contract, Omaha became the par deUvery point.
Before, Omaha had been a delivery point but at a 75φ
per hundredweight dis∞unt. At the same time,
Guymon, Oklahoma, which is in the Southern Plains
marketing region, was designated a delivery point at a
$1 per hundredweight discount. The discontinuity
caused by this structural change was accounted for in
the analysis by the use of within-contract variances
only, that is, the variances within the 2-month span
of each contract. Since the change occurred between
contracts, this procedure abstracted from its effect.
The Southern Plains market was considered to be a
distant market despite the location of a delivery point
within it. Justification for this lay in the fact that the
deUvery point was estabUshed only for the last year
of the 3-1/2 year study period and because, as shown
by Crow, Riley, and PurceU [2], the deUvery
discount is so UnreaUsticaUy large as to render the
point ineffective anyhow.
Three lengths of hedge were postulated for
purposes of the study. These were a long-term
(30-week), a medium-term (21-week), and a
short-term ( 13-week) hedge. These hedge lengths
correspond to feeding periods required to carry light,
medium, and heavy weight feeder steers to a finished
weight of 1,000-1,050 pounds, as shown by National
Research Council rate of gain standards [5].
HEDGING MODEL
The procedure used to calculate hedging revenue
was as foUows:
(ɪ) ¾jt = Pft + Sjm ~ ɛmt
where:
Rjj^ is hedging revenue in market i for hedging
period j in week t,
Pjt is the average price for choice steers in market
i, weekt,
Sjm is the price at which cattle were sold short in
the week prescribed by hedging period j in the
deUvery month m ∞ntract, and
Lmt is the price at which the same contract was
purchased in week t.
The model is descriptive of the hedging process that
was postulated, with calculation of hedging revenues
being oriented on the marketing date. Hedges were
assumed to be placed routinely at the weekly average
price 30, 21, and 13 weeks prior to the marketing
date. Choice steer futures contracts are established
for delivery every other month, February through
December. For marketings in a deUvery month,
hedges were assumed to be placed in that contract up
to the week containing the 20th of the month, the
date on which contracts normally expire. Marketings
for the latter part of the month were hedged in the
succeeding contract, as were marketings in the
following month.
The model takes an ex post view of the hedging
process, in which results are measured On the basis of
realized revenues? Hedging revenues are compared
on a hundredweight for hundredweight basis, so the
model does not provide for portfolio-type analyses of
hedging strategies, such as the derivation of
hedged/unhedged inventory ratios. The model
abstracts from commission charges and interest
charges on margin deposits since these would tend to
be equal in all markets.
Hedging revenues were generated on a weekly
basis for the study period, and pooled variances were
calculated for comparisons between markets. Pooled
cash market price variance is defined as:
1An ex ante view of hedging can also be adopted, as shown in [1] and elsewhere, in which hedge placement is the
reference period for measuring results. Ex ante measures have the advantage of being independent of length of hedge, but on the
other hand they are dependent on the hedgers’ basis expectations. Since expectations cannot be measured from market data, the
ex post formulation was adopted here.
74