### Abstract:

The economic efficiency related to the use of the range resource
is extremely important to an area such as Southeastern Oregon where
most of the gross income of the area is obtained from range livestock
production; important to the economic progress of the region and to
the development of a desirable social structure for the citizenry.
The investigation of the decision making problem under conditions
of uncertainty implies that the certainty problem can be solved for
all uncertain alternatives. The certainty alternatives are admissible
only when the economic efficiency criterion is met. The criterion
used in this study for decision making under uncertainty involved
maximizing expected utility through the use of a multivariate
Bayesian statistical model. The uncertain states of nature were
next year’s cattle price and forage production. The optimal strategy
involved the structure of the cattle inventory.
A set of equations was developed which generate the input-output
coefficients and the objective function values for a linear program
solution for any cattle inventory system with respect to expected
calving percentage, death loss, replacement policy, cattle weights
and prices, and for any number of time increments. A set of homogeneous livestock inventory systems were defined such that a
linear program model can be used to determine the optimal inventory
structure under certainty conditions. The link between the resource
equation system and the linear program improves the feasibility of
effectively getting large volumes of budgeted ranch data into an
optimizing framework.
Primary data were used to establish costs and returns and the
land use structure for representative units. Despite the variance
in physical structure that exist among the ranch population ranch
units, there are certain consistencies for which some general results
can be inferred:
1)Under conditions of certainty with respect to price and
forage productions, the optimal livestock inventory structure for the
study area would tend toward the production of yearling and other
steer beef. The study area ranch units tend toward the production
and sale of calve beef. It is postulated that this discrepancy
between what "should be" and "what is" is a result of the dependency
of the ranch unit on public lands. The public land input is commensurate
with an administrative definition rather than the physical
production relationship. This difference is defined as a misallocation
and the models developed can be used to quantify this
misallocation. This condition is independent of the present public
grazing fee structure.
2)The primary data indicate that the area could absorb an
increase of 20 percent in the total spring, summer, and fall range
forage with substantially the present resource structure. Range improvements have an expected marginal value product of $3.00 per
animal unit month.
3)It is meaningful through the use of the models developed
to think in terms of a general population utility function for
purposes of explaining the population's economic behavior with
respect to cattle inventory structure and for predicting economic
stimuli responses.
It is concluded that multivariate regression models in
obtaining a posteriori weights for decision making under uncertainty
can be formulated as an operational management tool. The combination
of subjective and objective evidence into the scientific
approach for decision making has wide use implications beyond firm
management problems.