Masters Theses

Date of Award


Degree Type


Degree Name

Master of Science


Agricultural Economics

Major Professor

Dan L. McLemore

Committee Members

Emmit Rawls, Keith Phillips


An efficient spatial organization of livestock auction markets in Tennessee would improve net prices received by livestock producers and/or reduce the cost of livestock to buyers. This study used separable programming to build a spatial model of Tennessee's livestock auction market industry. The purpose of the model was to determine the optimal sizes, number, and locations of auction markets that minimize the combined annual costs of assembling (transporting) and marketing livestock through auctions in the State. Once this optimal solution was found, the model was re-solved under varying assumptions about livestock numbers and cost levels to analyze the effects of changes in these parameters on the optimal solution. Livestock numbers were both increased and decreased by 10 and by 25 percent. Transportation cost and marketing cost were each increased by 10 and by 25 percent.

Since few barriers to interstate movement of livestock exist, the area of study consisted of Tennessee and parts of surrounding states within 50 miles of Tennessee's border. This area encompassed 238 counties, each of which was considered an origin for livestock and a potential market site in the mathematical model. Data for input into the separable programming model came from several sources. Estimates of the expected annual volume of livestock marketed in the supply area were derived from livestock inventory data for each county. Transportation cost was estimated using an economic engineering approach to develop transportation cost budgets for "typical" loads of livestock. Transportation cost was estimated to be $0,226 per mile per livestock unit in 1983. These "typical" loads of livestock were identified from the results of surveys of 275 individuals hauling livestock to auction markets in Tennessee. The cost of transporting a livestock unit was computed for all potential market destinations within 50 air-miles of each origin. The cost of marketing was estimated by Spielman in a previous study (1978) and was adjusted to reflect costs in 1983.

The basic model (Model I) was expanded to include the reduction in buyer operating costs due to increased market volume. This modified model was designated as Model II. It was hypothesized that buyers realize cost savings by attending auction markets with "large" volumes. These costs savings were thought to exist for two reasons. One reason is that at "large" sales, buyers should be able to acquire full, uniform loads for shipment, eliminating the cost of intermediate assembly between auction markets. The second reason cost savings may exist is that if a buyer must attend more "small" sales to fill orders than would have been necessary if "large" sales had been attended, extra costs accrue in the form of travel time, mileage, food, and lodging.

The negative relationship between buyer costs and market volume was included in the mathematical model by using a positive relationship between market price and market volume as a proxy for the buyer cost savings attributable to market volume. This positive price--volume relationship was estimated using regression analysis on 1982 and 1983 price and volume data from 16 Tennessee auction markets for feeder cattle, cull sows, and cull cows.

The results of Model I show the optimal number of auction markets in Tennessee to be 47, a reduction of seven from the 54 markets actually in operation during 1983. This solution was somewhat stable in response to changes in the model's parameters. With one exception, in all variations of Model I a reduction from the currently existing market number was shown to be desirable. For the model version with transportation cost increased 25 percent above 1983 levels, an increase in market numbers to 56 was found to be optimal for the State.

For Model II, which considers the reduction in buyer costs associated with larger market volume, the optimal number of markets in Tennessee was found to be 19. This solution was very stable under changes in the cost parameters of the model, but was sensitive to changes in livestock numbers. In response to a 25 percent increase in livestock numbers, market numbers increased by four in the optimal solution for Tennessee. When livestock numbers were decreased by the same percentage, market numbers declined by five.

The results of this study show that improvements in the efficiency of Tennessee's livestock auction market industry are possible through a reduction in the number of markets in the State. The policy implications of this study are that licensing of new markets in the State should be done with consideration of the optimal locations identified in this study, and with the long term goal of reducing the number of auction markets in the State.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."