A Farm Credit Debt Selection Model: Description and Application

Abstract

The primary funding objective of a Farm Credit Bank is to obtain the necessary debt funds for its lending operations at the lowest possible cost. Fulfilling this objective entails decisions of participation in the various Farm Credit System securities. Participation decisions in present and future securities are based upon present and future debt needs and costs. Because future debt needs and costs are not known with certainty, participation decisions are necessarily complex and difficult. Complexity exists because of the numerous debt participation options that are possible. Even if future interest rates could be known with certainty, they often peak and ebb at various times. It is, therefore·, a tedious job to determine the future debt participation options that would result in the lowest cost. However, because future interest rates are not known with certainty, participation decisions become very difficult. Participation decisions are often based upon expected debt needs and costs. Unfortunately, selecting the lowest expected cost participation strategy is no guarantee that the selected strategy will in fact be the lowest cost strategy, since actual costs may deviate greatly from expected costs. This manuscript discusses a technique, called quadratic risk programming, that sorts through all participation possibilities and selects strategies that have low expected costs and low risks, such that actual costs will not deviate greatly from expected costs. The model does not forecast interest rates or debt needs. Rather, it uses projections provided by the user to generate low expected cost and low risk debt participation strategies. The model is user oriented

it queries the operator for information and data that it needs. The remainder of this manuscript is divided into sections discussing the details and operation of the debt selection model. The first section covers the basic concepts of the quadratic risk programming model using a simple example. Then, the characteristics of the model for selecting Farm Credit securities are specified. The third section is a user's guide for operating the model. The final section is a technical section which specifies the assumptions and equations used in the model. The Appendices include a sample application as well as a listing of the computer programs.