Opportunity Costing: An Application of Mathematical Programming

The design of control systems in large companies is recognized as a formidable task. A large company usually contains many different productive activities which, although separately identifiable, have important connections with each other. Different products are often competing for the same inputs, or the outputs of one activity are the inputs of another. The problem, as is well known, is to design a system whereby each manager of a product or activity acting in his own interests also achieves the most profitable operation for the company as a whole. In this paper an attempt is made to develop a costing system which would assist a large organization in arriving at its optimal position. An example is used to indicate how the techniques of transfer pricing which have been developed at theoretical levels elsewhere [5, 9, 11] may be adapted to fit into an accounting system. Certain difficulties which are not discussed in the theory of the subject become apparent when an attempt is made to put the techniques into practice; these are discussed below. The system developed is based on the use of the shadow prices which are reflected in the solution of a mathematical programming problem, and these shadow prices are incorporated into a system of responsibility accounting. The use of such techniques in accounting is not new. One of the first articles appeared in Accounting Research in 1954 [2]. Although much work has been done with mathematical programming [3-6, 10, 11], little work has appeared in which programming techniques have been linked with the accounting tools for planning and control. Such co-ordination is the concern of this paper. Programming can be regarded by the accountant as the natural extension of the principle of marginal costing. If an organization has only one