Knowledge that Transforms

Overhead costs, Cost allocation

This study is a sequel to an earlier computer simulation of foreign accounting principles, the results of which were published in the Autumn, 1966 issue of this journal.' In the present study, the actual transaction data of two U.S. manufacturing firms, the same data that had been used previously for a multinational comparison, were used in a computer model to simulate prevailing U.S. and German Accounting practices. The aim was to obtain some notion of the different effects of U.S. and West German accounting practices on financial statements, assuming the underlying transactions are the same. The study is divided into five parts:

Valuation, Depreciation, Asset valuation, Historic costs

upon establishing a means of distinguishing which of two alternative disclosure treatments is the more -useful. The traditional means of attempting to make such distinction is through rational argumentation. Rational argumentation is very useful in exploring and drawing out the logical implications of alternative treatments, but may not be sufficient to enable selection of the better treatment. Recently attempts have been made to employ the Predictive as the means of distinguishing the better of two alternative accounting measurements. This criterion selects as the better of two alternative accounting measurements the one which has the greater to predict a given event considered to be of particular importance.' Surely the application of the predictive ability criterion in accounting research can accomplish much, but continual effort should be made to bring other research methods to bear on the problem of improving accounting disclosure. This paper presents an effort to develop a criterion

Depreciation, Straight-Line depreciation, Investment credit

In a recent article, F. K. Wright proposes a linear programming approach to the measurement of asset services.' He bases his work on the theories and shadow-pricing developed within linear programming. Similar uses of shadow prices are to be found in other papers on accounting research. Usually, in a linear programming context, one need not distinguish between the gain in having one more unit of a resource, as opposed to the loss in having one less unit. The purpose of this note is to show, by means of an example, the need for making such distinctions. I give a simple illustration and then discuss its relation to general linear programming theory and its implications for accounting research.

Performance Measurement, Direct costs, Absorption costing, Ex post optimum model

Depreciation, Income tax, Tax allocation

Although accounting has traditionally assumed a relationship between a firm's historical transaction data and its future performances, there is increasing interest in assessing this relationship directly in terms of specific predictions of the firm's future earnings. The most recent statement of accounting theory published by the American Accounting Association emphasizes the importance of this aspect of accounting in the following remarks:

This paper reports an empirical test of the usefulness of a prediction method for multivariate budgets suggested by Theil.' The method, based on information theory concepts, is described in the first three sections, and results of the test are presented in the remaining sections. These results show that the suggested prediction method outperforms both the actual forecasts made by the firms who provided the data and two naive forecasts.