FUNCTIONAL FIXAWON IN ACCOUNtING Ij

FUNCTIONAL FIXAWON IN ACCOUNtING
Ijiri, Jaedicke, and Knight [1966] discussed the conditions under which a decision maker might be unable to change his decision process in response to a change in the underlying accounting process which supplies him witt. decision data. If, for example, the accounting department changed from the FIFO method to the LIFO method of costing
inventory, under what conditions might a manager who used inventoiry cost data in decision making fail to adjust his decision process to take into account the change in accounting method? Ijiri, Jaedicke, and Knight suggested that functional fixation might inhibit such an adjustment. They stated: "Psychologists have found that there appears to be 'functional fixation' in most human behavior in which the person attaches a meaning to a title or object (e.g., manufa(3turing cost) and is unable to see alternative meanings or uses. People intuitively associate a value with an it«3m through past experience, and often do not recognire that the value of the item depends, in fact, upon the particu^lar moment in time and may be significantly diflferent from what it was in the past. Therefore, when a person is placed in a new situation, he views the object or term as used previously" [1966, p. 194]. The authors attempted
to place functional fixation in an accounting context by stating that "If the outputs from different accounting methods are called by the same name, such as profit, cost, etc., people who do not understand accounting well tend to neglect the fact that alternative methods may have been used to prepare the outputs. In such cases, a change in the accounting process clearly influences the decisions" [1966, p. 194].
While authors must be allowed some latitude in extrapolating concepts and findings from one discipline to another, it is important to recognize when such extrapolations may be tenuous. For example, a review of the literature indicates that, contrary to the claim of Ijiri, Jaedicke, and Knight, psychologists have not demonstrated that func-
tional fixation exists in most human behavior. Psychologists have only provided evidence that functional fixation pervades simple, problemsolving tasks involving common, everyday objects. To the best of my knowledge, psychologists have not investigated functional fixation involving data. And, of course, one cannot assume that the behavior of subjects trying to solve the box or two-string problem necessarily parallels that of a manager using data.
In addition to the distinction between objects and data, the Ijiri, Jaedicke, and Knight extrapolation clouds the distinction between/w/iction and output. The psychology literature dealt with fixation on the functions of objects, for example, a box was considered to function as a container rather than as a platform. Ijiri, Jaedicke, and Knight, however, suggested that fixation on outputs of an accounting system (e.g.,
profit or cost) may occur.^ In the case of inventory cost data, to maintain that a decision maker is functionally fixated, one would bave to argue, for example, that a decision maker might be accustomed to using such data for one function (e.g., to assist in making pricing decisions) and that this "preutilization" would inhibit his discoverj' of other functions of that data (e.g., to assist in making production decisions). Clearly, this was not tbe suggestion of Ijiri, Jaedicke, and Knight. Instead, they suggested that a decision maker migbt be fixated on accounting outputs
and migbt ignore the fact tbat the outputs were influenced by a change in accounting methods.
Despite these problems witb tbe Ijiri, Jaedicke, and Knigbt extrapolation, several other accounting researchers bave invoked functional fixation to explain tbeir own research results. While both the original functional fixation work and tbe Ijiri, Jaedicke, and Knigbt extrapolation were time-series oriented (dealing with the bebavior of subjects trying to discover a new function for an object after undergoing preutilization training), some accounting researchers bave given functional fixation a cross-sectional orientation; tbey bave applied functional fixation to alternative accounting metbods rather tban to changes in accounting methods over time.
Jensen [1966] conducted an experimental study of the effects of alternative depreciation and inventory costing methods on decision making by investors. In explaining his finding that alternative accounting methods affected, decisions made in the study, Jensen speculated that some subjects may have been functionally fixated on net earnings. Livingstone's [1967] empirical study of the effects of alternative methods
of interperiod income tax allocation on regulatory rate-of-retum decisions affecting lectric utilities indicated that some rate-making bodies focus on "raw" rates of return; that is, they do not consider the effects of alternative tsot allocation methods. Livingstone suggested that some jurisdictions may be functionally fixated on net operating revenue. Mlynarczyk [1969] studied the effect of alternative tax accounting methods on common stock prices of electric utility companies. He related
functional fixation to his work, asserting that "while the [Ijiri, Jaedicke, and Knight] paper focuses upon management decisions, the applicability to investor decisions is obvious" [1969, p. 70]. Actually, the application is not obvious because, like Jensen and Livingstone, Mlynarczyk applied functional fixation cross-sectionally.
Another area in which functional fixation has been applied in accounting is efficient market research, for example. Beaver's [1972] argument that the market is not functionally fixated. This represents an additional extrapolation of the original functional fixation notion; it was originally suggested for individuals, not for groups or larger entities.
The purpose of these remarks is not to suggest that functional fixation is a useless concept in accounting, nor to suggest that accounting researchers should not modify a general concept of one discipline to explain the particulars of another. However, we should recognize that the functional fixation hypothesis in accounting is a modified form (or forms) of the hypothesis in psychology. The modified functional fixation hypothesis should be subjected to research in accounting contexts, rather than relying entirely on the original functional fixation research as Ijiri, Jaedicke, and Knight and subsequent researchers appear to have done.
The experiment described below attempted to investigate certain aspects of the modified functional fixation hypothesis which appear to be pertinent in accounting.