Monday, February 28, 2000

Accounting Allocations

Arthur Thomas’ illustrates in his articles a belief that all financial accounting allocations are arbitrary and incorrigible. I will begin with a discussion of what Thomas means by “arbitrary” and will outline his three elements of a theoretically justifiable allocation method. Second, I will address Thomas’ notion of “incorrigibility”, and will address the implications of arbitrary and incorrigible allocations on financial accounting. I will then discuss what Thomas terms the “interaction problem” of process inputs, and his recommendations for providing more relevant financial reports. I will conclude with a discussion of the ability of traditional cost allocation methods to yield information useful for managers in their decision-making efforts.
Thomas defines as arbitrary “any allocation method for which theoretical justification cannot be found” (Eckel, 768). Equipped with this definition, allow me to segue to a discussion of Thomas’ idea of a theoretically justifiable allocation method. Thomas argues that an allocation method is justifiable if it possesses three qualities; unambiguity, defensibility, and additivity (Eckel, 768).
Thomas refers frequently to a range of ambiguity (Thomas, 474). This means that a given input in a given period may be allocated entirely during a specific reporting period, it may be partially allocated during a period, or it may not be allocated at all to a particular period. In such an extreme case, allocation for a given period could range from no cost allocation to the full historical cost of the item being allocated. The result here is referred to as “total ambiguity” (Thomas, 474). Thomas’ point is that for an allocation method to be justified in theory, there should be a restricted range of ambiguity. In fact, he prefers that the chosen allocation method “yield a unique allocation” (Eckel, 768).
The question becomes, why is it so important that an allocation method have a minimal range of ambiguity? What occurs currently in regard to allocations is that one or two groups of financial statement users (usually consisting of a firm’s management or some regulatory body) bargain for what Thomas terms “mutually satisfactory allocations” (Thomas, 477). The resulting allocation is good in that it suits the needs of the bargaining parties. The problem is that the allocation may ignore the needs of non-bargaining users of the financial statements and, thus, fail to suit their purposes. Thomas reminds us that financial reports should be “directed toward the common needs of users,” and that in order to enhance the neutrality of their product, financial statement preparers “should not try to increase the helpfulness of the information to a few users to the detriment of others who may have opposing interests” (Thomas, 476). When there exists a broad range of ambiguity, some users are likely to benefit at the expense of others.
Thomas’ second element of a theoretically justifiable allocation method is that it be “logically defensible against all competing alternatives” (Eckel, 768). Eckel interprets this to mean that so long as there is available another allocation alternative that is defensible to any degree, the method being considered is not “conclusively defensible” (Eckel, 771). The result here would be a choice between two or more allocation methods. In response to this choice between methods I believe Thomas would argue that; 1) the lack of defensibility renders the allocation method theoretically unjustifiable, and 2) the choice between methods supports his claim that allocations are arbitrary. Thomas’ concept of defensibility is important for much the same reason that the ambiguity issue is important – that is, if users are to have faith in the veracity of the financial reports, they must feel confident that their interests are not subordinate to those of other users. The existence of one defensible allocation method helps provide users this assurance.
The third element of a theoretically justifiable allocation method is that the results of the allocation be additive. This concept will be discussed further in conjunction with the discussion of “input interaction.” For now, it should suffice to say that when an amount of an input available for allocation is allocated over a period of time, those individual allocations, if added together, should equal the amount originally available for allocation. If this is not the case, Thomas suggests that the allocation method is not theoretically justifiable.
Now that the elements of a theoretically justifiable allocation method have been defined, allow me to reiterate Thomas’ notion of an arbitrary allocation: “Thomas defines as arbitrary any allocation method for which theoretical justification cannot be found” (Eckel, 768). The implications of arbitrary allocations for financial accounting are basically issues of providing users with financial elements as free from bias as possible. This neutrality in regard to financial reporting is key to maintaining and enhancing the reputation of the accounting profession.
Besides being arbitrary, Thomas argues that allocations are also incorrigible. An incorrigible statement, according to Thomas, is one that “can neither be refuted nor verified” (Thomas, 66). Eckel contends that Thomas’ idea of incorrigibility “amounts strictly to a lack of verifiability by reference to market values.” Eckel continues, “The concept of incorrigibility arises in answer to the following question: What is the relationship of financial accounting’s customary allocations to the ‘real world’?” (Eckel, 765). Thomas warns that if allocations are incorrigible “practicing accountants should be deeply concerned” (Thomas, 66).
It’s true that if allocations are indeed incorrigible that the implications for the profession are many. For example, can statements based on incorrigible allocations effectively serve the needs of a broad array of users? Thomas suggests that the answer to this question is no – “Unless you can suggest ways in which calculations that can neither be verified nor refuted assist decisions, our allocations…are irrelevant to investor needs” (Thomas, 68). I question the severity of Thomas’ conclusion, but do believe that standard setters would do well not to ignore the gist of his argument. It certainly would not hurt to have allocation methods that are verifiable.
A second implication for financial accounting is best viewed from the perspective of the independent auditor. Thomas mentions “accountants attest that financial statements present fairly the positions of companies and the results of their operations” (Thomas, 66). The question is, if allocation methods are incorrigible (or arbitrary for that matter), to what standard does the independent auditor attest? Eckel argues that the auditor’s report is based on generally accepted accounting principles (Eckel, 767). Even so, Thomas point has value. Just because GAAP permits the use of incorrigible allocation standards, this certainly does not always guarantee the protection of the auditor when faced with litigation from disgruntled users of audited financial data. All of this says nothing about the risk of the accountant’s reputation as it pertains to employing a reporting mechanism that can neither be verified nor refuted.
Thomas notes that “inputs to a process interact whenever they generate an output different from the total of what they would yield separately” (Thomas, 66). This synergy results in what Thomas calls the interaction problem. In order to properly match costs with revenues, it is essential to know how an input contributed to a given firm’s output. How can this contribution be measured when inputs interact? Imagine a garden for example. The gardener sustains a variety of costs throughout the year in order to produce the vegetables that he will sell at the farmer’s market. Most, if not all, of the gardener’s costs will benefit not only this accounting period, but future periods as well. In addition, the production of vegetables necessitates cooperation among inputs (i.e., seeds and water). Dismissing the idea that inputs interact in this scenario, how can the gardener’s costs best be allocated given that many of this period’s costs will benefit future periods? Some would argue that since the absence of a particular input would result in no produce, the entire cost of a vital input should be allocated to the current period. Now consider input interaction. How can the costs of interacting inputs best be allocated to the revenue of an accounting period? The point is that there are several ways to allocate the costs of producing revenue. One particular method of allocating costs is no better than another, so whatever method is chosen cannot be refuted as inferior to another or verified as superior to another (Thomas, 67). It appears that Thomas may be right. Many allocations are both arbitrary and incorrigible. The situation gets even more convoluted in the presence of interacting inputs. Thomas has identified what many perceive as a serious shortcoming of financial reporting. He also offers a potential remedy.