Knowledge that Transforms

[Cost allocation is a pervasive practice in accounting. Horngren and Foster (1987, 411) define it as "the assignment and reassignment of a cost or group of costs to one or more cost objectives." An important form of cost allocation is the apportionment of corporate-level costs to various decentralized profit centers. The reason usually given for this procedure is that it is "a major means of getting subordinates to behave as desired by top managers" (1987, 422). Although such allocations have been criticized in the agency literature (Demski 1981) as being irrelevant for motivating managers in the presence of compensation contracts, this article identifies a precise role for indirect cost allocations, in conjunction with other instruments such as participative budgets, in settings with multiple divisions. A survey by Fremgen and Liao (1981) showed that 84 percent of firms allocated at least part of their indirect costs to their profit centers, and 80 percent did so for the purpose of evaluating the performance of profit center managers. The major aim was to remind managers that indirect costs exist and that at least a portion of these costs had to be covered by profit center earnings. Further, Atkinson's (1987) poll reported that the primary objectives of allocating indirect costs were the motivation of employees and the provision of signals for resource allocation. There has been little analytical work, however, on the economic role played by cost allocation systems. Demski (1981) claims that allocation mechanisms are useful only if they provide additional information that can be contracted upon. Similarly, Baiman and Noel (1985) show that in certain multiperiod settings, it is optimal to compensate an agent on the basis of prior periods' realizations of costs that were not in the agent's control, provided these costs affect the principal's capacity decisions. Magee (1988) identifies conditions under which an agent is compensated according to the usage of some resource, in addition to output. The agency studies described above cannot address cost allocations across operating units since they do not model multiple productive divisions among which common costs are to be allocated. Further, because they model single-agent settings, the second-best incentive schemes they derive suffice to motivate managers efficiently; thus, there is no role for allocations unless they provide additional information to the owner. With multiple divisions, however, simple compensation contracts do not provide adequate incentives to guarantee the owner's desired outcome. The role for allocations, over and above that of second-best contracts, in providing assurance to the owner is demonstrated here with a one-period model of an entrepreneur who incurs fixed costs in the production of different products by two workers. When the workers have correlated private information about the productivity of a common, central resource, the entrepreneur may fail to recover the fixed costs because the optimal payment schedules encourage the workers to misreport the productivity of the resource. By asking each worker to submit a budget to determine overhead rates, and by evaluating managers on their divisional profit after allocation of overhead costs, the entrepreneur can obtain the second-best return on the fixed investment. This mechanism is then compared to allocation systems observed empirically and to those recommended in the accounting literature. The mechanism design approach pursued in this paper, with the game itself a variable, enables the study of procedures by which accounting numbers are derived for purposes of performance evaluation. The accounting system is not imposed as a monitor or source of information; its value arises from its procedures, which enable the owner to play off the managers' private information against each other to guarantee a second-best return on investment. In the absence of such a system, the division managers would gain from implicitly colluding in their use of the central resource. The allocation scheme coordinates their actions and induces them to act in the manner desired by the owner. This positive role for the allocation of indirect costs in conjunction with the budgetary process emphasizes the value of a cost allocation system as a set of linked motivational devices.]

[Analytical review, the diagnostic process of identifying and determining the cause of unexpected fluctuations in account balances and financial ratios, has become an increasingly important part of auditing. Recent research suggests that analytical review is beneficial for detecting such unexpected fluctuations (Kreutzfeldt and Wallace 1986; Wright and Ashton 1989). It is, however, difficult for an auditor utilizing analytical review to know when the "correct" cause for an observed fluctuation has been determined, especially since no official guidance exists to assist the auditor. Thus, accepting a plausible but incorrect cause is a dangerous and largely unresearched possibility (Kinney 1987). This article describes two experiments that investigate conditions under which auditors may be prone to accepting a plausible, but potentially incorrect, cause when performing attention-directing analytical review. Identifying such conditions is important because either audit effectiveness or efficiency would be compromised with such incorrect acceptance. Although there are no normatively correct causes for the analytical review problem situations employed in this study, relative evidence on such compromises will be obtained by investigating conditions that lead auditors to make different judgments about the cause of an unexpected fluctuation. Differences in auditors' judgments are important since they can lead to differences in decisions as to the cause of an unexpected fluctuation and subsequent auditor actions (cf. Asare 1991). Dichotomizing causes into error and non-error classes reveals several ways in which an auditor performing analytical review may accept a plausible but incorrect cause. One that has particularly significant implications for the effectiveness of the audit would be concluding that a non-error cause adequately accounts for an unexpected fluctuation when the "correct" cause involves a financial-statement error. Since it is unlikely that the auditor would increase planned tests of details if indications of a material error were not present (and it is possible that the auditor may decrease such testing), the likelihood of subsequent detection of the error would be reduced in this situation. Conditions that may lead to this type of decision error will be investigated in this study. The effects of explanation, counterexplanation (i.e., consideration of why an hypothesized cause may not be correct), and their order of performance were studied to determine how such factors affect the propensity for acceptance of a plausible but incorrect non-error cause. Investigating explanation is important since auditors typically document explanations for analytical review findings. Even though current auditing standards do not require auditors to counterexplain, investigating how counterexplanation and the order of counterexplanation and explanation affect auditors' judgments is nevertheless important since it enhances our understanding of the potential for acceptance of an incorrect cause. Auditors from five public accounting firms participated in two experiments conducted to determine the effects of the aforementioned factors on auditors' judgments. Consistent with expectations, auditors revised their beliefs about an hypothesized non-error cause of an unexpected fluctuation in the auditee's gross margin in an upward (downward) fashion when asked to document an explanation (counterexplanation) for that cause. Inconsistent with expectations, however, auditors who were asked to document an explanation initially and then a counterexplanation judged the non-error cause to be less likely to have occurred than did auditors asked to document a counterexplanation initially and then an explanation. Overall, these results identify conditions in which auditors may be prone to compromising audit effectiveness by accepting a potentially incorrect non-error cause.]

[This article examines the incentives for defense contractors that operate under cost reimbursement to manipulate pension costs. We investigate pension costs because of their high discretionary component and their importance to managers seeking to manipulate costs. For example, two of the 20 standards issued by the Cost Accounting Standards Board (CASB) focus exclusively on the determination of allowable pension costs. Also, the availability of detailed data on pension funding and actuarial variables across plans sponsored by the same firm and across plans of different firms allows for extensive tests of the hypothesized effects of cost reimbursement. Defense contractors are selected because they are closely associated with cost reimbursement (see, e.g., Chwastiak 1990; Pownall 1986), and the terms "defense" and "non-defense" are used here to denote the presence and absence, respectively, of reimbursed costs. The analysis suggests that overfunding is valuable to defense contractors only if they also have or plan to have non-defense business. Two generic funding strategies are described: across-time and across-contract. In both cases, plans are overfunded when employees work on defense contracts, then these excess pension assets are withdrawn when the employees work on non-defense business. The across-contract strategy is implemented by transferring employees and associated pension assets and liabilities between defense and non-defense plans. The across-time strategy is implemented by having individual plans alternate between extended periods wherein they first cover defense contracts, then cover non-defense business. The firm can achieve this effect either by selectively "cycling" defense contracts across plans or by changing its proportion of defense contracts over time. Evidence of inter- and intra-firm variation in funding levels and actuarial variables is presented for a sample of 80 major defense contractors. Across firms, funding levels increase with the proportion of defense revenues, reach a peak when defense revenues equal non-defense revenues, and then decrease as the proportion of defense revenues increases (an inverted-U-shaped relation). This last downward sloping portion of the relation is consistent with the across-contract strategy as it indicates reduced ability to make transfers as the pool of non-defense employees declines. Although evidence of such transfers between defense and non-defense plans could not be obtained, indirect support for this view is provided at the intra-firm level. Relative to a control sample, defense contractors sponsor more plans (which is consistent with their having separate plans for defense and non-defense activities) and exhibit a wider range of funding levels and actuarial variables across plans sponsored by each firm (which is consistent with differential funding of the two groups of plans). Two important features regarding the scope of this study are worth noting. First, cost reimbursement could also create incentives to inflate pension costs by increasing the level of promised pension benefits (more generous benefits), or by using various allocation strategies, whereby more common pension costs are allocated to defense contracts. Although we are unable to investigate these incentives due to data limitations, our inter- and intra-firm results are not confounded by these effects. Second, our results do not prove systematic "overcharging," as competition among suppliers and the ability of customers to identify and adjust for unusually high pension costs are not examined.]

[Demski and Sappington (1990) treat income disclosure as a process that conveys fully revealing information, and they establish two results related to this commentary: (1) an income-disclosure procedure is fully revealing if, and only if, the related end-of-period, present-value measures always are invertible in the realized information-structure signals, and (2) strict favorableness is sufficient for fully revealing income disclosure under the true probability measure. This commentary shows that the first result follows from a more interesting necessary and sufficient condition than invertibility. Essentially it shows that an income-disclosure procedure is fully revealing if, and only if, differences in perceptions about the discounted expected values of future cash flows or differences in current cash flows are possible in all periods for all cash flow/signal pairs. In the context of the second result, an alternative, more general, sufficient condition for fully revealing income disclosure under the true probability measure is identified. This outcome is assured if differences in perceptions about the discounted expected values of future-periods' cash flows are possible in all periods for all cash flow/signal pairs.]

[This study examines the stock market reaction to the adoption of long-term compensation agreements for top management that are based on accounting goals. Prior researchers (Baril 1988; Larcker 1983; Smith and Watts 1982) argue that these agreements, known as performance plans, motivate executives to improve firm performance by working harder, lengthening their decision horizons, and becoming less risk-averse in their investment decisions. However, because the agreements are based on accounting goals, performance plans can distract managerial attention from the maximization of share price and can also encourage the manipulation of the accounting system (Gibbons and Murphy 1989). Managers may also reject projects with large, positive net present values in favor of less valuable projects with higher accounting profits. In an earlier study, Larcker (1983) reports a significantly positive stock market reaction for a sample of 21 firms that adopted performance plans between 1971 and 1978. The reaction occurs on the trading day after the SEC receives the proxy statement announcing the plan adoption. However, we find no significant reaction during the two-day announcement period beginning with the SEC stamp date for a sample of 209 adoptions that occur between 1971 and 1980. Further, no evidence of abnormal stock performance is detected for the two-day periods beginning with either the date on which the board of directors voted to approve the plan or the proxy statement date. Our study highlights the difficulties inherent in analyzing the stock market reaction to announcements made in proxy statements. First, pinpointing the timing of information dissemination is problematic. Second, firm-specific information unrelated to the event of interest is often released in the proxy statement and at the annual meeting. This complicates the interpretation of any unusual stock price behavior. In our study, significantly positive abnormal performance is observed for both adopting and nonadopting firms during the period beginning two days after the SEC stamp date and ending the day after the shareholders' meeting date. This suggests that the reaction observed for the adopting firms is due to the impending shareholders' meeting rather than to the performance plan adoption per se. We suggest that the results of any event study involving announcements made around the time of the annual shareholders' meeting should be interpreted with caution.]