Activity-Based Budgeting--Part 1.
At the heart of ABB is a reversal of an ABC system. Instead of driving the cost of resources to activities and then to outputs, we now drive the demand for outputs to activities and then to resources. The classical North-South process of ABC is replaced by the South-North process of ABB (Figure 1). In theory, the ABB process is easy. You tell the ABC system the demand for outputs and run the model backwards to estimate the future demand for resources (see Figure 2, p. 86). What could be simpler?
Unfortunately, a simple reversal approach does not work well at all. Typically, the estimates for resource demand that are obtained this way are hopelessly inaccurate. The source of these errors is not a failure of the central activity-based model but fundamental differences between ABC and ABB. There are at least four reasons why a simple reversal process will give highly inaccurate estimates of resource demand: spending versus consumption patterns, secondary outputs, fungible resources, and unavailability of detailed knowledge. This column in the series describes the first two issues and why they are problematic. The next column will address the last two issues, and future columns will discuss how to overcome the challenges posed by all four issues.
Spending vs. Consumption Patterns
ABC systems necessarily model the way resources are consumed, not how they are acquired. That is, the systems relate the amount of resources consumed by activities to the outputs of those activities, not to the amount of resources acquired in the period, It is this property that allows ABC systems to report the same cost for products from period to period (assuming no change in the quantity of resources consumed and the price paid for them) even when the volume of production changes. This objective is achieved by removing the cost of unused capacity from the cost of the resources driven to the activities and then subsequently to outputs in the ABC model. In more formal terms, the ABC is capacity adjusted.
When an ABC system is reversed, because of the underlying resource consumption model the predictions for future resource consumption assume that resource acquisition and consumption are identical. This assumption works well for resources purchased using an "as needed" contract (for example, piecework). Therefore, contract resources consumed (that is, labor) are equal to resources acquired. For such resources, there can be no unused capacity, and therefore a capacity-adjusted ABC is not required.
But resources are not always purchased in the same way that they are consumed; some resources are purchased using "in-case" contracts. For such resources, the reversal of a consumption model does not always predict accurately the amount that is spent on them (even if the amount consumed is accurate). A simple example will illustrate this point. Suppose that supervisors are hired and paid on a monthly basis. Currently five supervisors are kept fully busy. The output for the next period is expected to be 10% higher than the previous period. Reversing a consumption model predicts that 5.5 supervisors are required. However, you can only hire integer supervisors. Thus, the firm has to hire six supervisors to satisfy the demand for supervision. Thus, the reversed ABC systematically underestimates the demand for supervisors. It is only when, fortuitously, the amount of consumed resources equals an acquirable amount that the ABB estimate is "accurate."
The majority of ABC systems report the cost of only three types of outputs: products or services, customers, and distribution channels. These outputs are unique in that revenues can be associated with them. Products can be associated with their selling prices (typically, their average selling prices), customers can be associated with the revenue they generate through their orders, and distribution channels can be associated with the business that is transacted through them. Subtracting the costs (derived from an ABC system) from these revenues gives the profits generated by these profit objects. The ability to report these profits is why most ABC systems are designed to report their costs on an ongoing basis. In contrast, the costs of other outputs of the activities performed at the firm, such as annual reports, typically are reported on an infrequent, special basis.
The decision to report only the costs of profit objects presented a few challenges for the design of ABC systems, but implementers solved them in the very early designs (because these systems only reported these costs on an ongoing basis). In particular, the cost of secondary activities, such as those performed in the human resources, information systems, and finance departments, are assigned to the cost pools of the primary activity (those performed directly upon profit objects) that they support. This assignment process simply assumes that the cost of secondary activities can be averaged across primary activities in a cost pool without significant loss of accuracy. In most settings this is an acceptable assumption.
But for ABB purposes, this design choice poses more fundamental challenges. Assume for the moment that the demand for profit objects for the upcoming period is known (an assumption we'll challenge in the next column)--let's say it's 10%. The reversed ABC simply will assume that the demand for all resources increases by 10%. But the demand for human resources typically will increase by some smaller or larger amount. Therefore, the simple estimate for human resources is inaccurate. The solution is to estimate the demand for the outputs of the human resources department (which is different from the demand for profit objects). For example, to support the 10% expansion, the workforce might need to be expanded by 20% (the same ratio as for the supervisors above). Therefore, the human resources department might need to increase its size by 20% to cope with the increased demand. The assumption that the amount of primary and secondary activities alter in strict proportion is rarely valid, and therefore more sophistic ated modeling is required.
Robin Cooper is professor in the practice of cost management at the Goizueta Business School of Emory University.
Regine Slagmulder is professor of management accounting, Tilburg University (the Netherlands), and visiting professor at the University of Ghent (Belgium).
|Printer friendly Cite/link Email Feedback|
|Author:||Cooper, Robin; Slagmulder, Regine|
|Date:||Sep 1, 2000|
|Previous Article:||The New Computers in School.|
|Next Article:||SEC Bars "Selective" Disclosure.|