What are the advantages and challenges of using activities as cost pools?

Using activities as cost pools

An activity is a sequence of tasks consuming resources to produce a desired outcome and triggered by the same event. Since they have the same underlying cost driver, or activity driver, these tasks are highly correlated and can thus be treated as a single activity.

Activities are homogeneous cost pools because, by construction, they gather the resources consumed by highly correlated tasks and separate the resources consumed by uncorrelated tasks. This is a major argument in favor of Activity-Based Costing.

Activity-Based Costing is an approach to costing that focuses on activities as the fundamental cost pools. Costs of activities are then used as the basis to assign costs to other cost objects, like departments or products.

Tasks composing an activity are not necessarily within a single department. For instance, a new vacancy (the event, activity driver or cost driver) will systematically results in writing a job description (hiring department), advertising for the position (HR), filtering resumes (HR), conducting preliminary interviews (HR), conducting interviews with the future manager (hiring department), preparing an employment contract (legal affairs) and entering a new account in the accounting system (accounting). Now, since these tasks have the same cost driver, they form a single activity with a single event (a vacancy) causing systematically more or less the same pattern of resource consumption. This brings forward another advantage of Activity-Based Costing: it shows the inter-dependencies between departments and functions so that they can be properly managed.

However, the transverse nature of activities is also a major obstacle to the proper implementation of Activity-Based Costing: accountants cannot readily identify for which activity a transaction was made. This would only be possible if each department had only one activity. Even it this were possible, it would be vastly inefficient. Such an organization prevents the dynamic reallocation of resources from one activity to another and complicates coordination between departments by multiplying them. Two different approaches have been suggested to overcome this obstacle through the division of department costs and their recombination into activity costs: the initial approach to Activity-Based Costing (Kaplan & Johnson, 1987) and Time-Driven Activity-Based Costing (Kaplan & Anderson, 2004).

Two approaches of activity base costing

The initial approach of Activity-Based Costing is a two-stage allocation process. In the first stage, department managers are interviewed to list their activities, estimate the proportion of each department’s resources dedicated to each activity and allocate department costs to activities according to these estimated proportions. The second stage consists in identifying each activity driver (i.e. the cost driver serving as allocation base), computing activity rates (i.e. allocation rates) by dividing activity costs by the quantity of activity driver and finally allocating activity costs to cost objects in proportion of cost objects’ consumption of each activity driver.

Figure 1: The two-stage allocation process of the initial approach to Activity-Based Costing

The second approach, Time-Driven Activity-Based Costing, consists in interviewing managers to list activities and decompose them into tasks. Then the resources which should be consumed to perform each task are identified (which resources?), quantified (how much should be consumed?) and valued (how much should it cost?). Based on this information, it is possible to compute an allocation rate for each activity by adding the estimated value of the estimated resource consumption of all the tasks composing this activity. Finally costs are allocated using these allocation rates and the cost objects’ consumption of each activity driver.

For instance, if we note “AR” the activity rate, “U” the usage or consumption of resource, “P” the price of the resource, “L” the labor, “HR” the human resources, “OP” the operations, “FR” the filtering of resumes, “INT” the interviews and “EC” the employment contract, we can build the following equation for the allocation rate of the activity “Hiring” made of these three tasks:

\begin{aligned} AR_{Hiring} = & U_{L.HR.FR} \times P_{L.HR.FR} + \\ & U_{L.HR.INT} \times P_{L.HR.INT} + \\ & U_{L.OP.INT} \times P_{L.OP.INT} + \\ & U_{L.HR.EC} \times P_{L.HR.EC} \end{aligned}

Where you can read $$U_{L.HR.FR}$$ as the consumption of HR labor for filtering resumes and $$P_{L.HR.FR}$$ the price of each unit of HR labor consummed for that purpose. Then each component or this equation can be used to allocate the costs of each contributing department to any other recruiting department (“D1”, “D2”, etc.). This allocation consists in multiplying each component of the activity rate by the number of open vacancies filled (the hiring activity driver, “AD”) for each recruiting department: $$AD_{Hiring.D1} \times (U_{L.HR.FR} \times P_{L.HR.FR})$$ is allocated from HR to Department 1, $$AD_{Hiring.D2} \times (U_{L.OP.INT} \times P_{L.OP.INT})$$ allocated from Operations to Department 2, etc. Fortunately we have computers nowadays…

Issues with both approaches of Activity-Based Costing

First, the very descriptions of these two approaches reveal how complex and therefore expensive their implementation must be even for simple activities. Second, as subtly indicated by my frequent use of the term “estimated” in these descriptions, each approach relies on a different cost estimation technique. The first approach relies on account analysis and the second on the engineering approach. I will define and describe both these estimation techniques in Chapter 3. For now, you just need to remember that since both approaches rely on estimations, they are considered as potentially dangerous by financial accountants (estimations are easy to manipulate) and therefore forbidden for inventory valuation.