What are the principles guiding the definition of cost pools and the selection of allocation bases?


Cost pool homogeneity

It is theoretically possible to gather several costs in the same cost pool without any loss in accuracy under one condition: these resources must be consumed in the same proportions by each cost object and therefore the related costs must be highly correlated with each other, whatever the underlying product mix.

In the bakery example, let’s assume that oven’s costs are made of energy costs and costs of supplies for maintenance. Both costs are likely to be highly correlated because they have the same underlying cause or cost driver: the time during which the oven was used, whether is was used for bread or pastries. The longer you use the oven, the greater both energy and maintenance costs. Oven time is the cost driver of these oven costs.

A cost driver is a measure of activity which triggers a consumption of resources. A cost driver has a cause-and-effect relationship with costs: a change in cost driver causes a change in costs.

Since energy and maintenance costs have the same underlying cost drivers (oven time) and are thus likely to be highly correlated, they can be pooled together and allocated in the same proportions between cost objects. This is the basis of the first fundamental principle of cost allocation, the principle of cost pool homogeneity:

The principle of cost pool homogeneity states that all costs gathered in the same cost pool should have the same underlying cause or cost driver. By extension, all costs which have the same underlying cause or cost driver can (and should) be gathered in the same cost pool.

A homogeneous cost pool is thus a cost pool in which all the costs have the same or a similar cause-and-effect relationship with a specific cost driver. These costs should therefore all be highly correlated with each other and with this cost driver.


Allocation based on causality

If a cost pool is homogeneous, all its costs can be allocated together without distortion using the same allocation base, providing the underlying cost driver is used as allocation base. This is the second fundamental principle you must respect for an accurate cost allocation: the principle of allocation based on causality:

The principle of allocation based on causality states that the cost driver of the costs gathered in a cost pool should be used as allocation base for this cost pool.

Gathering indirect manufacturing costs in homogeneous cost pools results in less expensive costing without loss in accuracy. Another great advantage of having cost pools clearly associated with their underlying cost drivers is that it tells managers if they can keep their costs under control. They can try to reduce either their consumption of cost drivers (e.g. use less oven hours) or the amount of resources consumed every time a cost driver is used (which is approximated by the allocation rate).


Theoretical causation, empirical correlation, and practical efficiency

To implement these principles, ask first whether some indirect cost items are likely to have the same underlying cost driver (theoretical causation) and make a first pre-classification based on this criterion. Then, if you have the data, check whether these costs are correlated with each other and correlate with the cost driver (empirical correlation). If there is both causation and correlation, it is safe to put them together in the same cost pool, using the cost driver as allocation base.

Solution. If we reorganize the table of correlation and format cells based on the magnitude of the correlation, we obtain the following results:

It is clear here that we have two different costs patterns or behaviors. A first group gathers energy costs and lubricant costs which are highly correlated with both machine hours and labor hours. From the causation perspective, energy and lubricant are likely to be used by machines rather than humans (well, I base this assumption on the context…). From a correlation perspective, energy and lubricant costs both have a stronger correlation with machine hours than labor hours. Therefore a first cost pool could gather energy and lubricant costs and have machine hours as allocation base.

The second group gathers maintenance and set-up costs around the number of batches processed which can therefore be used to allocate them.

This pattern tells a lot about the production process and how resources are consumed. Since the number of units and the number of batches are uncorrelated, the number of units per batch must vary a lot (some batches have few units, other batches have many). Moreover, the longer the machines run, the more labor (probably machine operators), energy and lubricant they consume. However, set-up probably happens before every batch and maintenance after every batch and require as much time irrespective of the batch size. To reduce costs, managers should increase the number of units per batch!


This is for the theory. In practice, these principles can be difficult to enact because companies are mostly organized out of a concern for operating efficiency and effectiveness, not for costing accuracy and efficiency. This is how it should be: management accounting is a tool to help operations, not the other way around. Moreover, many costs are not variable and thus have no cost driver and cannot be correlated. Constraints imposed by the way companies are organized and by cost behavior thus force less accurate but more practical choices: the use of departments or activities as cost pools and the reliance on surrogates of cost drivers as allocation bases.


Please indicate how clear and understandable this page was for you: