By Dr. Peter Turney
CEO, Cost Technology
Journal of Cost Management

This article examines ten myths of product costing identified from the comments of managers and academics. Each myth was heard several times from different sources. Each myth also appeared to have a kernel of truth, but had been generalized to apply to all activity-based systems. This article describes each of these ten myths and identifies the conditions under which they are true and those under which they are not.

Activity-based cost (ABC) systems maintain and process data on activities and products. These systems trace costs to products according to the activities performed to produce them. ABC systems are an important source of information on overhead activities and remove much of the distortion in product costs that is inherent to conventional unit-based systems.

Mythical beasts
This article examines several myths about ABC systems. The word “myth” as it is used here means a commonly stated view that is accepted without critical analysis. In fact, myths of this type are sometimes true – but only under certain limited conditions. It is this limited truth that gives them their persuasive force.

It is therefore important to understand when myths are true and when they are not. When viewed as rules of general applicability, the myths discussed in this article could lead to incorrect designs of cost systems. An incorrect cost system design could report inaccurate product costs and could provide a distorted view of the performance of activities. Either deficiency could cause management to make wrong decisions.

Common themes
The myths examined in this article contain certain common themes about ABC systems. One theme is that these systems are too difficult to implement and use. The second theme is that improving conventional systems makes ABC systems redundant. The third theme is that more accurate product cost information is unnecessary. The forth theme is that ABC systems play a limited role in managing a firm. The myths for each theme are as follows:

• ABC systems are too difficult to implements and use:
  • ABC systems are too costly; and
  • ABC systems are too complex to understand.
• Improving our existing system will do the job:
  • All that we need are more cost centers;
  • Machine-hour systems save the time; and
  • A cost system should be kept simple.
• We do not need more accurate product costs:
  • We know what our products cost; and
  • The market sets prices, so we do not need product costs.
• Cost systems play a limited role:
  • We cannot do anything about fixed costs;
  • Only manufacturing costs are product costs; and
  • Product costs are not useful for managing overhead activities.

The following sections provide a more detailed examination of each theme and its related myths.

ABC systems are too difficult to implement and use. Managers often express a concern that ABC systems are too difficult to implement and use. There is certainly a cost associated with designing and implementing a new cost system that has a different architecture and different data requirements than conventional cost accounting systems. There is also a cost to users of ABC systems who must adapt to the new cost information. Managers are therefore justified in questioning these costs, because they affect the net value of a shift to an ABC system.

Myth 1: ABC systems are too costly. A common response to ABC systems is to say that they cost too much. According to this myth, the additional cost drivers used in ABC systems increases the cost of system design, implementation, and maintenance to an unacceptable level.

In fact, ABC systems are more costly than unit-based systems. By design, an ABC system identifies activities and cost drivers, traces costs to each activity, and then traces the cost of the activities to products by means of the cost drivers. Cost driver usage is estimated for each component or products by means of the cost drivers. Cost driver usage is estimated for each component or product that consumes that driver. The driver “number of receipts”, for example, requires that the number of receipts be estimated for each component. Each time the ABC system is revised, this estimate must be updated.

Experience shows, however, that most people overestimate the incremental costs of tracking and data gathering for ABC systems. Mush of the data required by an ABC system already exist in the plant or can easily be captured. In a high-technology environment, for example, the number of insertions in an integrated circuit insertion machine or the number of holes punched in a printed circuit board are likely cost drivers for an ABC system that would already be stored in the manufacturing data base.

Even if data that a cost driver uses are unavailable, the data can be estimated from available data, or replaced with an equivalent measure. For example, the number of machine hours does not have to be measured directly; it can be estimated from the number of direct labor hours combined with the average number of machines worked on by an operator, Similarly, if the number of setups is not recorded, the number of production runs can be substituted, because typically there is a one-to-one relationship between the number of production runs and the number of setups.

ABC systems often reduce the cost of special studies required to supplement the information reported by the product costing systems. There are many firms with conventional systems whose accounting departments spend over 50 percent of their time conducting special studies. After implementing an ABC system, this percentage drops substantially; fewer studies are needed and also less effort per study. This reduction in frequency and effort reflects the additional information and increased accuracy of information in the ABC system.

Several firms have implemented ABC systems at a cost that was considered acceptable to management. One form, for example, recently implemented a system with 44 cost drivers, 30-activity center, and over 2,000 cost pools. Although this system was fairly complex, it was designed and implemented in about nine man-months. The system runs on a PC and computes product costs in minutes.

Myth 2: Activity-based systems are too complex to understand. Managers often state flatly that ABC systems are too complex to understand. They believe that the additional number and type of cost drivers make an ABC system difficult to understand. As a manager of one form commented:

“The objective with the cost system was to change the behavior of the management of the division. We did not go to a more complex system immediately – say, an eight-driver system – because we wanted to take incremental steps and change behavior permanently. If we had tried to introduce a massive change we might have created such confusion that the benefit would have been lost.”

When pressed for examples of complex systems that they have encountered, managers describe accounting systems that generate large numbers of variances and transactions, but not the product costing system itself. For example, in the firm whose manager was just quoted, the accounting system generated twenty-one different cost variances each month and required as many as fifty performance measurements by each operator each day (totaling over 30,000 measurements per month on one product line alone). The system processed more than 25,000 inventory transactions per month and divided the plant into more than 100 cost centers. It is therefore not surprising that management had difficulty understanding such a complex system.

The firm’s product costing system, however, was not the source of confusion. Overhead was directly traced at the first stage to the production cost centers or allocated using direct labor hours. The second stage used direct labor hours as the only cost driver. This simple unit-based design was actually quite easy to understand.

Although ABC systems use more cost drivers than unit-based systems, ABC systems are easy to understand because these drivers describe the major factors that create the demand for activities. ABC systems therefore match mangers understanding of the economics of their plants.

The driver “number of engineering changes,” for example, conveys the message that the more engineering changes there are on a product, the more costly the product will be.

Similarly, the driver “number of purchase orders” implies that the cost of purchasing some item is a function of the number of times it is bought. In both cases, it is clear to management that the cost per driver unit is determined by how efficiently engineering change and purchasing activities are carried out.

Therefore, the perception that ABC systems are too difficult to understand and use does not reflect the experience of firms that have implemented such systems. Although ABC systems are undoubtedly more costly than conventional systems because they require more measurements, this additional cost is often modest. In any case, even if the additional measurements increase the complexity of ABC systems, the resulting systems prove to be easier to understand than conventional cost accounting systems because they reflect economic reality better.

Improving our existing system will do the job. Critics of ABC systems frequently point out that a better alternative to starting with a completely different cost accounting system would be to improve the existing system. These reminders are appropriate, because generally ABC systems are compared with poorly designed conventional systems. Many well-designed conventional systems exist: These systems use multiple cost centers, often incorporate machine hours as a cost driver, and tend to be simpler than ABC systems. The question, therefore, is whether improving a conventional cost system reduces the need for an ABC system. The next sections address this issue.

Myth 3: All that we need are more cost centers. Another often-stated myth holds that dividing a plant into multiple cost centers will eliminate most of the distortions in product costs.

It is true that increasing the number of cost centers leads to more accurate product costs. However, there is a limit to the accuracy that can be achieved by simply increasing the number of cost centers. Creating multiple cost centers where only one or two existed does reduce distortion when costs are consumed differently in different parts of the plant. A plating department, for example, may have expensive equipment, whereas a painting department may have inexpensive equipment. Combining both departments into one cost center will result in inaccurate product costs if some products are routed through one department and some through the other.

The problem, however, is that activities that are not on the unit level cannot be handled by unit-based systems if multiple products pass through the cost centers. Instead, different types of cost drivers, such as those used in ABC systems, must be used.

Each batch of product passing through the plating department, for example, may require first-piece inspection. A unit-based driver, such as direct labor hours, attaches inspection cost to each product unit based on the amount of direct labor time required. This inspection cost results in inaccurate product cost when the batch size varies depending on the product inspected. A cost driver that would more accurately trace inspection costs to products, therefore, would be a batch-level driver, such as the number of inspections.

Myth 4: Machine-hour systems would save the day. Machine-hour systems are often proposed as solutions to the problems of unit-based systems. The belief is that replacing the cost driver direct labor with machine hours will result in more accurate product costs in today’s manufacturing environment.

Advantages of using machine hours. Using machine hours has two major advantages over direct labor; First, machine hours may provide more reasonable overhead rates. In many firms, the proportion of manufacturing cost attributable to direct labor has fallen below 20 percent; in some cases, it is as low as one percent. In these firms, the overhead rate per labor dollar may be 200 percent or more. Managers have trouble understanding the meaning of such high rates and question whether diminished use of direct labor can really account for such huge overhead rates. Since most companies now use more machine hours than labor hours, the use of machine hours as a cost driver generally makes overhead rates fall to a more reasonable level.

The second advantage of using machine hours instead of direct labor hours is that machine hours trace machine-related activities to products more accurately than direct labor hours when machine-base diversity exists. Machine-base diversity exists when the number of machines that an operator runs varies according to the product produced. A relatively complex product, for example, may need four machines to be produced, while a relatively simple product may be produced on only two machines. In a case such as this, the ratio of machine hours to labor hours differs from one product to another, and direct labor hours will not capture this variation.

Machine hours will not produce accurate product costs if much of a company’s costs is incurred by activities that are not at the unit level, because both machine hours and labor hours are unit-based cost drivers. Machine hours capture work performed on a unit of a product and therefore do not do a good job of tracing the cost of activities that are not at the unit level to products.

Example: Products A and B, for example, may require the same number of machine hours. Even if A has quality problems and requires twice as many engineering hours as B, an equal amount of engineering hours as B, an equal amount of engineering cost will be traced to each product if machine hours are the cost driver used. The reported costs of the products will therefore be inaccurate, because A actually consumes twice the engineering costs that B consumes.

Ten Myths that Create Barriers to the Implementation of Activity-Based Cost Systems
By Peter B. B. Turney, PhD
Journal of Cost Management

Myth 5: A cost system should be kept simple. Managers and academics often state that a cost system should be kept simple-that is, relatively few cost drivers should be used.

If the products in a plant are homogeneous, even a unit-based system with only one or two cost drivers produces accurate product costs. (Note that, in the extreme case, a plant that produces only one product does not need a product costing system in the first place. To assign costs to products would simply require dividing all costs of the plant by the output during the period.) In plants where product homogeneity exists, therefore, there is no need to install an ABC system with multiple cost drivers to improve the accuracy of product costs.

Product diversity. But using only one or two cost drivers may result in inaccurate product costs when product diversity exists in multiple product facilities. Product diversity exists, for example, when products are manufactured in batches of different sizes or when product differences affect how intensely activities are performed. When product diversity exists, more – and different – types of cost drivers are needed to capture the impact of the product diversity on the cost of the products.

The cost of machine setup, for example, is a batch-level activity. If setup cost is traced to products via a unit-based cost driver and if batch sizes vary, the result will be inaccurate product costs. In contract, a batch-level driver, such as the number of setups, traces setup cost to products based on their use of the activity in question. Changing the engineering specifications of a product is a product-level activity. Product costs will be inaccurate if the cost of engineering changes is traced using a unit-level driver. This results because engineering effort usually varies from product to product. To accurately assign this product-level cost, a cost driver such as the number of engineering – change notices would have to be used.

The belief that improving conventional cost accounting systems will increase the accuracy of product cost is correct, but the increased accuracy may still be insufficient. Increasing the number of cost centers separates unlike activities. Using machine hours helps reduce burden rates and copes with machine-based diversity. These changes, however, do not improve the ability of conventional systems to handle activities that do not occur at the unit level, nor do they account for differences caused by product diversity. Therefore, companies may need an ABC system to provide accurate product costs if the following situations apply:

• They have product diversity; or
• They incur significant costs in activities that are not at the product level.

We do not need more accurate product cost. Some people retort that conventional systems report product costs that are sufficiently accurate. Some add that additional accuracy is unimportant, because managers know intuitively what products cost. Others believe that accuracy is unnecessary on the ground that the cost to produce a product is not used to set prices. These beliefs are important because they challenge one of the major benefits of ABC systems-more accurate product costs. The following sections address this issue.

Myth 6: We know what our products cost. Many people believe those managers easily estimate product costs without the help of a cost accounting system. If this belief is true, why incur the cost of developing and maintaining a cost system when managers can estimate product costs to an acceptable degree of accuracy using their intuition? In other words, such managers only need a cost accounting system to value inventory to the satisfaction of the auditors and the taxation authorities.

Managers may be able to identity products that are overcosted or undercosted. These managers use their knowledge of the products and processes to identify inaccurately costed products, These managers know, for example, that a product that is difficult to manufacture must cost more than one that is easy to manufacture. They will understand this fact even if their unit-based cost accounting system reports that the two products the same.

Managers do not realize the size of the problem. Experience has shown, however, that managers are unable to estimate the size of product cost distortions. The controller (who chose to remain anonymous) of one plant that had designed an ABC system commented on this tendency:

“The cost of some products turned out to be close to what we had predicted, but the cost of others was quite different. The cost of prototypes, for example, turned out to be far less than we had anticipated. Prototypes required extensive discussions with customers, engineering time, and different handling. They were regarded as nuisance items. But when we quantified this nuisance factor, it turned out to be more emotion than reality. At the extreme, the cost of setups was far higher than we anticipated. We had failed to include the cost of inefficiencies in the process flow in the cost of the setups. These costs were not visible to us and were not clearly attributable to the setups.”

Myth 7: The market sets prices so we do not need product costs. Many managers believe they do not need product costs because the market sets prices. Since one of the main reasons for having product cost information is to set prices, managers in firms whose prices are set by the market often feel that product costs are unnecessary.

Pricing to market is common in several industries, which means that the role of product costs in setting prices is, therefore, limited. Pricing to market is common, for example, in high-technology industries, because product life cycles are short and margins are high relative to manufacturing cost. Pricing to market is also common for commodities and when market prices are readily available. Finally, pricing to market is also used when product costs are difficult to determine.

However, it is dangerous to believe that product costs are unnecessary even if they are not needed for pricing; Managers chase profits, and profits can be phantoms of the cost system. It is only natural that managers should devote attention and resources to the products they perceive to be the most profitable, and it is the cost system that reports the absolute and relative profitability of products.

The business of one company, for example, was to combine two products – attaching machines and fasteners – and market them to customers as a package. The firm rented the machines out at a deliberately low price and priced the fasteners to cover their cost and the remaining cost of the machines. The cost system, however, traced all costs to the fasteners and none to the machines. On the surface, the system worked fine. Customers were happy and loyal. But when the company redesigned the cost system so that it separated the costs of the two types of products, it became clear that the cost system was sending highly distorted signals. Since some of the fasteners were labor intensive, the old system had attributed a disproportionate amount of overhead (including the cost of the attaching machines) to them. Over the years, the company had put little effort into these product lines and consequently walked away from attractive markets.

Experience has shown, therefore, that improved accuracy is an important benefit of ABC systems. Managers may be able to identify products that are incorrectly costed, but they may not be able to judge the size of the distortion. Thus, even firms that do not use product costs to set prices should still devote attention and resources to products based on relative profitability.

Cost systems play a limited role. Another school of thought maintains that ABC plays a limited role in managing a firm. One related belief is that improved information about overhead cost is unnecessary because little can be done about this “fixed” cost. A second belief excludes non-manufacturing activities from the focus of ABC systems. A third belief grants that ABC systems can help in making improved strategic decisions but does not acknowledge their usefulness in helping to manage overhead activities. If these beliefs are true, they clearly diminish the value of ABC systems. Each of these issues is discussed next.

Myth 8: We cannot do anything about fixed costs. Many people believe that little can be done to reduce fixed costs. (Fixed costs are costs that remain constant during a limited period despite fluctuations in volume. This assumption of constancy is relaxed only if the time period is long enough and if the change in volume is sufficiently large.)

The source of this myth is the belief that most costs can be divided into the categories of variable and fixed. Managers and academics a like observes that the cost of certain activities, such as the cost of connecting a power supply to a chassis, varies with the number of product units produced. Such activities are unit-level activities because they are performed each time a unit of the product is produced. Both other activities, such as engineering change activities, are not performed each time that a unit is produced and are, therefore, fixed.

While the cost of these “fixed” activities does not vary with the number of batches and different products produced. Batch-level activities are performed each time a batch of the product is produced. Product-level activities support the production of a product type. The only activities that are not variable are those at the process level that relate to the entire facility. These are costs that support all the products produced within the facility.

Batch-and-product-level activities are common in many manufacturing settings. Example of batch-level activities include the following:

• Setting up a machine
• Scheduling the production of batches of products; and
• Issuing purchase orders to buy a batch of components or materials.

Product-level activities include the following:

• Making an engineering change to a product; and
• Establishing vendor relations for a new component.

A failure to recognize batch-and product-level activities diminishes the perceived benefit of ABC systems. Conventional systems are unit based. They recognize only one level of variability-the unit level-and use only unit-level cost drivers (traditionally, direct labor hours). ABC systems, however, recognize multiple levels of variability and use unit-level, batch-level, and product-level drivers to capture this variability.

Myth 9: Only manufacturing costs are product costs. Another widely held belief is that only manufacturing costs is product costs. Conventional product costing systems are consistent with this belief because they exclude the cost of activities performed outside the factory walls. These excluded activities include research and development, selling, and distribution.

The exclusion of non-manufacturing activities from product cost reflects the influence of generally accepted accounting principles (GAAP) on product costing systems. GAAP defines product cost to mean full manufacturing cost; non-manufacturing costs are classified as period costs rather than product costs. This bias reflects the importance of inventory valuation as an objective for unit-based systems. For some, this bias also reduces the perceived value of ABC systems is their ability to handle non-manufacturing activities.

While excluding non-manufacturing activities may make sense for inventory valuation, it is counter-productive for managerial purposes. It is an inevitable fact that products require many activities outside manufacturing including the following:

• Designing products;
• Processing customer orders; and
• Distributing products to customer.

The cost of these activities can be traced to products using drivers such as the number of customer orders. Failing to include the cost of these activities distorts product costs if the following situations occur:

• The cost of the activities is high; and
• The activities are consumed differently by different products.

Note that in many firms the cost of non-manufacturing activities exceeds 20 percent of total revenues.

Tracing non-manufacturing costs. Several companies with ABC systems trace the cost of non-manufacturing costs to products. A manufacturer of printed circuit boards, for example, traces the cost of order entry, sales, and marketing activities to products using the number of customer orders as a cost driver. This tracing results in quite different product costs, because the size of customer orders varies significantly from product to product.

Under the firm’s old system, the cost of order entry, sales, and marketing activities was expensed and not traced to the products. All products were therefore under-costed; products ordered in small batches were significantly under-costed. This under-costing created an incentive to accept small customer orders. Management has since instituted a charge per customer order (using the ABC system as a guide). As a result, the average size of customer orders has increased.

Myth 10: Product costs are not useful for managing overhead activities. Some people maintain that product costs are not useful for managing overhead activities. For those who subscribe to this opinion, a decision to implement an ABC system must be based on strategic benefits from the system instead of cost reduction.

This belief stems from the experience of using unit-based product costs for cost reduction purposes. Unit-based systems use drivers such as direct labor hours that do not accurately measure the use of activities that are not at the unit level. Therefore, managers who use this inaccurate information to reduce cost may make changes (to products or processes) that increase rather than reduce cost. In one firm, for example, the redesign of the products to reduce direct labor cost actually increased total cost because of the increased demand for overhead activities.

ABC systems, however, provide useful information for cost-reduction programs. ABC systems use more and different types of cost drivers that enable them to determine product costs accurately. As a result, it is possible to manage these cost drivers and reduce costs. ABC systems also provide information on activities that facilitates managing those activities.

One firm, for example, used the number of part numbers as a cost driver in its ABC system. An important goal in this firm was a reduction in the numbers. Management believed that reducing the number of part numbers in new product designs would reduce the demand for material-related activities and thus reduce overhead. Management also felt that a reduced part count and the simpler product designs that resulted from using fewer parts would facilitate the introduction of just-in-time (JIT) manufacturing.

Another firm used its ABC system to determine the cost of engineering-change activities. Careful study showed that the engineering-change process could be redesigned to reduce the cost and time required completing an engineering change. The ABC system was used to model the change and to confirm the cost savings that resulted.

Overall, there is no justification to the belief that ABC systems play a limited role in reducing costs. ABC systems provide new opportunities for cost analysis by recognizing what proportion of supposedly “fixed” overhead is actually variable. Including activities outside the factory provides additional information on the cost of products that require these activities. Finally, ABC systems provide rich sources of information for managing overhead activities. In some cases, firms have implemented ABC systems solely for this reason.

Conclusions
The rapid emergence of ABC systems challenges forms to assess the value of ABC by carefully balancing the costs and benefits re-designing and implementing a new cost system. Many firms are likely to find that ABC systems provide a distinct advantage and should be implemented.

This article has explored a number of myths about ABC systems. These myths, each of which casts doubt about the value of ABC systems, actually represent only partial truths about ABC. The myths may be valid under limited circumstances, but they break down when they are extended to ABC systems at large. Companies that are considering the redesign of their product costing systems may, therefore, make the wrong decision if they accept these myths as always being true in all circumstances.

In sum, experience shows that ABC systems are not difficult to implement and use. There are also limits to how much improvement can be derived from conventional systems. In addition, companies usually benefit from an ABC system that reports more accurate product costs. Finally, the value of ABC systems is derived from a number of uses, including the management of overhead activities.

About Cost Technology
Cost Technology designs and implements solutions to support fact-based decision making.  These solutions transform existing data into strategically meaningful and actionable insights in areas such as cost and profit management, financial forecasting, analytics, and performance management.  Since 1991, the firm has helped public and private sector organizations across the world improve performance by creating the knowledge needed to make every decision count.