Activity base costing for manufacturers.

Activity-Base Costing (ABC) has grown in popularity in recent years as a substitute for the traditional methods of allocating factory overhead. This method bases the overhead allocation on many activities instead of only one. Although not yet widely adopted, ABC could mean the difference between bankruptcy and profitability for many manufacturers. Because of the relative newness of the technique, many accountants are not yet familiar with the advantages offered by ABC. In fact, most textbooks still do not cover ABC extensively. However, any accountant who has a manufacturer for a client should become familiar with ABC.

An ABC system assigned factory overhead on more than one basis. Although direct labor (hours or dollars) has long been the most common basis of allocating overhead, it has recently become obvious that such an allocation is not always relevant. Traditional cost accounting may be the cause of the declining competitiveness of some companies because cost accounting systems have simply not kept pace with changes in manufacturing technology. Manufacturers who have adopted ABC now use dozens of different allocation bases depending upon how production activity affects overhead costs. Some of these other allocation bases include machine hours, raw material costs, set-up time, waiting time, number of requisitions or purchase orders and number of units produced. One company reported using 37 different bases to allocate overhead. Each of these bases are typically called "cost drivers" because they represent activities that drive costs higher.

Historically, direct labor was a large part of total production cost, while overhead represented a smaller component. In most such cases, direct labor was the activity that drove (caused) overhead costs. Thus, it was appropriate to use direct labor as the allocation base. However, such is no longer the case because overhead now makes up a very large component of total production costs, while direct labor represents a small percentage of the total.

Overhead now comprises a large portion of total production costs because of the increased use of computers and robotics. In many companies, direct labor makes up less than 10% of total production costs. Allocating a very large cost (overhead) on the basis of a very small cost (direct labor) does not make rational sense. Also, a small change in direct labor on a particular product can make a significant difference in total production costs for that product (which may not be a valid assumption). For instance, if overhead is 800% of direct labor, a $1 savings in labor on a specific product would show an $8 lower cost for overhead, when actual overhead costs may not have changed at all.

Most overhead costs today vary in proportion to product diversity and the complexity of an operation. For a high technology company, direct labor is not a cost driver for most types of overhead costs.

ABC is not a new concept. Pioneer cost accounting writer Alexander Hamilton Church suggested the idea as early as 1908. However, ABC was not cost beneficial until the advent of widespread computer use because manually allocating overhead over several different bases would be time consuming. Nor was ABC needed when overhead costs were low relative to direct labor. For example, assume that in a traditional direct labor-based system direct labor totaled $100 and factory overhead was $10. If the direct labor were allocated on the basis of $60 to product A and $40 to Product B, the overhead allocation would be $6 to Product A and $4 to Product B, making the total conversion costs of the two products $66 and $44.

Alternatively, assume that the $10 of overhead was in the form of set-up costs, represented by $5 for setting up Product A and $5 for setting up Product B. An ABC system would assign the $10 of set-up costs on the basis of set-up time rather than direct labor. Thus, an ABC system would result in total conversion costs of $65 for Product A and $45 for Product B.

Alternatively, assume that the $10 of overhead was in the form of set-up costs, represented by $5 for setting up Product A and $5 for setting up Product B. An ABC system would assign the $10 of set-up costs on the basis of set-up time rather than direct labor. Thus, an ABC system would result in total conversion costs of $65 for Product A and $45 for Product B.

Now the latter figures in Table 2 ($65 and $45) are not much different from the $66 and $44 in Table 1 which were obtained under the traditional direct-labor-based system. Because the degree of difference between the two methods is so small, management was not materially harmed by any errors created in situations where direct labor was not a cost driver. Essentially, when overhead costs were low (in relation to direct labor costs), there was little damage caused by the error of making all allocations on the basis of direct labor (if there even was an error).

If the above manufacturer replaces its direct laborers with robots and computers, the cost structure would be more heavily weighted on the overhead side. Assume that overhead is $100 while direct labor is $10. If the direct labor represents $6 on Product A and $4 on Product B, the allocation for overhead would be $60 for Product A and $40 for Product B. The total conversion costs for each product would be $66 and $44 respectively.

The ABC system produces a significantly different answer. If the $100 was incurred in setting up the two jobs, then ABC might result in an allocation of $50 for each product, or total conversion costs of $56 for Product A and $54 for Product B. These latter costs represent a 23% difference between direct-labor-based costing and reality. This large difference between the two methods occurs because overhead is so large relative to direct labor, which is the traditional allocation base.

Table 1
Traditional Allocation:       Product A   Product B
Direct Labor                     $60         $40
Factor Overhead (10% of DL)        6           4
Total Conversion Cost            $66         $44
Table 2
Activity-Based Costing:       Product A   Product B
Direct Labor                     $60         $40
Factor Overhead (ABC)              5           5
Total Conversion Cost            $65         $45
Table 3
Traditional Allocation:         Product A   Product B
Direct Labor                        $6          $4
Factor Overhead (1000% of DL)       60          40
Total Conversion Cost              $66         $44
Table 4
Activity Based Costing:       Product A   Product B
Direct Labor                      $6          $4
Factor Overhead (ABC              50          50
Total Conversion Cost            $56         $54

ABC is more useful to a company when overhead costs are relatively high compared to other types of costs. Also, the more diverse a factory's product line, the more helpful ABC will be. It is only when all products are absolutely uniform that simple averaging procedures such as direct-labor-based costing are valid. For instance, a simple allocation basis in a factory with both large and small machines, and both high-priced and low-cost labor, all working together, would not be very exact.

An ABC costing system will also be quite beneficial whenever there are significant volume differences among a company's products. For example, assume that a company produces two similar products (perhaps golf clubs for adults and a similar product for children). Raw materials are $20 per unit, direct labor is $70 per unit and factory overhead totals $20,000. The company produces 1,000 units of the first product and 100 units of the second product. Using direct labor as the allocation basis, the costs would be as shown in Table 5.

Alternatively, assume that the overhead is represented by set-up costs, with one set-up necessary for each product. Thus, the $20,000 would be allocated equally under an ABC system. The ABC costs would be as shown in Table 6.

Because of the low volume of Product 2, the difference between the traditional allocation basis and ABC is quite significant. If the company were selling Product 2 at $150 each (resulting in a nice profit based on the $108.18 direct-labor-based cost), it would be losing money on every sale. However, things get significantly worse when overhead becomes a higher proportion of total costs. Assume that direct labor costs are only $10 per unit and that overhead totals $140,000. The traditional allocation basis would result in the costs shown in Table 7.

Table 5
                           Product 1        Product 2
Raw Materials               $20,000           $2,000
Direct Labor                 70,000            7,000
Overhead (25.97% x DL)       18,182            1,818
Total Cost                 $108,182          $10,818
Cost per Unit            $108.18 each     $108.18 each
Table 6
                         Product 1        Product 2
Raw Materials             $20,000           $2,000
Direct Labor               70,000            7,000
Overhead                   10,000           10,000
Total Cost               $100,000          $19,000
Cost per Unit          $100.00 each      $190.00 each
Table 7
                         Product 1        Product 2
Raw Materials             $20,000          $2,000
Direct Labor               10,000           1,000
Overhead (% of DL)        127,273          12,727
Total Cost               $157,273         $15,727
Cost per Unit             $157.27         $157.27
Table 8
                          Product 1       Product 2
Raw Materials               $20,000         $2,000
Direct Labor                 10,000          1,000
Overhead                     70,000         70,000
Total Cost                 $100,000        $73,000
Cost per Unit            $100.00 each   $730.00 each

Using the ABC system, the production costs would appear as shown in Table 8.

Thus, the combination of relatively high overhead and diversity in product volume results in costs that are four times those computed under the traditional costing method. Think about how a manager facing a competitive bid situation for Product 1 would react. The manager at the direct-labor-based company would bid some amount slightly greater than $157.27 and would wonder how a competitor could make a profit with a bid just over $100. Perhaps the competitor used the ABC system and had more relevant costing figures. Alternatively, what if the manager bid just over $157.27 on Product 2 and won the contract. His company would lose money simply because costs were not being properly allocated.

ABC is not for everybody. In some cases, direct labor is the predominant cost driver and the traditional method of overhead allocation is appropriate. But for any company with a high level of overhead relative to direct labor and multiple products which sell in widely differing volumes, ABC is the logical choice. Starting up an ABC system is not easy. Every step in the production process has to be evaluated and every individual component of factory overhead must be evaluated to determine the cause (driver) behind the cost. In summary, the process of implementing ABC is as follows:

a. Document the flow of the production process.

b. Determine input and output requirements, including staffing requirements for each process.

c. Define activities within each process.

d. Classify activities as value added or non-value added. Eliminate the non-value-added activities.

e. Determine the cycle time for each activity.

f. Determine which activities cause (drive) costs and the amount of those costs.

When the above steps have been accomplished, the result will be an overhead application formula much more complex than the simple percentage of direct labor to which most people are accustomed. For instance, an overhead application formula might be something like this: $1 x direct labor hours, plus $12 x each materials requisition, plus $1.20 x each hour spent awaiting processing, plus $2 per finished unit, plus $24 x each setup, plus $3 x each hour of cycle time. This is much more complex than the traditional method and a computer is a virtual necessity, but the result can mean business success for the manufacturer.

Conclusion

The traditional direct-labor-based method of assigning overhead has been used primarily because of its simplicity. The cost of collecting the information needed to use more than one activity base has, until recently, been quite high relative to the benefits generated. Now, because of the advantages offered by computers and bar code technology, the cost of processing cost information has decrease. Thus, an ABC system -- based upon multiple cost drivers -- can now be profitably used. Companies have begun adopting ABC because of its capability to deal with costing problems that conventional cost accounting either creates or fails to address.

Examples of these problems include suboptimal pricing decisions, poor allocation of costs and incorrect direction by management. For example, with overhead allocated on the basis of 700% of direct labor, managers will try to reduce direct labor costs by $1 in order to reduce the amount of overhead allocated by $7. In reality, the company would be better off if the manager ignored direct labor. Cost cutting efforts should be concentrated on cost drivers such as set-ups, engineering changes and movement of materials.

The advantages of activity-based costing can perhaps best be summarized by quoting Alexander Hamilton Church's 1910 justification for the technique:

...few of the expenses in the profit and loss account have any relation either to each other or to wages or to time. To rely upon an arbitrary established percentage which may actually be either much over, or much under, the real incidence of a number of varied factors on a particular order, may be a good way of getting rid of figures and giving an air of finality to cost accounts, but it is very little else. As a guide to actual profitableness of particular classes of work it is valueless and even dangerous.(1)

Footnotes

1 Alexander Hamilton Church, "Organization by Production Factors," Engineering Magazine (April, 1910), pp. 80-81.

Dale L. Flesher, PhD, CPA, CMA, is the Arthur Andersen Alumni Professor of accountancy at the University of Mississippi. He received both his bachelors and masters degrees from Ball State University. He holds a PhD in accounting from the University of Cincinnati. He has published more than 150 articles in numerous professional journals, and is the author of 26 books. He is currently the editor of The Accounting Historians Journal. He is the recipient of a number of research and teaching awards.