Dynamo's Second Dilemma
Dale Dynamo, owner of Dale's Miraculous Medical Devices (DMMD), is a forward-thinking business owner. We learned of his strategy to adopt lean thinking and some of the costing issues he faced. At the same time, he realized that many of his manufacturing operations were not conducive to cellularization; they would, at least for the foreseeable future, have to remain deployed in a job shop configuration with like-kind equipment located in groups and the parts being manufactured moved from machine to machine as the manufacturing process takes place.
One afternoon, as Dynamo reviewed a group of customer quotations, he noticed something that didn't appear to make a lot of sense. Two of the quotations required the use of the same mid-sized machine center, but the cost per hour for using that machine to produce one part was twice as much as the cost per hour for producing the other. He called in his Controller, Marvin Myopia, for an explanation.
"That's simple," explained Myopia, "one part is large and difficult to handle so two workers are required to load and unload the part when we do the machining; the other one is much smaller and easier to handle, so it only requires one worker. As a result of needing two workers instead of just one, the hourly cost for the large part is double the cost of the other." Myopia then backed up the explanation with the analysis of the company's 376 percent direct labor-based overhead rate shown in Exhibit 1.
Exhibit 1: DMMD's Direct Labor-Based Overhead Calculation
"You see," continued Myopia, "we take all of the costs we can attribute to the machining operations, like the workers' fringes and taxes, shift and overtime premiums, depreciation, utilities, purchased maintenance services and the supplies we buy to support machining, and we assign them to the overhead pool. We then analyze the company's support departments and assign portions of them to the pool as well. This includes a charge for the space it occupies, human resource support, maintenance performed by our employees, the department's supervisors and the work done by our engineers to support the machine centers. When we add them all up, it comes to $1,084,000. We then divide this by the $288,000 paid to the workers and arrive at the 376 percent rate. Since two workers are needed to produce one of the parts, the total hourly cost is twice as much for that part as it is for the part that only requires one worker."
After mulling over Myopia's explanation for a few minutes, Dynamo said, "Marvin, this still doesn't make any sense to me at all. First, we have three different sizes of equipment in that machine shop. I'm sure the small units don't use as much power, have as much depreciation or consume the same maintenance and operating supplies as the large ones. And the mid-sized machines probably require a level of those costs something in between those two. Second, we've got two different levels of pay for the workers; the more skilled ones make $16 per hour and those with less skill $12 per hour. Shouldn't those factors be considered when we measure the cost of operating these machines? Not only that, I can't believe that all those overhead costs actually double when there are two workers—not just one—running the same machine."
"That may be the case," answered Myopia, "but our auditors think our methods are perfectly okay, and they're the experts, aren't they?"
An Engineer's View
Dynamo didn't find Myopia's explanation very satisfactory, so he set his engineer's brain working on the problem and soon arrived at three conclusions:
The first conclusion he reached was that the cost of labor and the cost of operating the equipment should be treated as two separate issues; there's no real rationale for linking the two together when measuring costs. After all, economists break the factors of production into man, material, machine and money, so why do accountants insist on combining man and machine into one category? It also seemed unlikely to him that if a lower-skilled individual tended a machine instead of a higher-skilled worker, it would reduce the cost of the machine's operation by $15.04 as would be the case using the company's 376 percent direct labor-based overhead rate. Dynamo also reasoned that there should be two different labor costs; one for the more highly-skilled workers with a higher wage rate, and another for those having less skill and a lower wage rate.
He further reasoned that the cost of operating the equipment could not possibly be uniform; the cost of his large machines should be higher than the small ones and the cost of the mid-sized machines somewhere in between.
To test his ideas, Dynamo took Myopia's analysis and broke the costs down into five categories—two categories for the labor and three for the equipment. By using some simple measures, data available from the accounting system and estimates by those actually involved in the manufacturing process, he prepared the breakdown shown in Exhibit 2.
Exhibit 2: Dynamo's Analysis of Machining Costs
Using his engineer's logic, he assigned certain headcount-driven fringe benefits—like health insurance—between the two labor categories based on headcount. Other benefits and taxes he split on the basis of payroll dollars. He also reasoned that the human resource and supervision cost should be divided between the two categories of production labor based on hours worked. Depreciation, purchased maintenance and manufacturing supplies were divided among the three machine groups based on an examination of the underlying accounting records. Utilities were split on the basis of each machine category's relative power consumption and the number of hours they operated during the year. Building and grounds was assigned based on each machine group's footprint. Finally, maintenance and engineering were split after getting the best estimates from those actually performing those services.
Anxious to see how this would compare to the "per hour" cost on the two quotations that started his thought process, Dynamo calculated both quotations' hourly costs "his way" and compared them to Myopia's calculation. The equipment in question was a mid-sized machine and the worker was a highly-skilled individual, so he took his $30.00 labor rate and added the mid-sized machine's $60.00 rate for a total hourly rate of $90.00—18 percent higher than taking the $16.00 labor rate and adding 376 percent for a total of $76.16. The difference was almost reversed, however, when two workers were involved. Using Dynamo's method, the hourly labor cost doubled to $60.00, but the machine rate of $60.00 remained the same—a total of $120.00 –21 percent lower than Myopia's calculation, which adds 376 percent to a $32.00 labor cost. These two calculations, which are summarized in Scenarios A and B of Exhibit 3, raised serious concerns in Dynamo's mind about the quality of the decisions he and his other managers are making whenever they must rely on the accounting department's cost information.
Exhibit 3: Dynamo's Comparison of Costing Methods
As he pondered the situation, it occurred to Dynamo that the customer quotations that brought the issue to his attention represented only half of the problem; not only do his machine centers sometimes require multiple workers, but there are times when they only require fractional workers. The cycle times required to produce some parts are so long that a worker assigned full-time to that machine would be idle a great deal of the time. To make better use of the workforce, DMMD often schedules work so that one individual can tend two machines; while one machine goes through its cycle, the worker loads/unloads the other machine. What impact might that have on the measurement of costs?
To answer his question, Dynamo compared the hourly cost that would be determined by his method and accounting's method for two additional scenarios: one where a low-skilled worker would spend one-half of his time tending a large machine vs. one where a high-skilled worker would tend a small machine. The results appear as Scenarios C and D in Exhibit 3. A half hour of low-skilled labor operating a large machine would result in a $28.56 hourly rate under the current method, while that same situation would result in a $92 per hour rate under his—a 222 percent difference. A half hour of highly-skilled labor operating a small machine would result in an hourly cost of $38.08, while his method would measure that cost a $45 per hour—an 18 percent difference that seems small compared to the 222 percent difference, but one that could still mislead a decision maker.
Not one to jump to conclusions, Dynamo tried two more situations: one in which a two-man crew comprising a high-skill and a low-skill worker operate a small machine, and one where a high-skill worker operates a large machine (Scenarios E and F in Exhibit 3). The costs came out 37 percent lower and 44 percent higher, respectively, using his method. Dynamo knew he had a problem.
When he presented the results of his analysis to his Controller, Myopia responded, "Yeah, but your method is a lot more complicated and the auditors have already accepted the percentage of direct labor method." (Apparently, the finance department believes that the purpose of cost information is to make their job easy, not to help the company become more profitable.)
Bottom Line Impact
The value of accurate and relevant information has never been greater than in today's ultracompetitive, worldwide marketplace. The financial success of an organization is dependent upon the quality of the decisions made by its managers, and the quality of those decisions is dependent upon the quality of the information given to them as support. Although cost information as inaccurate as that generated by DMMD's accountants will negatively impact all sorts of decisions—from capital spending to vendor selection and from outsourcing to process improvement—I will focus here on one decision category: core business pricing.
The quotations that initially brought the issue to Dynamo's attention indicated that the manufacturing cost of the smaller product was $76.16 per hour and the larger one $152.32 per hour. Dynamo's more reasoned measures were $90 and $120. Undercosting the smaller product can lead the company into believing that it will make money at a price that doesn't, in reality, cover its costs. On the other hand, overcosting the larger product misleads the company into believing a very high price—probably one the market will not accept—is required to sell the job at a profit. Winning contracts at a price that won't cover all the company's costs while losing out on potentially profitable contracts whose costs were overstated by a dysfunctional costing model, is not a formula for financial success.
A manufacturer who fails to segregate those indirect and support costs driven by the operation of equipment* from those driven by the work of individuals will seldom arrive at product or process costs that reflect economic reality. Although it may not remedy all of a manufacturer's costing issues, accurately linking costs to both the work of production workers and the operation of equipment and then assigning those costs to jobs, products or contracts based on separate labor and machine (or equipment/cell/line) rates will eliminate one of the most damaging features of a direct labor-based costing model.
*The concept is the same with regard to cells and lines; a cost per hour for operating the cell's or line's equipment separated from the cost of individuals working in the cell or on the line.