In the StoreBox example, we discussed variable-, fixed-, and mixed-cost functions using information about future cost structures StoreBox was considering. Often, however, managers use cost estimation to measure a relationship based on data from past costs and the related level of an activity. Managers are interested in estimating past cost functions primarily because they can help them make more accurate cost predictions, or forecasts, of future costs. For example, to choose the design features for its new TV models, LG’s managers use past cost functions to evaluate the costs of alternative designs and compare it to what customers are willing to pay. Similarly, marketing managers at Audi identify cost drivers of customer-service costs (for example, the number of new car models introduced or the total number of cars sold) and the fixed and variable components of these costs. They use this information to prepare the customer-service budget.
The most important factor in estimating a cost function is identifying the cost driver—a variable such as the level of activity that causally affects costs. Without a cause-and-effect relationship, managers are less confident about their ability to estimate, predict, or manage costs. We use the terms level of activity and level of cost driver interchangeably when estimating cost functions. The cause-and-effect relationship might arise as a result of the following:
Managers must be careful not to interpret a high correlation between two variables to mean that either variable causes the other. Consider Winston Furniture, which makes two types of (otherwise identical) tables, one with a granite surface and the other with a wooden surface. Granite tables have higher direct material costs but are available in precut blocks and so have lower direct manufacturing labor costs. Winston currently sells 10,000 granite tables and 30,000 wooden tables.
If Winston sells 20% more of each type of table (12,000 granite tables and 36,000 wooden tables), then total direct materials costs and total direct manufacturing labor costs will each increase by 20%. The two cost categories are highly correlated in this case, but neither is a cost driver of the other. Using one cost to predict the other is problematic.
To see why, suppose again that Winston sells 20% more tables (or a total of 48,000 tables), but now 18,000 are granite and 30,000 are wooden tables. Total direct manufacturing labor costs will increase by less than 20% because granite tables require less labor. Total direct materials costs will increase by more than 20% because granite is more expensive than wood. Total direct manufacturing labor costs would be a poor predictor of total direct materials costs. The best predictor of total direct materials costs are its cost drivers, the number of each type of table produced.
Only a cause-and-effect relationship—not merely correlation—establishes an economically plausible relationship between the level of an activity and its costs. Economic plausibility gives analysts and managers confidence that the estimated relationship will appear repeatedly in other sets of data. Identifying cost drivers also gives managers insights into ways to reduce costs.