2.1 Target costing
After preliminary market research and future sales volume planning, the target costing model starts by calculating the AC:
$$AC = Price \cdot \left( {1 - Profit\;Margin} \right) \cdot Volume - SG{\& }A\;Expense.$$
(1)
The AC are then compared to the product’s DC to determine the cost gap. If the DC are higher than the AC, the implication is to perform cost splitting and to optimize the cost structure in a customer-oriented way to approximate the DC to the AC (e.g., Kato
1993). For the customer-oriented optimization of cost structure, the block of target costs is split with respect to customer utilities instead of using total AC. Within the customer preference analysis, utility values are assigned to each feature of the product (Woods et al.
2012). The product’s features result from the product’s components used and the assigned feature utilities are proportionally matched with their components. To derive the allowable costs (
\(AC_{i}\)) per component
i, total AC are split upon the product’s components based on the utility proportion each component provides (Woods et al.
2012). Conversely, if DC are smaller than or equal to the AC, then the product’s cost structure is optimal if the proportion of component
i’s utility value is equal to the proportion of component
i’s drifting costs (
\(DC_{i}\)). The ratio of both is then defined as the Target Cost Index (
\(TCI_{i}\)) of component
i:$$TCI_{i} = \frac{{\mathop \sum \nolimits_{j = 1}^{J} \left( {u_{j} \cdot v_{ij} } \right)}}{{\frac{{DC_{i} }}{DC}}},$$
(2)
where J denotes the total number of features,
\(u_{j}\) denotes the utility of feature
j, and
\(v_{ij}\) the share of component
i in contributing to feature
j with
\(\sum\nolimits_{{i = 1}}^{I} {v_{{ij}} = 1}\). This index contains information about the relative cost-(in)efficiency of each component
i. If the
\(TCI_{i}\) is not equal to one for all components of a product, costs should also be optimized through customer-oriented optimization of the cost structure, irrespective of DC being smaller than or equal to AC. This step is economically intuitive because even though the DC might be smaller than or equal to the AC, the costs for some components are too high or too low relative to their contribution to the product’s total utility. Through a cost-value analysis by using the
\(TCI_{i}\), management can analyze into which component
i it should invest (divest) in accordance with a high (low) customer utility of the component. Therefore, the customer's willingness to pay increases for features with high utility and cost decreases for features with low utility. In case the component is too costly relative to the attributed customer utility,
\(DC_{i}\) have to be reduced to reach
\(AC_{i}\). Cost optimization is completed when it applies that
\(DC_{i}\) are equal to
\(AC_{i}\) for each component
i (Sakurai
1989; Woods et al.
2012).
2.2 Target costing including a minimum variant
The theoretical framework above hinges on the assumption that customer preferences are independent and, thus, on the additivity of their utility functions (Woratschek
1995). In accordance with Götze and Linke (
2008), we argue that this assumption is economically plausible only if specific minimum requirements are met.
4 As an example, consider a car without wheels but an engine that leads to high customer utility. Such a car should still have a total utility of zero since it cannot function without wheels. Accordingly, we cannot add up utility values of a product’s components until the product’s intended minimum functions are provided. Coenenberg et al. (
2003) introduce a minimum variant of a product that has features with a customer value of zero. Their example implies that these features are not leading to an increase in costs compared to the costs for the minimum variant. Nevertheless, features with zero utility for the customer, still face costs in production. The implication of a strict interpretation of target costing would be to reduce costs to zero for all features that have a customer utility of zero. However, for a minimum variant of a product, it is reasonable to implement the features while having costs greater than zero. To support this thought, we revisit the car example and consider a functioning car without any special features as our minimum variant. According to the framework, this car has zero customer utility. Yet, the production of this car still leads to costs for the company. Only after reaching the minimum requirements of the car, we can go further and add features that lead to customer utility. Coenenberg et al. (
2003) acknowledge the existence of a minimum variant, but do not account for a solution for the model’s violation to have costs for features with zero utility. Furthermore, an additive utility function is still used for the total amount of costs including the cost for the minimum variant. Therefore, the aforementioned limitation of target costing is not solved. Götze and Linke (
2008) start to address this limitation. They assume that specific basic requirements of a product must be attained before an additive utility function can be used. More precisely, they do not include the costs for the basic requirements in the calculations for cost-optimization, which is in line with our framework. Yet, they do not specify how to treat costs for the basic requirements considering that they have zero utility for the customer but lead to costs for the company.
Based on the limitation of target costing not accounting for the dependence of features and preferences until the minimum requirements of a product are reached, we integrate a minimum variant in the modified target costing approach. To facilitate comprehension of our following theoretical model framework, we provide a numerical example in the appendix. We argue that a customer-oriented optimization for product components is economically unreasonable until the production of a minimum variant is achieved. Our basic idea, which distinguishes our note from previous work, is to disaggregate AC into allowable costs to achieve the minimum variant (
\(\underline{AC}\)) and allowable costs to achieve the presently favored variant (
\(\overline{AC}\)). The favored variant of a product is characterized by offering features that go beyond the minimum requirements and are demanded by customers in the market. For the process of customer-oriented splitting of target costs we only take into account the block of costs that does not include
\(\underline{AC}\):
$$AC^{mod} = \overline{AC} - \underline{AC} ,$$
(3)
where
\(AC^{mod}\) denotes the modified allowable costs. Along with the definition above, the customer-oriented optimization of costs of product components is conducted for the DC that do not include
\(\underline{AC}\). With
\(\overline{DC}\) denoting the drifting costs of our favored variant, we define our modified drifting costs (
\(DC^{mod}\)) as follows:
$$DC^{mod} = \overline{DC} - \underline{AC} .$$
(4)
It is crucial that we do not subtract the drifting costs (\(\underline{DC}\)) of our minimum variant to receive \(DC^{mod}\). In general, \(\underline{DC}\) are higher than \(\underline{AC}\). Using \(\underline{DC}\) to calculate \(DC^{mod}\) would lead to biased modified drifting costs. Therefore, the cost would not be completely optimized since the gap between \(DC^{mod}\) and \(AC^{mod}\) would be smaller. At this point, it is important to underline that the modified target costing leads to the same amount of required adjustment in cost as the original target costing model. The difference lies in the allocation of required adjustments on the component level. With the modified target costing approach, allowable costs are assigned more reasonably to the product’s components.
As for original target costing, the implication for the customer-oriented optimization is to stop the optimization process when
\(DC^{mod}\) are smaller than or equal to
\(AC^{mod}\). Hence,
\(\overline{DC}\) are smaller than or equal to
\(\overline{AC}\)5:
$$\overline{{DC}} - \underline{{AC}} \le \overline{{AC}} - \underline{{AC}} \Leftrightarrow \overline{{DC}} \le \overline{{AC}} .$$
(5)
For the minimum variant, we define a customer value equal to zero. If the customer value of a product is zero, it means that the minimum variant is produced and that the favored variant of the product matches the minimum variant. According to this assumption,
\(\underline{AC}\) are equal to
\(\overline{AC}\), leading to
\(AC^{mod}\) of zero in Eq. (
3). Therefore, customer-oriented splitting of target costs is not conducted since
\(AC^{mod}\) are the only block of costs for which one would conduct customer-oriented splitting of target costs. In case a company only produces the minimum variant, it is required that the product is produced with the respective
\(\underline{AC}\). For a company to be able to produce a product with the respective
\(\underline{AC}\),
\(\underline{AC}_{i}\) need to be computed. The calculation of
\(\underline{AC}_{i}\) differs from the determination of
\(AC_{i}\), since we do not use customer-oriented splitting of target costs. To define
\(\underline{AC}_{i}\), drifting costs for the minimum variant per component
\(\underline{DC}_{i}\) are reduced by an identical factor
\(\underline{AC} /\underline{DC}\). This factor can be considered as a measure for the gap to a hypothetical ideal company in the “low-cost segment”:
$$\underline{AC}_{i} : = \frac{{\underline{AC} }}{{\underline{DC} }} \cdot \underline{DC}_{i} .$$
(6)
For this calculation we only need to determine
\(\underline{AC}\), assuming information about
\(\underline{DC}\) and
\(\underline{DC}_{i}\) are given. Even though we do not conduct customer-oriented splitting of target costs for the minimum variant, the framework still determines the total
\(\underline{AC}\) with respect to market-orientation. We use
AC from Eq. (
1) with the implementable price and sales volume for a minimum product. Here, the implication is to stop the adjustment of costs when
\(\underline{DC}\) equaling
\(\overline{DC}\) are smaller than or equal to
\(\underline{AC}\) equaling
\(\overline{AC}\):
$$\underline{DC} = \overline{DC} \le \overline{AC} = \underline{AC} .$$
(7)
We further highlight that, by introducing \( \underline{{{\text{AC}}}}\) in our model for modified target costing, we do not imply that companies should only have costs up to \(\underline{AC}\). This applies only, if companies are actually producing the minimum variant as discussed above. In the case of a company, which operates in a differentiated high-quality segment instead of a low-price segment, it could be reasonable to produce a favored variant of the product for which actual costs exceed the costs of a minimum variant. E.g., most car manufacturers do not produce cars which are only supposed to fulfill the minimum requirements of a car. Consequently, their production costs for the favored variant are not restricted to \(\underline{AC}\).
If a company produces a favored variant of a product that adds features to the minimum variant, these costs in excess of
\(\underline{AC}\) can be optimized on the basis of customer-oriented target cost splitting. As argued above, the independence of preferences and therefore, an additive utility function can be implied after reaching the minimum variant of a product and does not cause any distortions in our model. Consequently, utilities can be added or subtracted based on customer preferences.
\(AC^{mod}\) can be seen as the respective costs for these added features that lead to higher customer utility. Consequently, it is reasonable to use
\(AC^{mod}\) from Eq. (
3) for the customer-oriented splitting of target costs. Since production goes beyond the minimum variant, the company is not required to reach
\(\underline{AC}\). However, to calculate
\(AC^{mod}\),
\(\overline{AC}\) have to be determined. We use
AC from Eq. (
1) with the implementable price and sales volume for the favored product to determine
\(\overline{AC}\). With available information on
\(\overline{DC}\),
\(DC^{mod}\) can be calculated according to Eq. (
4). To conduct the customer-oriented optimization of the cost structure of a product, total
\(AC^{mod}\) have to be split based on the customer utilities to receive the modified allowable costs (
\(AC_{i}^{mod}\)) per component
i. Further, the modified drifting costs per component
\(DC_{i}^{mod}\) have to be calculated. To receive
\(AC_{i}^{mod}\), we use the known customer-oriented splitting of target costs for the residuum of costs after deducting
\(\underline{AC}_{i}\). The total
\(AC^{mod}\) are weighted by the utility proportion of the
\(i^{th}\) component.
$$AC_{i}^{mod} = \mathop \sum \limits_{j = 1}^{J} \left( {u_{j} \cdot v_{ij} } \right) \cdot AC^{mod} ,$$
(8)
where
\(u_{j}\) denotes the utility of feature
j, and
\(v_{ij}\) the share of component
i in contributing to feature
j with
\(\mathop \sum \limits_{i = 1}^{I} v_{ij} = 1\). The total allowable costs (
\(\overline{AC}_{i}\)) for the favorable variant of component
i are the sum of
\(AC_{i}^{mod}\) and
\(\underline{AC}_{i}\).
$$\overline{AC}_{i} : = \frac{{\underline{AC} }}{{\underline{DC} }} \cdot \underline{DC}_{i} + \mathop \sum \limits_{j = 1}^{J} \left( {u_{j} \cdot v_{ij} } \right) \cdot AC^{\bmod } .$$
(9)
For given
\(\overline{DC}\) and drifting costs (
\(\overline{DC}_{i}\)) of the favored variant per component
i,
\(DC^{mod}\) and modified drifting costs (
\(DC_{i}^{mod}\)) per component
i can be calculated by subtracting
\(\underline{AC}\) and
\(\underline{AC}_{i}\) from
\(\overline{DC}\) and
\(\overline{DC}_{i}\), respectively.
$$DC^{mod} : = \overline{DC} - \underline{AC}$$
(10)
$$DC_{i}^{mod} : = \overline{DC}_{i} - \underline{AC}_{i} .$$
(11)
To perform a cost-value analysis for the product components, the proportion of
\(DC_{i}^{mod}\) to
\(DC^{mod}\) is compared to the proportion of utility of component
i to the product’s total utility. This comparison leads to the calculation of the target cost index, as defined in (2) using modified drifting costs. For the modified target costing, the same implications of the
\(TCI_{i}\) apply as for the original target costing. If the
\(TCI_{i}\) is equal to one, the costs are optimized. If the
\(TCI_{i}\) is not equal to one, costs need to be adjusted. In case the cost structure is not optimized, the proportion of
\(DC_{i}^{mod}\) to
\(DC^{mod}\) differs from the proportion of utility of the
\(i^{th}\) component. The required adjustment in costs can be calculated as follows:
$$required\, adjustment\, in\, costs = \left( {\frac{{DC_{i}^{mod} }}{{AC^{mod} }} - \mathop \sum \limits_{j = 1}^{J} \left( {u_{j} \cdot v_{ij} } \right)} \right) \cdot AC^{mod} .$$
(12)
Hence, we subtract the proportion of utility one component has from the proportion of costs of the same component based on \(AC^{mod}\) and multiply the result by \(AC^{mod}\). Overall, our modified model and the original target costing approach result in the same required cost adjustment. Yet, the required adjustments per component i differ. Therefore, indicated required cost adjustments for product components will be biased in the original target costing model. In contrast, our modification based on \(AC_{i}^{mod}\) leads to more reasonable cost adjustments per component i since they take only the portion of costs into account that can actually be optimized with respect to customer utility.
For an easier understanding of our modified procedure, the illustrative example is based on a rather simple product understanding. This holds in particular for the minimum variant. We acknowledge that the majority of real-world applications of target costing will have to deal with complex products and their complex minimum variants. Hardware components are often no longer sufficient to characterize today’s product concepts. For instance, manufacturing firms are increasingly shifting to products in the form of hybrid offerings that combine goods and services (e.g., Ulaga and Reinartz
2011) so that certain minimum requirements for software, maintenance contracts, real-time and ex-post monitoring may well be necessary depending on the specific product.
6 This is not only observable in the B2C business but also in the B2B business, in which customers evaluate holistic business solutions instead of isolated products (e.g., Macdonald et al.
2016). Although these more complex product definitions might not be imperative to enable a product’s most basic function, they can still be a quite obvious requirement for the product to be acceptable for the consumer such that it makes sense for a producer to consider them in defining the minimum variant. Our modified target costing is able to capture these complex products by defining the minimum variant as a bundle of components belonging to different product functions. In any case the determination of a minimum variant should undergo a considerable amount of scrutiny to ensure that it really fulfills the customer’s minimum requirements with respect to all important product functions.