|
Article Excerpt
1. Introduction
In 2004, a consumer research study in the auto industry reported initial product quality as the second most important factor affecting consumers' purchasing decision after product price (J.D. Power and Associates 2004). (1) The large number of product recalls and lawsuits in the auto industry demonstrate how undetected quality problems and related production delays can lead to a huge profit loss and degrade a company's brand equity. For instance, in 2007, Ford's concerns about a design related quality problem in cruise control switches resulted in a recall of 3.6 million vehicles manufactured between 1992 and 2004, increasing the total number of vehicles recalled for the same quality problem to 9.6 million (Associated Press 2007). In the electronics industry, in April 2007, Sanyo agreed to share with the manufacturer Lenovo, the $17 million cost of recalling 205,000 Sanyo made laptop battery packs that can overheat because of a flaw in the product design (Nystedt 2007). In May 2007, the Consumer Product Safety Commission, the National Highway Traffic Safety Administration, and Evenflo Company Inc. announced a recall of Evenflo embrace infant car seat/carriers because of a malfunctioning handle. A total of 450,000 units, manufactured in the United States and China, were sold nationwide through department stores and baby items stores (CNNMoney 2007).
These are just a few examples that demonstrate that recalls are common in a variety of industries and often are associated with substantial present and future costs to a company. The cost and scale of recalls necessitate a deeper understanding of how to manage the quality improvement incentives of multiple supply chain partners to ensure better product performance.
Product recalls result from a lack of quality assurance in the manufacturing and/or design processes of one or many supply chain partners and can affect a large number of products manufactured over extended periods of time. For example, in a recent study, Ford reported that 76% of the company's quality problems stem from its first tier suppliers (Sherefkin 2002).
Today, extended quality improvement efforts take various forms. For instance, manufacturers in the auto industry are more willing to involve suppliers during the product development process to ensure early detection and elimination of quality problems (Kisiel 2007). In addition to preventive initiatives, it is also becoming a common practice among manufacturers to present suppliers with quality cost sharing agreements to ensure accountability of quality problems and to create incentives for process improvement (Balachandran and Radhakrishnan 2005). In this paper, we address the optimal design of a recall cost sharing contract when both the supplier's and the manufacturer's quality improvement efforts are subject to moral hazard and when the manufacturer has uncertainty regarding the quality of the supplier's process. In this context, we discuss the optimal use of product failure root cause analysis information in the design of cost sharing schemes.
Previous research on contract design in quality management has studied the fixed share rate contract (2) (Contract F) as the external quality cost (such as recall cost) sharing scheme between a manufacturer and a supplier. In this paper, we introduce two new contract formats to share product recall related external quality costs in the supply chain: (i) the selective root cause analysis contract (Contract S), which is characterized by a unit part price (p), a fixed recall cost share rate (R), paid by the supplier, (3) and a threshold product failure time ([bar.T]); and (ii) the partial cost allocation contract (Contract P), which is characterized by a unit part price (p), a fixed cost share rate ([R.sub.m]) paid by the supplier if the manufacturer is responsible for the product failure, and a fixed cost share rate ([R.sub.s]) paid by the supplier if the supplier is responsible for the product failure. A critical component of both contract formats is the root cause analysis information, which reveals the supply chain member who is responsible for the quality problem in the product. Contract S uses this information only if the product failure occurs before a threshold time([bar.T]), which we will refer to as the root cause analysis threshold, and allocates the total recall cost to the party at fault. Otherwise, the cost is shared according to a fixed rate (R). Under Contract P, root cause analysis information is always used in the cost allocation process, the total recall cost is always shared between the parties, and the supply chain member who is responsible for the recall incurs a larger share of the recall cost.
Considering a single-manufacturer, single-supplier supply chain structure, we address the following research questions regarding these contractual agreements:
* How effective are selective root cause analysis and partial cost allocation contracts in coordinating the manufacturer's and the supplier's quality improvement efforts when the effort levels are not observable and therefore are subject to moral hazard?
* Which contract format is optimal for the manufacturer? How do the fixed share rate contract, selective root cause analysis contract, and the partial cost allocation contract compare with respect to the manufacturer's profits as well as the quality of the final product?
* If the exact information about the supplier's initial quality is not available, can the manufacturer design a menu of selective root cause analysis contracts to screen supplier type as well as to induce quality improvement effort? Under what circumstances does knowing the supplier's initial quality result in significant savings for the manufacturer? How much does the quality of the final product under the menu of contracts differ from that under perfect information (i.e., when the manufacturer knows the exact initial quality of supplier's product)?
In the following section, we present the contribution of our paper in the context of the existing literature on quality and supply chain management. Then, we present the basic modeling framework and the assumptions of our model in [section]3. Section 4 develops a detailed analysis of Contract S and Contract P under the complete information assumption. We present an extensive numerical study in [section]5, which examines the profitability of these contracts for the manufacturer as well as for the total supply chain. Section 6 investigates the case of information asymmetry and derives the optimal menu of selective root cause analysis contracts. Section 7 presents a numerical study to compare the cost efficiency and product quality under different contracts in cases of information asymmetry. A summary of our findings and directions for future research are presented in [section]8.
2. Literature
The main focus of this research is on modeling the process improvement incentives of supply chain members when their effort choices are not observable and there is information asymmetry with regards to their existing process capability. Therefore, by jointly modeling moral hazard and adverse selection issues, this paper contributes to several streams of research each of which we review below.
In operations management, a group of papers discusses the design of quality cost sharing contracts among manufacturers and suppliers. In a game theoretic set-up, Reyniers and Tapiero (1995a, b) and Lim (2001) model suppliers' choice of process quality and manufacturers' choice of inspection strategy. Their model characterizes the Nash equilibrium of the supplier-manufacturer quality game in terms of the cost sharing parameters for internal (rework) and external (warranty) quality costs, assuming a fixed rate for sharing external quality costs between the parties. We, however, model a more general contract format for sharing external quality costs resulting from a recall. A special case of our contract of interest is the fixed share rate contract studied in the above papers.
Baiman et al. (2000) analyze the relationship between product quality, cost of quality, and the information that can be contracted on. In a risk neutral setting, the supplier invests in reducing the process defect rate and the manufacturer invests in the inspection quality of the incoming part. Both decisions are subject to moral hazard. Like Reyniers and Tapiero (1995a, b) and Lim (2001), they also assume that the external quality costs are shared at a fixed rate. In a subsequent paper (Baiman et al. 2001), the authors investigate the link between product design, contractible information, and the supplier's investment in process quality. In contrast to this paper, which considers fixed share rate contract, we focus on a broader set of contract formats to share external product failure costs and show that, even though the root cause analysis can perfectly determine the party responsible for product failure, it is not optimal for the supply chain to share quality costs based on this information for all failures occurring during the contract period. We propose a contract with selective root cause analysis which differentiates early failures from late failures to coordinate the quality improvement efforts of supply chain members.
In a subsequent paper, Baiman et al. (2003) examine a product structure exhibiting the weakest link property and investigate how the internal and external failure cost sharing mechanisms impact supplier selection when there is an adverse selection problem. Their analysis considers moral hazard only on the supplier side, whereas we model moral hazard both on the manufacturer and on the supplier side. Furthermore, like previously cited work, their analysis also assumes that the external quality costs are shared at a fixed rate, which is, in fact, a special case of the contract we investigate.
Balachandran and Radhakrishnan (2005) consider a double moral hazard situation in quality investment effort, in which the final product consists of components made by a buyer and a supplier. Although their paper focuses on the best use of incoming inspection information to achieve the first best effort levels from the supply chain partners, in this paper we investigate the best use of root cause analysis information about external failures to achieve first best effort levels from supply chain members. Furthermore, Balachandran and Radhakrishnan (2005) model the fixed share rate contract for allocating the costs of internal failures, whereas we consider a more general contracting arrangement for external failures.
In a recent paper, Zhu et al. (2007) look at a buyer who designs a product and owns the brand, yet out-sources the production to a supplier. Both the buyer and the supplier incur quality related costs that are shared by a fixed share rate contract. The Zhu et al. (2007) model captures the effect of the buyer's involvement in ensuring product quality. They also endogenously model the effect of operational decisions such as the buyer's ordering quantity and the supplier's production lot size. Unlike Zhu et al. (2007), we look at a setting where the manufacturer is involved in the production process and his effort affects the final quality of the product and discuss two new contract formats to share external quality costs.
A related supply chain management paper by Corbett and DeCroix (2001) discusses the use of a shared savings contract (assuming a fixed share rate between a supplier and a buyer) to induce supplier and buyer effort that reduces indirect material consumption. Although the modeling of effort in our paper has some similarity to their modeling constructs, we investigate the use of contractual formats to share external quality costs resulting from a recall rather than the cost of indirect materials.
Based on data from the automotive and the pharmaceutical industries, a number of political economy research papers investigate the real total cost of a recall for a manufacturer. For instance, Jarrell and Peltzman (1985), Barber and Darrough (1996), and Rupp (2004) study the cost attributes of recalls in the U.S. automotive industry and find that the indirect costs such as brand equity loss, consumer goodwill loss, and loss in firm value are in fact much larger than the direct costs of a recall such as product collection and repair cost. The findings of this stream of empirical research serve as a basis for some of our assumptions regarding the manufacturer's unit recall cost.
In summary, this paper introduces two new contractual formats for sharing the external quality costs of product recalls; in particular, we focus on the best use of root cause analysis information and its impact on the quality of the final product under both complete and asymmetric information assumptions. In this respect, our findings enrich the growing literature in this area and help managers to better understand the cost efficiency of these contractual agreements and their impact on product quality.
3. Modeling Framework
We investigate the implications of contract choice to share product recall costs on manufacturer's and supplier's quality improvement efforts and on supply chain profits. To this end, we consider a manufacturer who produces a product that consists of two components, one of which he procures from a single supplier at a unit price p. The manufacturer procures a total of M components from the supplier and uses them to manufacture M units of the good to be sold in the market. The product generates a unit revenue of r for the manufacturer. We denote the manufacturer's unit production cost by [u.sub.m], and the supplier's by [u.sub.s]. At the time of contracting, both the manufacturer and the supplier know M....
|
