On the Advantage of Quantity Leadership When Outsourcing Production to a Competitive Contract Manufacturer Yulan Wang Institute of Textiles and Clothing, Hong Kong Polytechnic University, Hong Kong, yulan.wang@inet.polyu.edu.hk Baozhuang Niu Lingnan (University) College, Sun Yat-sen University, Guangzhou, Guangdong, China, niubz@lingnan.net Pengfei Guo Department of Logistics and Maritime Studies, Hong Kong Polytechnic University, Hong Kong, lgtpguo@polyu.edu.hk T his study investigates a supply chain comprising an original equipment manufacturer (OEM) and a contract manufac- turer (CM), in which the CM acts as both upstream partner and downstream competitor to the OEM. The two parties can engage in one of three Cournot competition games: a simultaneous game, a sequential game with the OEM as the Stackelberg leader, and a sequential game with the CM as the Stackelberg leader. On the basis of these three basic games, this study investigates the two parties��� Stackelberg leadership/followership decisions. When the outsourcing quantity and wholesale price are exogenously given, either party may prefer Stackelberg leadership or followership. For example, when the wholesale price or the proportion of production outsourced to the CM is lower than a threshold value, both parties prefer Stackelberg leadership and, consequently, play a simultaneous game in the consumer market. When the outsourc- ing quantity and wholesale price are decision variables, the competitive CM sets a wholesale price sufficiently low to allow both parties to coexist in the market, and the OEM outsources its entire production to this CM. This study also examines the impact of the supply chain parties��� bargaining power on contract outcomes by considering a wholesale price that is determined via the generalized Nash bargaining scheme, finding a Stackelberg equilibrium to be sustained when the CM���s degree of bargaining power is great and the non-competitive CM���s wholesale price is high. Key words: outsourcing contract manufacturing competitive CM quantity leadership Cournot competition History: Received: April 2010 Accepted: November 2011 by M. Eric Johnson, after 3 revisions. 1. Introduction Outsourcing the manufacturing function to contract manufacturers (CMs) is common practice today for many original equipment manufacturers (OEMs). In the personal computer industry, for example, Apple and Hewlett-Packard outsource all of their assembly functions to Foxconn, Flextronics, and other CMs in Taiwan and mainland China (Smith 2008). Due to the intense competition among CMs, the services they provide now go beyond the pure manufacturing func- tion. In the electronics industry, increasing numbers of classic CMs (which have no design capabilities) are becoming original design manufacturers (ODMs) that offer value-added services in addition to product manufacturing. Foxconn and Flextronics, for instance, have built large R&D centers to offer product design services to OEMs (Baljko 2006), a welcome develop- ment that allows OEMs to shorten new product development lead-times and introduce greater prod- uct variety. However, allowing CMs to handle an increasing number of business functions, from innovation and design to production and even logistics, can prove a double-edged sword for OEMs, as the former are becoming increasingly capable of producing and selling their own self-branded products. A number of interesting business cases have been observed in which CMs act as both upstream partner and down- stream competitor to OEMs. For example, BenQ, Moto- rola���s CM, produced its first own-brand cellular phone in 2005 (Hilmola et al. 2005). Asustek, a Taiwan- based CM for Apple, Dell, Sony, and Toshiba, designs, produces, and sells its own Asus brand of notebook computers (Shilov 2007). Acer Inc., origi- nally a CM for IBM and Apple, actually became the third largest computer manufacturer in the world (by sales) in 2007 (Nystedt 2007). 1 Vol. 0, No. 0, xxxx���xxxx 2012, pp. 1���16 DOI 10.1111/j.1937-5956.2012.01336.x ISSN 1059-1478|EISSN 1937-5956|12|0|0001 �� 2012 Production and Operations Management Society

If a CM performs a single role, whether upstream partner or downstream competitor, then its relation- ship with the OEM is relatively simple. In the former case, the OEM decides the production quantity as a monopoly, and the CM is responsible only for manu- facturing. In the latter case, according to traditional oligopoly theory, the party that first decides the pro- duction quantity is able to capture a larger share of the market and obtain a higher profit, thereby exhibit- ing first-mover advantage (see, e.g., Vives 2001). However, a competitive CM is not only an OEM���s competitor, but also its business partner. Its revenue is generated both from producing and selling its own self-branded products and from contract manufactur- ing. The outcome of Cournot competition between an OEM and its competitive CM and the incentives of both in choosing quantity leadership/followership remain unclear, which provides the motivation for this study. In practice, it is common for an OEM to act as a Stackelberg leader in contracting with a competitive CM, although there are cases in which the latter assumes the leadership role. For example, at Compu- tex 2007 in Taipei, Asustek announced its production of a low-cost sub-notebook based on Intel���s Classmate PC reference design. It also reported a sales target of 200,000 units by the end of 2007 and of between three and five million by 2009 (Laptops 2007, Vilches 2007). One year later, one of its OEMs, Dell, entered the same market with a target production of more than 3.6 million (Dannen 2008). This study considers a setting in which the end market includes an OEM and a competitive CM. The OEM outsources part of its production to the competi- tive CM and the remainder to non-competitive CMs. All of the CMs, whether competitive or non-competi- tive, are capable of both design and manufacture.1 There is no intellectual property (IP) conflict between the OEM and competitive CM���s products in this study. Further, the products offered by the competi- tive CM and the OEM are imperfectly substitutable that is, the OEM���s products can be fully substituted for those of the CM, but the reverse does not hold true. There exist three basic Cournot (quantity) com- petition games between these two parties: a simulta- neous game, a sequential-move Stackelberg game with the OEM as the leader, and a sequential-move Stackelberg game with the competitive CM as the leader. To explore the endogenous quantity leader- ship issue, this study adopts the extended two-stage game in Hamilton and Slutsky (1990). In the first stage, the two players simultaneously choose a lead- ership or followership role. In the second stage, they play a simultaneous game if both players choose lead- ership or followership in the first stage, and a sequen- tial game otherwise. To provide a full picture of the outcomes of the three games, the study first considers a scenario in which the wholesale price and the proportion out- sourced from the OEM to the competitive CM are exogenously given. This scenario is realistic in certain settings. For example, intense price wars among CMs can result in price alliances or associations among them and lead to an industry standard price, which can be deemed as given. At the same time, an OEM may have multiple reasons to outsource production to several CMs, one of the most important of which is to avoid supply risks (see Tomlin 2006, for details of supply chain disruptions). The proportion of produc- tion the OEM outsources in these settings can be deemed exogenous. Both first- and second-mover advantages may exist for the OEM and the competi- tive CM. The advantage of quantity leadership depends on multiple factors, such as the market size, the wholesale price, the product substitution rates, and the percentage of production that the OEM out- sources to this competitive CM. When the wholesale price or the proportion of production outsourced to the CM is lower than a threshold value, both parties prefer Stackelberg leadership and, consequently, play a simultaneous game in the consumer market. As the degree of homogeneity between the OEM and com- petitive CM���s products increases, the more difficult it becomes to keep the CM as the follower and the more likely it is that a simultaneous game appears. The second scenario this study considers is one in which both the wholesale price and the proportion of production outsourced to the competitive CM are endogenized. In this scenario, the OEM determines the proportion of production that it outsources to the competitive CM, whereas the competitive CM endog- enously determines the wholesale price. Interestingly, the OEM is found to prefer outsourc- ing entirely to the competitive CM as long as its wholesale price is no more than that of non-competi- tive CMs. Further, when the competitive CM sets the wholesale price, it always sets it sufficiently low to allow both parties to coexist in the market. This finding implies that a rational competitive CM will not readily give up its contract manufacturing business and that a rational OEM will be cautious about employing the outsourcing quantity as a weapon against a competitive CM. A win���win solu- tion for both may be to allow coexistence in the market. Otherwise, the loss of orders from OEMs may actually spur CMs to develop and sell their own- brand products, thereby turning them into aggressive competitors. Many CMs in Taiwan and the Pearl River Delta region of China were reportedly forced to create their own brands to compensate for lost OEM orders following the global financial crisis that began in September 2008 (Liu 2009). At the same time, Wang, Niu, and Guo: On the Advantage of Quantity Leadership 2 Production and Operations Management 0(0), pp. 1���16, �� 2012 Production and Operations Management Society

however, there are many examples of OEMs and com- petitive CMs coexisting in harmony. Some competi- tive CMs even put considerable effort into retaining long-term relationships with OEMs, such as by divid- ing themselves into two companies, with one respon- sible for self-branded business and the other for contract manufacturing business (Chung 2004). Other CMs, such as Arima, Clevo, Elite, TPV Technology, and Twinhead, choose to maintain their self-branded and ODM businesses within the same organization. In the case of any conflict, they place priority on the latter, satisfying outsourced orders first by reducing the output of their own branded products (Wang 2008, Yang 2006). As a result, many OEMs choose to retain long-term relationships with competitive CMs rather than terminate their business with them, espe- cially when those CMs have accumulated special expertise, such as trained workers and good produc- tion control systems and policies. The final scenario considered here is that the CM and the OEM negotiate the wholesale price via a gen- eralized Nash bargaining (GNB) scheme. Paradoxi- cally, a weak CM is found to behave aggressively in the end-product market and, consequently, to play a simultaneous game with the OEM, whereas a powerful CM is rather cooperative, prompting a sequential-move game. The explanation is that greater bargaining power allows the CM to obtain a larger revenue share from contract manufacturing, which weakens its incentives to sell its own-brand products. The remainder of this paper is organized as follows. Section 2 reviews the related literature. Section 3 pre- sents the model notations and assumptions. Section 4 analyzes the way in which the OEM���s outsourcing decision affects the production quantity and leader- ship preference of both itself and the competitive CM, and carries out sensitivity analyses of the product substitutability parameter. Section 5 extends the discussion to a setting with endogenized quantity and wholesale price, and section 6 investigates the endogenous wholesale price via a GNB scheme. Sec- tion 7 concludes the article. All of the proofs are rele- gated to Online Appendix A. 2. Literature Review The issue of subcontracting to a rival/potential entrant has been discussed in the economics litera- ture. Spiegel (1993), for example, shows that if the transfer payment can be shared via Nash bargaining, then outsourcing production to a potential rival always renders both the incumbent and the potential rival better off, meaning that the latter has fewer incentives to build up its self-branded business. However, the issue of outsourcing to a competitive CM is relatively new to the operations management literature. Arrunada �� and Vazquez �� (2006) provide a number of business cases of competition between an OEM and a competitive CM. Horng and Chen (2007) empirically examine why some Taiwanese CMs have shifted toward own-brand management, and Arya et al. (2007) investigate a Cournot competition model between a retailer and its supplier. In an encroach- ment setting, they assume that the supplier has the right to set the wholesale price and that the retailer maximizes its profit by choosing the retail quantity. In a non-encroachment setting, they assume the whole- sale price to be exogenously given. By comparing encroachment and non-encroachment settings, they demonstrate that supplier encroachment can achieve Pareto improvement by inducing lower wholesale prices and increasing downstream competition. Ozkan and Wu (2009a) explore the market entry tim- ing problem from the perspective of a competitive CM by adopting a product life-cycle model, and they further consider such a CM���s capacity allocation issue (Ozkan and Wu 2009b). Lim and Tan (2010) investi- gate an OEM���s make, buy, and make-and-buy deci- sions by considering its interactions with its supplier (a CM in our context) over two periods. They show that the OEM���s high degree of brand equity can prevent the potential market entry of its CM. Chen et al. (2010) examine the OEM���s component sourcing decision in the face of a competitive CM, that is, the decision concerning whether to buy and resell compo- nents or delegate the procurement function to the competitive CM. This study, in contrast, investigates how the OEM���s outsourcing decisions affect the Stack- elberg leadership/followership preferences of both itself and a competitive CM, and also considers the endogenous wholesale price and outsourcing proportion decisions. This work is closely related to the study of firms��� outsourcing decisions. Elmaghraby (2000) presents a survey of the operational issues related to outsourc- ing, and Cachon and Harker (2002) consider two com- petitive firms facing economies of scale. McGovern and Quelch (2005) summarize the reasons that firms engage in outsourcing, and discuss what should be outsourced and the responsibility borne by marketing managers. Ulku �� �� et al. (2007) investigate whether the OEM/CM should bear the inventory/capacity risk. Arya et al. (2008a) are concerned with a firm���s make- or-buy decision, in which the firm can either produce inputs internally or outsource them to a monopoly supplier. Gray et al. (2009a) explore the impact of cost-reduction ability and the OEM���s outsourcing decision in a two-period game setting. They (Gray et al. 2009b) further test the OEM���s outsourcing pro- pensity by jointly considering cost and quality issues. Kaya and Ozer �� (2009) discuss the quality risks of out- sourcing, and Feng and Lu (2009) characterize OEMs��� Wang, Niu, and Guo: On the Advantage of Quantity Leadership Production and Operations Management 0(0), pp. 1���16, �� 2012 Production and Operations Management Society 3