Typology and Effects of Co-opetition in Buyer-Supplier Relationships: Evidence from the Chinese Home Appliance Industry. 供应链关系中竞合分类及其影响：来自中国家电业的实证研究

Abstract

Building on the co-opetition perspective, this study takes a unified approach to examine relationship benefits and transaction costs in buyer–supplier relationships. We argue that buyer–supplier vertical dyads fall into one of four distinct co-opetition types based on their respective degrees of cooperation and competition. Each co-opetition situation then corresponds to unique levels of relationship benefits and transaction costs. A sample of 225 supplier–buyer dyads from the Chinese home appliance industry confirms that competition has a positive effect on transaction costs and a negative effect on relationship benefits, while cooperation stimulates greater relationship benefits but affects transaction costs mainly at low competition levels. The results advance prior research on buyer–supplier relationships and co-opetition and offer valuable avenues for future research. 基于竞合的观点，本文采用了一个整合的模型对供应链关系中的关系收益和交易成本进行了研究。本文认为，依据双方合作和竞争程度的不同，买卖双方间的交易关系可以划分为四种不同的竞合关系类型，且每种竞合类型分别对应不同的关系收益和交易成本水平。利用来自中国家电行业225对买卖双方的配对样本，本文发现，买卖双方间的竞争对交易成本有正向作用，对关系收益有负面影响；而双方间的合作会促进关系收益的提高，但只在竞争水平较低时，合作才会对交易成本产生影响。本文研究发展了以往关于买卖双方关系和竞合的相关文献，并指明了有价值的未来研究发展方向。 Securing partnership benefits while reducing governance transaction costs is a persistent challenge in the management of buyer–supplier relationships (Cannon & Homburg, 2001; Dahlstrom & Nygaard, 1999; Kotabe, Martin, & Domoto, 2003). On one hand, firms seek to form vertical relationships in pursuit of economic rewards, knowledge transfer, and efficiency (Geyskens, Steenkamp, & Kumar, 1999; Jap, 1999; Terpend, Tyler, Krause, & Handfield, 2008). Achieving such benefits is enhanced by factors that stimulate information and knowledge exchange, such as relationship satisfaction (Gassenheimer & Ramsey, 1994), justice perceptions (Luo, 2007a), and relationship duration (Heide & Stump, 1995). On the other hand, vertical relationship partners are subject to multiple transaction costs associated with forming, monitoring, and evaluating partnerships (Buvik & Reve, 2001; Cannon & Homburg, 2001; Dahlstrom & Nygaard, 1999). Problems stem from behavioural or environmental uncertainties, as well as from the potential for building non-reciprocated relationship-specific capital, introducing dangerous lock-in or value appropriation by one partner at the expense of the other (Williamson, 1985). Meanwhile, in an emerging line of research, scholars recognize that the broader challenge of managing both the generation of benefits and the reduction of costs in business relationships is often holistically captured by co-opetition – simultaneous cooperation and competition (Bengtsson & Kock, 2000; Chien, 2005; Gnyawali, He, & Madhavan, 2006; Gnyawali & Park, 2011; Luo, 2005, 2007b). The co-opetition framework has been increasingly applied to describe business relationships in a number of different contexts, such as among the subsidiaries of multinational enterprises (Luo, 2005), within business networks (Chien, 2005; Gnyawali & Madhavan, 2001; Håkansson & Ford, 2002), in global competition (Govindarajan & Gupta, 2001; Luo, 2007b), in strategic alliances (Afuah, 2000; Dyer & Singh, 1998; Khanna, Gulati, & Nohria, 1998), and in multifaceted relationships incorporating buyer–supplier transactions (Dowling, Roering, Carlin, & Wisnieski, 1996; Garcia & Velasco, 2002). In all cases, co-opetition emphasizes parties' interdependence in business relationships and corresponding syncretic rent-seeking, dynamically balanced behaviours (Lado, Boyd, & Hanlon, 1997) by uniting two seemingly incompatible phenomena. Whereas cooperation promotes the mutual pursuit of shared value-adding activities, competition emphasizes firms' independent actions to improve their individual performance, often at the expense of the firms with which they are cooperating (Luo, 2004). We connect those two separate lines of research to examine how the relative levels of cooperation and competition affect relationship benefits and transaction costs in buyer (distributor)–supplier (manufacturer) relationships. To date, buyer–supplier relationships are seen as predominantly cooperative entities mainly in the areas of marketing and supply chain management (e.g., Anderson & Narus, 1990; Terpend et al., 2008). In their review of two decades of research on buyer–supplier relationships, Terpend et al. (2008) concluded that firms in such settings seek to derive value mainly through cooperative means. That research has mainly focused on investigating cooperative mechanism effects for improving relationship benefits and reducing transaction costs (Dahlstrom & Nygaard, 1999; Heide & Miner, 1992; Jap, 1999). However, that line of research has overlooked competition as another dimension for analyzing and managing transaction costs and relationship benefits in buyer–supplier dyads. Broadly, competition between buyers and suppliers in a dyad occurs over the makeup and distribution of the total gains from serving the same end markets and concerns their individually optimal approaches to serving particular market segments. The sociological view of competition sees it as a contest between entities over resources, position, prestige, awards, or status (e.g., Andersen & Taylor, 2012). The economic view sees competition more narrowly as the market-based efforts of independent sellers to secure the business of prospective buyers by offering the most favourable terms (e.g., Kasper, 2008; Kirzner, 1997). In business, the concept of competition is used more broadly to include economic competition, as well as sociological forms of competition that can occur within organizations (Birkinshaw, 2001; Luo, 2005) or in business networks (Bengtsson & Kock, 2000; Gnyawali & Madhavan, 2001). Our perspective is more closely aligned with this broader view of competition. We are concerned with the contest between buyers and suppliers in a dyad to directly increase their individual gains by undercutting gains for the other side. To account for both cooperation and competition in buyer–supplier relationships, we adopt the co-opetition typology (Luo, 2007b), where adapting corresponds to high competition–high cooperation situations; isolating corresponds to low competition–low cooperation situations; contending corresponds to high competition–low cooperation situations; and partnering corresponds to low competition–high cooperation situations. We argue that buyer–supplier dyads are clustered within these four types of co-opetition. We then examine the effects of each type on relationship benefits and transaction costs in a sample of 225 supplier–buyer dyads from the Chinese home appliance industry. By empirically validating Luo's (2007b) co-opetition typology, we make three main contributions to the literature. First, we offer a co-opetitive perspective to conceptualize buyer–supplier relationships. While prior research emphasizes the role of cooperation in this context, we offer an analytical platform permitting examination of dyadic behaviours along two dimensions – cooperation and competition. Second, we expand the range of possible contexts where co-opetition may apply. In addition to the common context of horizontal relationships, we apply co-opetition to vertical relationships. Third, the insights we derive from our approach allow us to enhance current knowledge about the relational perspective of business transactions and the related concepts of trust, reciprocity, and guanxi (Chen, Chen, & Huang, 2013; Luo, Huang, & Wang, 2012). Buyers and suppliers engage in co-opetitive relationships when they simultaneously cooperate to create greater common value from selling manufacturers' products through distributors' networks while competing to extract more of the total gains from the sales. The co-opetitive perspective recognizes that buyers and suppliers can simultaneously pursue individual and common goals. Generally, organizations cooperate when they interact on the basis of a common interest structure (Dagnino & Padula, 2002), seeking mutual benefits by exploiting complementary resources (Anderson & Narus, 1990). Organizations compete when companies interact on the basis of conflicting interest structures, often exemplified by struggle to access a common pool of resources (Bengtsson & Kock, 2000). The literature has well-documented the cooperative aspect of buyer–supplier relationships. Buyers and suppliers create vertical partnerships to exploit complementary resources corresponding to their respective upstream and downstream activities (Terpend et al., 2008). Buyers and suppliers enhance the cooperative aspect of their relationship through greater trust, improved communication, mutual experience, and shared problem solving (Anderson & Narus, 1990; Heide & Miner, 1992). Such cooperative efforts may increase common gains (Cannon & Homburg, 2001; Heide & Stump, 1995) and suppress differences and friction (Dahlstrom & Nygaard, 1999; Heide & Miner, 1992). However, the competitive aspect of buyer–supplier relationships is less known. Distributors and manufacturers may continuously compete over the generation and division of gains from selling the manufacturer's products in the distributor's network. For the distributor, the gains depend on the total amount of sales revenues, the portion of the sales received, the fees collected from the manufacturer, and the individual cost of carrying and selling the manufacturer's products. For the manufacturer, the gains depend on the total amount of sales revenues, the portion of the sales received, ther individual cost of selling products in the distributor's network, and all costs associated with manufacturing the products. Since the total gains are a function of the total sales revenues and the relationship-specific costs related to selling the products, competition may take place over each element. The fundamental source of such competition is the difference in the parties' individually optimal strategies to serve the same end-market, including how the strategies fit with the other aspects of their business, such as the distributor's overall marketing and growth strategies, or the manufacturer's own stores or brand image. Although buyers and suppliers may try to settle their strategic differences with contractual provisions, few aspects of their relationship are fixed formally because of the complexity and specificity of each transaction. Such relative contract incompleteness may also provide a weaker institutional platform guiding the cooperation, restraining opportunities for trust and reciprocity to nurture continuance and long-term payoffs (Luo, 2002). Most commonly, manufacturers and distributors in the Chinese context agree on revenue distributions from single items, such as fixed final sales prices and fixed proportions. They agree on explicit costs for manufacturers to sell their products in the distributor's stores, such as annual allowances based on store position for the merchandise, shelving fees for booth/counter maintenance, fee for managing manufacturers' salespeople, and sponsor/promotion fees related to assistance in setting up new stores or during promotional events and campaigns. However, the agreements typically omit clauses about approaches to generating total sales revenue, are relatively broad in specific amounts, frequency, and types of fees, and are open with respect to other costs that can be created or modified such as payment time frames, inventory backlogs, after-sale services, or installation services. These circumstances give distributors and manufacturers numerous opportunities to manipulate their individual gains by influencing total sales revenue or transaction costs. Because firms are heterogeneous, they may have incongruent strategies to extract value and achieve growth. The first broad area of incongruence may occur over how buyers and suppliers plan to generate current sales and expand future sales. Depending on their strategic positioning, parties may have various low-cost (lower margin and higher volume) or differentiation (higher margin and lower volume) approaches to serving the market. Differences in approaches may generate buyer–supplier competition over sales revenues. For example, the manufacturer may follow a differentiation approach, while the distributor may have a low-cost approach, with each pursuing a different optimal strategy. For example, GOME, China's largest home appliance retailer, and Haier, China's largest white goods maker, have an agreement over the final sales price of Haier's products, but GOME often undertakes unilateral promotional campaigns to expand market share by lowering sales prices (Winshang, 2005). GOME, not the contract, determines the frequency and intensity of such campaigns. To obtain a greater share of the gains, GOME often demands the original share from the contractual final sales price although the goods might have sold for less, while Haier receives only the remaining portion of the actual sales price. However, because of its differentiation strategy, Haier prefers to obtain higher margins for its products to cover its higher costs. To show its displeasure at GOME's practice, Haier once temporarily withdrew its products from GOME's stores. On other, less extreme, occasions, Haier attempted to renegotiate aspects of its contract with GOME or lower some of its costs in the relationship. Strategic conflict also may arise from the extent of vertical integration and mutual dependence. If the manufacturer has its own stores, it may be particularly sensitive to the loss of customers to the distributor's market penetration or expansion moves. For example, after Gree, China's leading air-conditioning unit manufacturer, invested substantial resources in developing its own brand stores, it resisted GOME's price-cutting marketing initiatives (Guo, 2004). In addition, the relatively smaller Gree had to bear substantial costs from a smaller proportion from the sales price, after-sale service, and installation fee to sell its products in GOME's stores. After a period of complete secession, GOME and Gree restarted their relationship, but Gree imposed restrictions on GOME in terms of marketing, customer base, and overall costs within the relationship (Chen, 2007). Their prolonged competitive battle motivated other household electrical appliance manufacturers to build their own brand stores to reduce their dependence on large retailers and increase their gains from serving the same end market. Examples include TCL's ‘Happy Tree’ program, Konka's ‘1000 Stores in 1000 Counties’ program, and Midea's ‘One Hundred 4s Stores’ program. The second area of incongruence occurs over how buyers and suppliers manage and distribute the costs of selling the manufacturer's products in the distributor's chain of stores. Because contracts do not completely fix such costs, the parties can manipulate both the types and amounts of the costs. For example, distributors may add extra fees and manufacturers can refuse to bear costs or can renegotiate to eliminate or reduce some costs, as has happened often in the GOME, Haier, and Gree cases. In addition, even when a contract details the costs, vertical partners may see environmental changes as opportunities to transfer greater portions of costs to the other party. For example, in 2010, Master Kong, China's leading maker of instant noodles, tried to increase its prices in response to higher raw material costs (Xinhuanet, 2010). Carrefour, China's seventh largest chain store, was a Master Kong distributor, and it was facing higher operation costs. It demanded half of the additional price rather than the typical ten to fifteen percent proportion. Master Kong withdrew its products from Carrefour's stores until the two firms renewed their contractual agreement (Zhan, 2010). Overall, the competition between buyers and suppliers can take opportunistic forms when self-interest compels partners to break contractual rules and coercive forms when one partner pressures the other to conform to demands that benefit one but disadvantage the other. As the examples have shown, over time the two sides can mix the competition forms. With higher strategic incongruence, competitive behaviour is more likely, regardless of the form of competition. Within a co-opetitive framework, buyer–supplier relationships can experience various levels of co-opetitive intensity depending on the respective levels of cooperation and competition in a given dyad. For example, although intense competition characterizes the Haier–GOME and Gree–GOME dyads, high levels of cooperation also characterize the Haier–GOME dyad. We adopt the co-opetition typology (Luo, 2007b) – partnering, adapting, isolating, and contending – to describe four situations that can have different behavioural and outcome implications. The partnering co-opetition type corresponds to high cooperation–low competition situations. A typical example is the relationship between Procter & Gamble and Walmart. Since the late 1980s, Procter & Gamble and Walmart have assisted each other in various functions such as finance, inventory, and production. In a specific instance, after they launched their production and marketing alliance, Procter & Gamble's diaper inventory turnover in Walmart increased by 70 percent and the sales by 50 percent. For buyer–supplier relationships, high cooperation means that both sides share common goals and interests, depend heavily on each other, and commit to the focal relationship (Heide & Miner, 1992), while low competition implies that they have little disagreement about the strategic approach to serving a particular end market. The adapting co-opetition type corresponds to high cooperation–high competition situations. For example, besides engaging in intense competition, since 2005, Haier and GOME have cooperated extensively in joint market expansion, bilateral information exchange, and combined sales promotion. As a result, Haier's colour TV sales through GOME's network have grown annually by a steady ten percent. Although the vertical partners have substantial incongruence in their individual approaches to serving the same end market, they continue to cooperate because of their mutual interdependence and relationship-specific investments. The isolating co-opetition type corresponds to low cooperation–low competition situations in which buyers and suppliers have unimportant exchange relationships with little or no interdependence. These may often be sporadic, such as the purchase of nonessential operation and office supplies. This situation characterizes a convenient, uninvolved, and transaction-based relationship (Wilkinson & Young, 1994). The contending co-opetition type corresponds to low cooperation–high competition situations implying interest divergence, low interdependence, and substantial conflict and opportunistic behaviour (Luo, 2005). Buyers and suppliers have incentives to compete with each other because they substantially differ in their plans to extract maximum market value. In this situation, buyers and suppliers are more likely to use rigid contracts or coercive power to control transactions (Wilkinson & Young, 1994), and may experience more strained relationships (Young & Wilkinson, 1998). The GOME and Gree relationship, as mentioned, was historically dynamic because they had substantial strategic incongruence and low interdependence, and neither side had significant relationship-specific investments. Buyer–supplier relationships are subject to both benefits and costs. While substantial research attention has been paid to relationship benefits in dyadic business relationships, relatively fewer attempts have been made to analyze their inherent transaction costs (e.g., Noordewier, John, & Nevin, 1990). Nevertheless, reducing costs is important to creating value and a critical factor in relationship development and stability (Cannon & Homburg, 2001). Co-opetitive relationships operating in the horizontal rival context have provided the mechanisms by which cooperation and competition affect relationship benefits and transaction costs. Prior developments in the area of co-opetition strongly suggest that cooperation and competition are mutually compounding factors. Simultaneous cooperation and competition exhibits syncretic rent-seeking behaviour benefitting both firms (Lado et al., 1997). Additionally, simultaneous competitive pressures and cooperative desires motivate global players to seek ‘the positive-sum, efficiency-enhancing effects of competition and cooperation’ (Luo, 2007b: 131). In buyer–supplier relationships, cooperation and competition are additive factors and act in opposite directions. Unlike horizontal cases where firms are competitors by default, in buyer–supplier contexts the firms are fundamentally partners. Consequently, their relationship is much more strongly influenced by cooperative notions such as trust, reciprocity, communication, and justice perceptions than are relationships between industry rivals. Buyer–supplier behaviours based on self-interest are counterproductive and even destructive. In cases of embeddedness, the narrow pursuit of immediate economic gains is incompatible with the enrichment of relationships through trust and reciprocity (Uzzi, 1996). Cooperation inspires win–win solutions that promote relationship continuity while win–lose solutions are zero-sum and act in the opposite direction – unwillingness to extend the relationship beyond the focal transaction (Gupta, 2011: 21). In essence, when a buyer–supplier relationship is based on trust, communication, and reciprocity, acting in self-interest may reverse cooperation effects. Nevertheless, in some cases the relationship approach is subject to lock-in and power coercion, preventing mutually beneficial adaptation to environmental contingencies (Gupta, 2011: 22). In such circumstances, a transactional approach may reduce lock-in. Although we do not emphasize the latter dynamics in our analysis, we assert that the principle of additivity remains valid even in those situations. Figure 1 illustrates the idea of additivity and opposite effects, where the levels of relationship benefits and transaction costs are presented as functions of cooperation and competition for each of the four co-opetitive situations. We assume that even in the isolating type where cooperation and competition are minimal, base levels of relationship benefits and transaction costs exist. Purely transactional relationships can be beneficial and preferred in many circumstances (Gupta, 2011), but parties must safeguard against possible transaction costs and opportunism (Buvik & Reve, 2001; Liu, Luo, & Liu, 2009; Poppo & Zenger, 2002). When cooperation increases, relationship benefits increase and transaction costs decrease compared to these base levels (the partnering type). In contrast, when competition increases, relationship benefits decrease and transaction costs increase compared to the base levels (the contending type). Following the principles of additivity and opposite influences, when both cooperation and competition are high (the adapting type), the levels of relationship benefits will be higher than the base levels due to higher cooperation but lower than in the partnering type due to higher competition. Similarly, transaction cost levels will be higher than the base levels due to higher competition but lower than in the contending type due to higher cooperation. In the following sections, we apply that logic to derive our two hypotheses. Relationship benefits and transaction costs in the co-opetition typology a Cylindrical shapes indicate relationship benefits; cubical shapes indicate transaction costs. b Shapes with only solid lines correspond to base levels; patterned shapes correspond to additions to base levels; and shapes with dashed lines are shallow corresponding to reductions in levels compared to other types. Based on the principles of additivity and opposite effects, each co-opetition type is associated with different relationship benefits. In the isolating type, buyers and suppliers achieve some increased economic performance pertaining to the complementary nature of their resources. However, the gains are likely to be limited because the interaction is sporadic and has low interdependence. Although the vertical partners may be satisfied with their arms-length transaction, their insufficient social contact and communication contribute little to relationship-specific satisfaction or knowledge-sharing opportunities (Hansen, 1999). Relationship benefits are likely to increase more substantially for the partnering type than for the other three types because buyers and suppliers willingly invest in the development of dyadic tangible and intangible transaction-specific assets in pursuit of their cooperative objectives (Morgan & Hunt, 1994). These assets serve as a bilateral bonding (Anderson & Weitz, 1992; Rokkan, Heide, & Wathne, 2003) to improve economic performance (Brown, Dev, & Lee, 2000; Jap, 1999) and act as an intermediary for partners' knowledge transfer and sharing. The benefits, along with frank and open communication, strengthen partner perceptions of fulfilment in relationship interactions (Mohr, Fisher, & Nevin, 1996). Relationship benefits are likely to weaken for the adapting type because of high competition levels. Attrition in a relationship inhibits buyer–supplier communication and understanding of the other party's needs, problems, and resources (Inkpen & Tsang, 2005), hinders active problem solving, and obstructs bilateral trust (Sanzo, Santos, Vazquez, & Alvarez, 2003). The parties lose their generally positive evaluations of their partners (Anderson & Narus, 1990) so that it becomes more difficult to agree on decisions related to economic performance (Jap, 1999), and to share knowledge. Nevertheless, the relationship benefits remain sizably higher than those in the isolating type. While firms in adapting dyads suffer some attrition and obstruction of their already high levels of relationship benefits, firms in isolating dyads cannot construct relationship-specific benefits because transactions are short-lived and generic. The benefits from relationship-specific investments in the adapting type, even when diminished with intensified competition, remain higher than the benefits achieved without such investments. Hypothesis 1: In buyer–supplier dyads, relationship benefits will be significantly higher in the partnering type than in the adapting type, significantly higher in the adapting type than in the isolating type, and significantly higher in the isolating type than in the contending type. Because cooperation and competition in buyer–supplier relationships have an additive property and act in opposite directions, transaction costs will be different across the four co-opetition types. In the isolating type, low frequency and volume of transaction allow buyers and suppliers to adopt a standard contract template and facilitate a relatively easy contract evaluation process for addressing potential problems (Williamson, 1979). The low levels of interdependence and asset specificity are associated with low termination costs and limited asset appropriation risks, reducing the need to build relational governance mechanisms (Rindfleisch & Heide, 1997; Wilkinson & Young, 1994; Williamson, 1985). Nevertheless, even buyers and suppliers in the isolating type are subject to some transaction costs and opportunism stemming from behavioural or environmental uncertainties (Buvik & Reve, 2001; Liu et al., 2009; Poppo & Zenger, 2002). Transaction costs are likely to increase substantially in the contending type, and be highest among all four types, because of substantial divergence in strategic approaches and interests, as well as low communication between buyers and suppliers. This situation burdens the process of reaching agreement on contract clauses, especially with regard to pricing and payment (Wilkinson & Young, 1994) and prevents accurate and timely information exchange about each side's fulfilment of contract obligations (Milgrom & Roberts, 1990). Furthermore, firms in such dyads turn to competition more than they do to cooperation, so the inevitable surge in conflicts and friction would discourage fulfilment of contract obligations (Wathne & Heide, 2000) and propagate the design and implementation of strict control mechanisms (Milgrom & Roberts, 1990). Moreover, the two parties may substitute long-term relationship orientation for immediate self-interest gains (Anderson & Weitz, 1992; Luo, 2006), permeating the exchange relationship with opportunism (Brown et al., 2000) that needs safeguarding. Higher cooperation is likely to diminish transaction costs in the adapting type compared with those in the contending type. As the firms in the dyad become more interdependent and increasingly pursue mutual benefits, they have less anxiety regarding the fulfilment of contractual obligations and feel freer to exchange information and allow transparency (Lee & Cavusgil, 2006). In addition, as buyers and suppliers try hard to establish reliable relational norms and trust, they can more efficiently address problems arising from unforeseeable changes (Luo, 2007c), opportunism (Wathne & Heide, 2000), or challenges in executing contract obligations (Poppo & Zenger, 2002). Although higher cooperation reduces transaction costs in the adapting type compared with those in the contending type, the cost levels are likely to be higher than those in the isolating type. While the isolating type has limited sources of transaction costs due to a dearth of relationship-specific assets and low interdependence, high competition in the adapting type raises substantial potential problems from behavioural and environmental uncertainties. The resulting transaction costs remain high, despite cooperation's mitigating influence. Hypothesis 2: In buye