The Strength of Churning Ties: A Dynamic Theory of Interorganizational Relationships* SERGIO G. LAZZARINI John M. Olin School of Business – Washington University 1 Brookings Drive, Campus Box 1133 St. Louis, MO 63130-4899 Phone: 314-935-4538; Fax: 935-6359 E-Mail: LAZZARINIS@olin.wustl.edu TODD R. ZENGER John M. Olin School of Business – Washington University 1 Brookings Drive, Campus Box 1133 St. Louis, MO 63130-4899 Phone: 314-935-6399; Fax: 935-6359 E-Mail: zenger@olin.wust.edu June 2002 * We thank the insightful comments by Lyda Bigelow, Chih-Mao Hsieh, Gary Miller, and especially Jackson Nickerson, who provided critical suggestions to improve our argument. We also benefited from comments by anonymous referees of the Business Policy and Strategy Division of the 2002 Academy of Management Conference. Remaining errors and omissions are our own. The Strength of Churning Ties: A Dynamic Theory of Interorganizational Relationships Abstract In many circumstances, firms seek to combine or balance the cooperation-based advantages of strong ties (partnerships) with the autonomy-based advantages of weak ties (arm’s-length exchanges) into a single relationship. However, we contend that such a combination is unstable, because tie strength is influenced by informal processes—socialization and learning—that are constantly shifting. Thus, managers cannot directly control the functionality of ties, but they can alter the trajectory of those informal processes through changes in formal mechanisms that periodically alter tie strength. Managers initiate weak ties and then strengthen them through formal mechanisms that promote cooperation, such as long-term contracts and joint ownership. Whenever ties become excessively strong and autonomy has been undermined, they are formally restructured or terminated. This generates ties that are neither permanently weak nor strong, but churning. We identify conditions that make churning ties an appropriate strategy to manage interorganizational relations, and also the costs and competencies of alternative formal mechanisms that managers can employ to reinforce and weaken ties. Keywords Interorganizational collaboration, strategic alliances, networks, cooperative strategy. The close buyer-supplier relations of Japan are often viewed as a benchmark in structuring interfirm alliances. Many argue that such partnerships promote mutual trust, knowledge sharing, and relationship-specific assets that enhance performance (e.g. Asanuma, 1989; Dyer & Nobeoka, 2000; Holmstrom & Roberts, 1998; Nishiguchi, 1994). Such partnerships, however, are far from a panacea. In 2001, Toyota, long viewed as possessing model types of supplier relationships, discovered it was paying significantly higher prices for parts inside its close network of suppliers than were available outside this network. Predictably, they encouraged their managers to “seek business ties outside the group” (Dawson, 2001: 43). Chrysler, which actively sought to replicate “Japanese-style” partnerships in the U.S. (Dyer, 1996), faced a similar problem in 2001, with supplier relationships “so cozy” that competitors were extracting better prices for similar parts (Ball, 2001: A8). Moreover, efforts by automakers to push component purchasing into electronically-managed markets suggests movement away from close ties, toward more arm’s-length supply relationships (Lucking-Reiley & Spulber, 2001). The above events reflect a fundamental tradeoff that managers face in crafting their interorganizational relations. While a long-term exchange promotes cooperation, which is often referred to in the literature as a “strong” tie, an arm’s-length exchange or “ weak” tie is easily severed and thus enables access to new relationships that may yield cost reduction or innovation. If we view tie strength as a choice, how then do firms select an appropriate tie strength for their interorganizational relations? Most authors adopt a simple contingency argument: the choice of tie strength depends on the attributes of the exchange or the industry in which it occurs. Thus, while exchanges involving relationship-specific assets require the support of strong ties, simple exchanges that do not require specialized assets can be managed through a series of weak ties, i.e., arm’s-length relations with multiple firms (e.g. Dyer, Cho, & Wujin, 1998; Williamson, 1985). However, very often managers seek both cooperation and autonomy in crafting a particular exchange relation. For instance, an exchange involving relationship-specific investments within an industry with high technological uncertainty may benefit not only from a strong tie, but also 1 from a series of weak ties (e.g. Harrigan, 1988a; Rowley, Behrens, & Krackhardt, 2000). Weak ties provide incentives for exchange partners to remain on the cutting edge in terms of cost and innovation and provide the flexibility to easily sever ties when opportunities emerge. Strong ties, however, foster investments in relationship-specific assets, but in the process managers may socially obligate themselves to partners whose capabilities become obsolete as potential new partners with lower costs or better technologies emerge (Afuah, 2000). On the other hand, while weak ties maintain the managers’ flexibility to pursue new opportunities and acquire new knowledge, they discourage relationship-specific investments and other forms of cooperation. In many circumstances, managers seek to both promote cooperation in the form of relationshipspecific investments and at the same time maintain autonomy to access alternative partners. In other words, managers would like to maintain an intermediate level of tie strength that balances cooperation and autonomy. We posit, however, that managers cannot craft interorganizational ties that combine the autonomy of weak ties with the cooperation of strong ties in any stable manner. The basic problem is that informal processes influence tie strength and these informal processes are neither static, nor directly controllable by the managers, except indirectly by the rather blunt instruments associated with formal changes in the nature of the exchange. Thus, the degree of tie strength that develops over time has an uncertain component that can lead to greater or lesser tie strength than originally desired by managers. For instance, an exchange relationship, which begins as an arm’s-length tie, may evolve through repeated interaction into a strong tie. Over time idiosyncratic routines and social attachments will tend to emerge and eventually constrain flexibility to access new knowledge and new exchange partners. The tie then becomes excessive in strength, showing high cooperation, but low autonomy. An opposite path of change is also possible: a firm with a strong tie supporting a particular exchange may initiate a new tie with another exchange partner and in the process unravel its existing strong tie. This existing tie then becomes weak, showing high autonomy, but low cooperation. 2 We contend that a static theory of tie strength is misguided, because managers have no capacity to statically choose the informal processes that influence the performance of interorganizational relations. Rather, building upon Nickerson and Zenger (forthcoming), we argue that managers can only dynamically approach the desired tie strength by essentially vacillating between formal mechanisms that alter expectations of relationship continuity, thereby providing incentives to shape the direction by which tie strength increases or decreases. Namely, managers monitor the evolution of informal processes that alter tie strength and implement, ex post, formal changes to adjust strength to the desired level. Thus, when repeated interaction in a strong tie damages autonomy, managers terminate this long-term relationship or alter its formal structure in ways that reverse the trajectory of tie strengthening. If managers sever the tie, they simply build new relationships. While initially weak, the newly formed tie is strengthened over time with the aid of formal governance structures such as long-term contracts and joint ownership. However, such changes are costly: managers will resist adopting formal changes until the benefit of adjusting tie strength exceeds the costs of doing so. Since informal processes that influence tie strength are not directly controllable, it is likely that at some point in time the degree of misalignment between desired and actual strength will be sufficiently severe to afford the cost of changing formal governance. Thus, the achievement of superior interorganizational performance may require ties that are neither permanently strong nor weak, but churning: managers may need to promote periodic changes in formal mechanisms that constantly alter the path of informal processes influencing tie strength. We submit that such patterns of escalating and weakening ties, considered by some as dysfunctional (Dyer, Cho et al., 1998: 73), may actually be the only way in which managers can approach an “ideal” tie when the achievement of superior performance requires both cooperation and autonomy. We identify conditions that make churning ties an appropriate strategy to manage interorganizational relations, and also the costs and competencies of alternative formal mechanisms that managers can employ to reinforce and weaken ties. By providing a dynamic theory of how firms adjust the governance of ties as a response to changes in interorganizational 3 performance, we contribute to the growing body of research examining the performance implications of external relationships (e.g. Dyer & Singh, 1998; Gulati, Nohria, & Zaheer, 2000). The paper is structured as follows. In the next section, we review the literature relating tie strength to performance, and then discuss in detail the difficulty in managing ties when industry or exchange contingencies demand at the same time cooperation and autonomy. We next present the logic of churning ties and the formal instruments available to firms to weaken or reinforce ties. Since our main focus is on dyadic transactions, we then briefly discuss how a similar logic applies to multiple transactions in networks. Concluding remarks follow. TIE STRENGTH AND PERFORMANCE Granovetter (1973: 1361) originally defined tie strength as a “combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie.” This collection of complex social elements that characterize close relationships tends to be associated with patterns of interaction that deliver a long time horizon of expected exchange which, as the relationship unfolds, creates a long history of past exchange (e.g. Gulati, 1995a; Ring & Van de Ven, 1994; Zajac & Olsen, 1993). Thus, to facilitate exposition, we simplify the original definition of tie strength by considering a variable that mediates most processes supporting close relationships: the “commitment” of partners, i.e., the degree to which they expect to continue their relationship despite the existence of alternative partners (Blau, 1964; Cook & Emerson, 1978; Kollock, 1994; Seabright, Levinthal, & Fichman, 1992). We therefore define tie strength as the degree of commitment that supports an exchange relationship for the transfer of goods, services, or information.1 We view tie strength as a governance device, though we recognize that a strong tie may emerge for reasons other than the functional support of an exchange. Notice therefore that our unit of analysis is a dyadic 1 This definition also allows a direct application of the concept to interorganizational ties, since some elements present in Granovetter’s (1973) original definition, such as emotional intensity and intimacy, are more applicable to interpersonal relations. 4 exchange that is potentially (although not necessarily) recurring.2 Based on this definition, we next discuss the performance consequences of tie strength. Benefits of Strong Ties The primary benefit of a strong tie is that it provides the necessary commitment to promote cooperation in an exchange. On the one hand, commitment provides a “shadow of the future” to discourage opportunism, meaning that the long-term payoff from cooperation surpasses the short-term payoff from defection (Anderson & Weitz, 1992; Axelrod, 1984; Heide & Miner, 1992; Parkhe, 1993). Within this perspective, many authors note that one of the main advantages of commitment is that it encourages ex ante investments in relationship-specific assets (Dyer & Singh, 1998; Helper, 1991; Hennart, 1988; Madhok & Tallman, 1998; Williamson, 1985). Parties engaged in arm’s-length exchanges are less willing to invest in such assets since there is a risk, ex post, that a party will leave the relationship or attempt to renegotiate contractual terms in such a way that the other party will be disfavored. Relationship-specific assets include not only physical investments but, perhaps more importantly, human-specific assets or co-specialized knowledge generated from joint learning (Bureth, Wolff, & Zanfei, 1997; Foss, 1996; Holmstrom et al., 1998; Poppo & Zenger, 1998; Williamson, 1985). On the other hand, repeated interaction and familiarity also create a “shadow of the past” which promotes the emergence of relational governance based on social attachments, trust, and norms (Gulati, 1995a; Larson, 1992; Luhmann, 1979; Macneil, 1980; Ring et al., 1994; Zajac et al., 1993).3 This further reinforces the cooperation-based benefits of strong ties. Thus, strong social attachments reduce conflicts and promote cohesion (Krackhardt, 1992; Nelson, 1989). 2 This rules out one-shot transactions such as when a firm acquires standard machinery (Williamson, 1985: 72-73). In such cases, exchange is occasional and thus the likelihood that parties will be able to meet in the future is negligible. At first glance, project collaborations with finite duration may appear to fall into this category. However, most exchanges of this sort actually involve a series of project collaborations. Consider the movie industry: although the production of a movie has a finite horizon, parties oftentimes continue their relationship in subsequent projects (e.g. Faulkner, 1983). 3 To be sure, social attachments are, in their essence, between people rather than firms (Seabright et al., 1992). However, informal relationships between firms can arise due to “closely interwoven personal relations” (Inkpen & Currall, 1997: 312) between individuals representing each firm. Outcomes of repeated interaction between these individuals are likely to disseminate to other members of the partnering firms and generate collective expectations and norms (Zaheer, McEvily, & Perrone, 1998). 5 Relational norms such as reciprocity also curb firms’ willingness to defect and thus support mutual trust (Berg, Dickhaut, & McCabe, 1995; Bradach & Eccles, 1989; Dore, 1983). Empirical research has found, for instance, that personal interaction and norms enhance joint learning and knowledge sharing between firms (Dutton & Starbuck, 1979; Dyer et al., 2000; Kale, Singh, & Perlmutter, 2000; Liebeskind, Oliver, Zucker, & Brewer, 1996; Rulke, Zaheer, & Anderson, 2000). In addition, since repeated interaction implies increasing amounts of time and effort devoted to the relationship (Madhok et al., 1998), psychological commitments towards particular partners will tend to escalate, thus increasing firms’ willingness to continue their cooperative exchange (Ring et al., 1994; Seabright et al., 1992). Benefits of Weak Ties Weak ties are advantageous because they provide autonomy. Managers can flexibly shift to access new knowledge and valuable exchange opportunities with alternative firms (Granovetter, 1973). Indeed, the absence of a strong tie is what frees the manager to access knowledge, goods, or services through a wide range of weaker ties. The firm can flexibly exchange with a variety of firms which possess specialized knowledge and capabilities (Granovetter, 1982; Schilling & Steensma, 2001). Interfirm specialization creates diversity of knowledge and competencies in a network (Kogut, 2000; Liebeskind et al., 1996; Richardson, 1972) and enhances opportunities for innovation through knowledge spillovers among firms (Feldman & Audretsch, 1999). Since agents connected through weak ties are not committed, they can adjust or sever their existing ties to benefit from new external opportunities. Weak ties also circumvent the dysfunctional outcomes unleashed by the escalating commitment of strong ties. Continuous interaction with the same partners restricts firms’ exposure to new ideas and information (Adler & Kwon, 2002; Greif, 1997; Uzzi, 1996) and undermines the diversity of knowledge available in the alliance, thus diminishing its value (Dussauge, Garrette, & Mitchell, 2000; Hamel, 1991; Nakamura, Shaver, & Yeung, 1996). The same mechanisms that promote commitment also restrict firms’ capacity to access new knowledge. Thus, joint learning and knowledge sharing that occurs between two firms can lock 6 them into one particular field of knowledge and lock them out of external opportunities (Cohen & Levinthal, 1990; Leonard-Barton, 1995; Poppo et al., 1998; Young-Ybarra & Wiersema, 1999). Routines emerging from repeated interaction tend to create idiosyncratic language and procedures, which limit firms’ ability to interpret external sources of information and recognize the importance of those sources (Katz, 1982; Tushman & Katz, 1980; Zenger & Lawrence, 1989). Thus, even if strong ties promote knowledge sharing and co-specialization (Dyer et al., 2000; Kale et al., 2000), this becomes a liability if partners are trapped into the “wrong” knowledge (Afuah, 2000; Grabher, 1993; Kern, 1998). The strong social attachment that accompanies repeated interaction aggravates the problem of gaining access to new knowledge. Excessive socialization may undermine partners’ ability to find discrepant information and innovative solutions (Gruenfeld, Mannix, Williams, & Neale, 1996; Kern, 1998; Levinthal & March, 1993). Additionally, firms connected through strong ties may be unwilling to pursue better opportunities, if pursuing these opportunities threatens existing ties or violates existing social obligations (Gargiulo & Benassi, 2000; Halpern, 1994; Jeffries & Reed, 2000). Constrained by such obligations, partners will tend to favor the stability of their relationship over efficiency (Dore, 1983; Seabright et al., 1992; Yamagishi, Cook, & Watabe, 1998). For instance, Humphrey and Ashforth (2000: 719) document that U.S. automakers expressed “concerns that interpersonal relationships could lead buyers to award contracts to suppliers who had higher unit costs, lower quality and slower delivery times.” Contingency Arguments The simple recognition that strong ties support cooperation, while weak ties provide autonomy suggests a simple contingency argument in which tie strength is matched to exchange conditions. Thus, transaction cost economics posits that relationship-specific assets at a minimum require the support of long-term arrangements to reduce the hazards of opportunistic behavior. Absent the need for relationship-specific assets, the weaker ties of market-like exchanges may provide sufficient governance (Williamson, 1985). Social exchange theorists (Cook et al., 1978; Kollock, 1994; Yamagishi et al., 1998) and organizational economists 7 (Barzel, 1982; Holmstrom & Milgrom, 1994) highlight additionally the role of measurement difficulty in the choice of governance. More difficult measurement of exchange performance requires longer-term relations to discourage opportunism. Other scholars view the transfer of tacit, complex knowledge as an exchange requiring strong social ties and the transfer of codified knowledge as well suited to weak ties (Hansen, 1999; Hennart, 1988; Kogut, 1988; von Hippel, 1994). Still other theories propose a fit between tie strength and industry conditions. Drawing from the concepts of exploration, the search for new resources, and exploitation, the use of existing resources (Koza & Lewin, 1998; Levinthal et al., 1993; March, 1991), Rowley et al. (2000) propose a fit between tie strength and industry characteristics. They argue that the autonomybased benefits of weak ties are crucial to enhance a firm’s innovativeness in industries with high technological uncertainty, which demand exploration-oriented tasks, whereas the cooperationenhancing properties of strong ties deliver superior performance in stable industries, which call for exploitation (see also Harrigan, 1988a; Schilling et al., 2001). Some authors also emphasize the role of market-level uncertainty, such as volatile demand, favoring transactions with many firms to absorb temporary supply or demand shocks (Eccles, 1981; Jones, Hesterly, & Borgatti, 1997; Kranton & Minehart, 2000). Table 1 summarizes the fit or alignment between exchange/industry conditions and tie strength, according to the contingency arguments discussed above. <Table 1 around here> WHAT IF FIRMS WANT TO COMBINE THE ADVANTAGES OF BOTH WEAK AND STRONG TIES? While these contingency arguments adequately describe the relationship between tie strength and various exchange or industry conditions, two factors plague managers’ choice of tie strength. First, managers commonly face a set of conditions that mandate a conflicting pattern of tie 8 strength. Second, managers cannot directly choose or control informal processes that influence tie strength. We consider these issues next. Conflicting Contingencies In many settings, managers want the benefits of both weak and strong ties. Consider the case of crafting R&D alliances in uncertain, high technology environments. In such settings, firms require continuous access to novel knowledge (Hagedoorn & Schakenraad, 1994; Powell, Koput, & Smith-Doerr, 1996). Such continuous exploration calls for the autonomy of weak ties. At the same time, creating effective R&D alliances often requires significant investments in relationship-specific human assets (Oxley, 1997; Pisano, 1989) and transfer of tacit knowledge (Hennart, 1988; Kogut, 1988). Such conditions require the support of strong ties (Table 1). Thus, in this context, firms will want both cooperation and autonomy in a single relationship; they will want a tie to be strong enough to create commitment while at the same time weak enough to enable access to a range of alternative firms. Recognition of this tension between weak and strong ties pervades the literature. Blau (1964) points out that while social attachments increase cooperation, they restrict individuals’ mobility to pursue opportunities with alternative partners. Resource-dependence theory posits that organizations desire autonomy to allocate their resources, even though they are commonly faced with commitments to existing relationships (Galaskiewicz, 1985; Oliver, 1991; Pfeffer & Salancik, 1978). This dilemma is also present in the literature on partnerships and strategic alliances. While scholars generally view alliance longevity as a proxy for performance (e.g. Geringer & Hebert, 1991; Park & Russo, 1996), they also recognize that such longevity reduces partners’ flexibility to learn and benefit from other firms (Afuah, 2000; Hamel, 1991; Lambe, Spekman, & Hunt, 2000; Parkhe, 1991; Singh & Mitchell, 1996; Young-Ybarra et al., 1999). Eccles and Crane (1988: 55) in their study of investment banking find that banks and their customers often “want the advantages of [long-term] relationships and the advantages of a more transactional [arm’s-length] orientation” (emphasis in the original). Finally, Doz and Hamel 9 (1998: 21) conclude that while committed relationships facilitate cooperation, “companies must be free to pursue better opportunities when they appear.” The tension between weak and strong ties exists whenever an exchange is subject to conflicting contingencies. Consider, for instance, a buyer-supplier relationship that involves high asset specificity, but is also subject to disruptive technological change that may cause the competencies of incumbent suppliers to become obsolete.4 Or, consider a service relationship that requires both a high degree of customization and enough flexibility to explore opportunities brought by alternative service providers.5 In the presence of such conflicting contingencies, managers will seek an intermediate level of tie strength that balances the cooperation-based advantages of strong ties with the autonomy-based advantages of weak ties. We argue below, however, that such a combination is unstable: it is not possible to structure relationships that permanently combine cooperation and autonomy. Unfortunately, firms have no capacity to structure a permanent relationship that is at once flexible, thereby easy to exit and providing easy access to new information, and also sufficiently committed, thereby supporting cooperation. The Dynamics of Tie Strength The fundamental difficulty facing the firm is that tie strength is largely influenced by informal processes—processes that managers can only indirectly control. Indeed, tie strength is constantly changing as personnel within the two firms both jointly learn and socialize. Through repeated interaction personnel within the firms develop idiosyncratic routines and co-specialized knowledge, as well as relational governance due to emerging social norms and attachments. 4 Analyzing the microprocessor industry, Afuah (2000: 389) asserts that in the advent of technological change “staying with the old supplier means that the manufacturer can build on existing close relationships but must grapple with the problems that the supplier faces in making the transition to the new technology. If the change is radical enough to suppliers, they many not be able to supply components with the type of quality that the firm needs to be competitive with the new technology… [However] switching suppliers means having to build new relationships and, in doing this, a firm may be handicapped by the organizational procedures and routines that it developed in its old relationships.” 5 In their study of investment banking, Eccles and Crane (1988) find that both customers and investment banks would like to develop strong ties to support cooperation. While corporate customers would like banks to offer services tailored to their needs, investment banks would like to increase the amount of information sharing in their relationship. However, by establishing a strong tie, banks and their customers necessarily reduce their autonomy to make better deals with other firms. Financial transactions require continuous exploration of arbitrage opportunities, which is properly carried out through weak ties (Zaheer & Zaheer, 1997). 10 These informal processes cannot be directly controlled or contracted for ex ante for two main reasons. First, they involve complex, intertwined social cognitive mechanisms: not only is learning a process resulting from collective efforts, but also the accompanying social connections influence individuals’ choice and solution of problems (Argote, 1999; Arrow, 1974; Levinthal et al., 1993). Second, the outcome of such processes is affected by stochastic factors that boundedly rational managers cannot perfectly foresee before firms interact in a given period. For instance, joint learning and socialization will be reinforced over time if for some reason individuals within the firms become motivated to proceed with an ongoing project, but they will be undermined if changes in industry conditions or firm-level strategies shift individuals’ attention to activities other than those involved in the alliance. Therefore, complexity and unforeseen changes will make it difficult for managers to make, ex ante, appropriate choices to avoid events that may alter the path of joint learning and socialization. Due to the difficulty in adjusting informal processes that influence tie strength, ties that initially exhibit intermediate levels of strength tend to be altered in a self-reinforcing manner. Over time, ties tend to become either very strong—with high cooperation but low autonomy—or very weak—with high autonomy but low cooperation. Consider the first possibility: ties becoming increasingly strong. Suppose that individuals within partnering firms jointly learn, at the outset, how to develop certain processes or technologies that are judged to be promising. Since joint learning requires social interaction, it delivers not only co-specialized knowledge, but also shared norms, values, and procedures (Kogut & Zander, 1996; Leonard-Barton, 1995). Thus, when knowledge is jointly generated, it tends to be socially embedded and supported by idiosyncratic routines (Argote, 1999; Wegner, Erber, & Raymond, 1991). As Brown and Duguid (1998: 100) remark, it is “socially embedded knowledge that ‘sticks,’ because it is deeply rooted in practice.” This makes it difficult for outsiders to interpret that collective knowledge, as well as for insiders to interpret external knowledge embedded in other social contexts (Afuah, 2000; Cohen et al., 1990; Levine & Moreland, 1991; Tushman et al., 1980; Zenger et al., 1989). In addition, individuals within firms 11 may become less receptive to new perspectives and more willing to deal only with people showing similar interests and views (Gargiulo et al., 2000; Hansen, 1999; Katz, 1982; Nelson, 1989). Consequently, individuals will tend to downplay opportunities with new firms and favor firms with which they have developed relationships. The resulting increase in commitment will tend to further solidify social attachments and deepen co-specialization (Bureth et al., 1997). The relationship may therefore reach a status of excessive tie strength associated with a pattern of functionality involving high cooperation, but low autonomy.6 An opposite path of change is also possible: ties can weaken over time. The processes generating such change are in theory symmetrical to the processes described above, though in practice tie strength may be more quickly unraveled than enhanced. If firms are to some extent autonomous, they will be able to use alternative firms for a particular exchange or pursue new types of activities that may emerge as a result of external events such as changing industry conditions. Individuals within the firms may therefore devote diverging effort towards new projects instead of projects where firms’ interests align. The resulting decrease in interfirm interaction reduces co-specialization and social attachments which, in turn, implies that it will be increasingly less costly to exit the relationship. But then firms will become increasingly reluctant to continue sharing knowledge and co-specializing assets, as they perceive an increased risk that their partners will pursue opportunities with alternative firms. Consequently, the tie will tend to become weak, showing high autonomy but low cooperation.7 This process of tie strengthening is consistent with what Williamson (1985) calls the “fundamental transformation.” According to Williamson (1985: 62) the initiation of a relationship triggers a process where “additional transaction-specific savings can accrue at the interface between supplier and buyer as contracts are successively adapted to unfolding events and as periodic contract renewal agreements are reached. Familiarity here permits communication economies to be realized: Specialized language develops as experience accumulates and nuances are signaled and received in a sensitive way. Both institutional and personal trust relations evolve. Thus the individuals who are responsible for adapting the interfaces have a personal as well as an organizational stake in what transpires.” Notice that Williamson (1985: 62) does not restrict the occurrence of this phenomenon to exchanges requiring specific investments at the outset. He also considers exchanges where asset specificity deepens as a by-product of repeated interaction, due to socialization and joint learning. 7 Eccles and Crane (1988) find this process of tie weakening in their study of investment banking. They note that “… when the customer can choose from among a group of investment banks, a profound change takes place in the expectations each has about the future… Because a relationship is based on both transactions and the expectation of future transactions, uncertainty about the future diminishes the quality of the relationship… The customer’s unwillingness to share information with the investment bank beyond what is necessary to do a particular deal further 6 12 Figure 1 illustrates the changes described above. In the presence of conflicting contingencies, firms’ preferences increase Northeast: they would like to reach a position with high levels of cooperation and autonomy, such as point X. The permanent use of either a strong tie or a series of weak ties (i.e., points S or W) is sub-optimal for firms that would like to procure, for instance, customized products in industries with uncertain technological paths. However, due to the forces described above, any tie close to point X will gradually become either strong, showing high cooperation but low autonomy (point S), or weak, showing high autonomy but low cooperation—i.e., the relationship will move towards either point S or W. Which outcome will be eventually “selected” is uncertain at the outset, since there will be forces pulling the relationship towards either extreme.8 Thus, the process of tie change will be path dependent (Arthur, 1989): a host of stochastic factors outside our model will eventually decide to which particular point the relationship will tend to converge. In sum, while managers can in theory identify an ideal tie strength for a given exchange relationship, such tie strength cannot be maintained in the presence of conflicting contingencies. The fundamental question is therefore how to increase relationship performance under those conditions, since partners will wish an intermediate level of tie strength that balances cooperation and autonomy and such combination is unstable. <Figure 1 around here> The arguments above suggest that firms cannot cope with the weak vs. strong tie dilemma statically. Rather than pursuing stable but sub-optimal outcomes—i.e., either a strong or a weak tie permanently—we contend that firms can do better in the presence of conflicting contingencies by promoting dynamic adjustments in tie strength in order to reap temporarily the benefits from autonomy and cooperation. diminishes the quality of the relationship. When interaction between deals decreases, the relative significance of transactions decreases, further reinforcing the perception of an increased transactional [arms’s-length] orientation on the part of the customer” (Eccles et al., 1988: 58, emphasis in the original). 8 For instance, Doz (1996) discusses the role of stochastic “initial conditions” that influence future alliance outcomes. 13 CHURNING TIES: DYNAMICALLY ADJUSTING TIE STRENGTH The preceding discussion provides us with two assumptions about the nature of interorganizational relationships. First, there is a wide range of circumstances under which both cooperation and autonomy in an exchange relationship are desired. Second, intermediate levels of tie strength that balance cooperation and autonomy are difficult to sustain, because informal processes that influence tie strength are dynamically changing. But although it is not possible to directly control such informal processes directly, managers can indirectly influence their trajectory through periodic, marked changes in formal governance mechanisms. Changes in formal governance essentially alter expectations of relationship continuity, thereby providing incentives to change the direction by which tie strength increases or decreases. For instance, when an exchange relationship is weak, managers can lengthen the terms of a contract or develop a joint equity relationship, which heighten expectations of relationship continuity and hence increase commitment. The repeated interaction that is thereby encouraged tends to promote socialization and joint learning. However, given that informal processes are to a large extent uncontrollable, changes elicited by formal governance will at best set the trajectory by which such informal processes are likely to evolve. Thus, the role of managers is to constantly monitor the performance of an exchange and intervene, through formal changes, whenever tie strength migrates in either direction significantly beyond the desired level.9 Building upon Nickerson and Zenger (forthcoming), the above assumptions of (1) conflicting contingencies, (2) dynamically changing tie strength due to informal processes that are not directly controllable, and (3) the capacity to alter these dynamics through formal mechanisms lead to the conclusion that exchange performance is optimized not by static matching of tie strength to exchange or industry conditions, but rather by dynamic management of tie strength. Firms initiate weak ties and then strengthen them through formal mechanisms that promote By “managers,” we mean individuals who constantly monitor and attempt to increase the performance of an exchange. To be sure, if managers are deeply involved in the informal processes of socialization and cospecialization, they will not be willing to promote any kind of change anyhow. We provide more comments on this issue in the conclusion section. 9 14 commitment and hence create incentives for socialization and joint learning, such as long-term contracts and joint ownership. Whenever ties become strong and autonomy has been undermined, they are formally restructured or terminated. Then, whenever ties become weak, managers adopt again formal changes to increase commitment. Through this pattern of churning ties the firm dynamically achieves levels of tie strength more closely aligned with the desired level than could be achieved by simply allowing a tie to continuously strengthen or weaken over time. The Logic of Churning Ties Churning ties involve periodic changes in formal governance that alter the path of informal processes influencing tie strength. Please refer to Figure 2. When the relationship is extremely weak (i.e., close to point W) and a stronger tie is desired, managers can pursue formal changes that promote commitment. For instance, managers can extend the duration of a contract or contractually specify shared asset investments. Such changes, which enhance expectations of relationship continuity, encourage repeated interaction, which in turn favors knowledge sharing and the formation of shared norms, values, and routines. If formal changes are successful, then the relationship will reach a point like X, where it attains maximum performance. Cooperation in the form of knowledge sharing will coexist with the ability of firms to learn from one another, as well as from other firms, since they will tend to carry distinct experiences and knowledge sets. Unfortunately, informal processes that influence tie strength will continuously shift. As a result of increased commitment, co-specialization and social attachments will likely escalate. The relationship will thus tend to move towards point S, where cooperation will increase at the expense of reduced autonomy. Limited ability to interpret and acquire new knowledge, added to the dysfunctional effects of socialization—e.g., friendship ties inducing favoritism—will cause a sharp reduction in performance. To avoid having the relationship reach point S, managers can do the reverse: they can formally reorganize the terms of their exchange. For instance, they can shorten the contractual duration. More drastically, they can sever the tie. If either firm severs the tie, both parties must craft new relationships. However, the lack of experience and 15 familiarity with their new partners will be an impediment to promoting investments in relationship-specific assets and knowledge sharing. Thus, the new or existing tie will be close to point W, exhibiting high autonomy but low cooperation. To restore cooperation, managers must again employ formal changes to increase commitment. <Figure 2 around here> However, formal changes designed to adjust tie strength are costly. For instance, costs to strengthen ties may encompass expenses to craft devices such as long-term contracts, whereas costs to weaken ties may encompass the costs of severing relationships, including the associated legal costs to alter clauses or settle disputes. Therefore, decisions to shift tie strength closer to the desired level must be balanced against the costs of change. Due to these costs, managers will need to wait until the benefits of adjusting tie strength exceed the costs of the necessary formal changes. Were change costless, managers would simply churn the ties at small deviations from the desired level of autonomy or cooperation. Thus, while intermediate tie strength cannot be achieved statically, it can be achieved dynamically by using formal mechanisms to periodically increase and decrease tie strength. Although this pattern of strengthening and weakening ties cannot achieve the “ideal” point X, it can at least temporarily achieve balanced levels of cooperation and autonomy that accompany intermediate levels of tie strength, such as point X. In other words, churning ties by continually stimulating, then diminishing tie strength may deliver higher performance than can be achieved by either a string of short-term exchanges or by one extended strong tie.10 The formal mechanisms available for the implementation of churning ties differ in two ways. First, they have differential ability to promote either strengthening or weakening—i.e., formal 10 Rigorously speaking, this occurs if increases in cooperation or autonomy have decreasing marginal effects on performance. Thus, when a tie is strong (high cooperation, low autonomy), further increases in cooperation have little impact on performance compared to even small increases in autonomy. By contrast, when a tie is weak (low cooperation, high autonomy), small increases in cooperation contribute much more to performance than further increases in autonomy. Thus, when the relationship is, say, at point W, movement towards point S will bring cooperation gains that will tend to outweigh initial losses in autonomy. We note that the trajectory depicted in Figure 2 is only illustrative: other paths may be described and may also enhance performance, as long as they are not strongly “convex” (e.g., when, by strengthening a tie, a firm decreases autonomy at a much higher rate than it promotes cooperation). 16 mechanisms vary in terms of the likelihood that they will alter the trajectory of informal processes influencing tie strength, and the resulting pace of change. For instance, formal changes that more likely promote commitment at the outset tend to accelerate the process of socialization and joint learning. Second, such formal mechanisms are associated with distinct levels of implementation and downstream costs, which must be factored in the decision to implement churning ties. We next discuss in detail the formal instruments that can be used to reinforce and weaken ties, emphasizing their relative costs and competencies. Strengthening Weak Ties When ties are weak, lack of commitment inhibits cooperation, and lack of past cooperation induces low willingness to commit. To disrupt this self-reinforcing process, managers must create expectations that the relationship will endure. Basically, managers must adopt formal mechanisms that credibly commit firms to a repeated interaction and, consequently, create incentives for the development of social attachments, co-specialization and routines (Zenger, Lazzarini, & Poppo, 2002). In addition, since crafting contracts and other formal devices requires parties to jointly determine procedures to deal with unexpected changes, the process of contracting itself creates opportunities for social interaction (Poppo & Zenger, forthcoming).11 We discuss below three possible changes in formal governance, arranged increasingly in their ability to create expectations of relationship continuity, and also in their associated costs: loose contractual arrangements, long-term contracts, and joint ownership. Loose contractual arrangements serve to signal intentions for future collaborative ventures. Renewable short-term contracts and narrow agreements based on particular projects to test the competencies of possible partners are examples. Even though such loose agreements are not too costly to implement and their termination is relatively easy, they do not create enough commitment to strengthen ties. Firms are not likely to develop shared routines and co11 There is a debate on whether formal mechanisms promote (Baker, Gibbons, & Murphy, 1994; Poppo et al., forthcoming) or undermine (Fehr & Gächter, 2000; Macaulay, 1963: 64; Sitkin & Roth, 1993) relational governance. This discussion, however, is centered on the effects of contractual incentives or punishments. Our argument that long-term contracts promote tie strengthening is actually simpler: such instruments guarantee that parties will be, to some extent, engaged in a recurring exchange. 17 specialized knowledge merely with a short-term contract. Thus, although loose contractual agreements are not costly, they are not very effective at altering the trajectory of the relationship towards strengthening. Long-term contracts take a step further by formally bonding parties in a recurring relationship or restricting exchange with alternative firms, such as in the case of exclusivity agreements. The degree of commitment enabled by long-term contracts is stronger than in the case of loose contractual agreements since the former explicitly defines the extent of repeated interaction. However, implementation and downstream costs increase. The anticipation of future events and the establishment of procedures to respond to such events engender major challenges for the implementation of long-term contractual relations (Crocker & Masten, 1991; Joskow, 1988). Furthermore, long-term contracts induce downstream costs to alter or terminate them when necessary. To exit from a contractual obligation, it is necessary to incur substantial legal expenses and, in some cases, accomplish side monetary transfers to compensate for losses faced by the other party (Argyres & Liebeskind, 1998). Thus, although long-term contracts are better able to strengthen the tie than loose agreements, they are more costly as well. Finally, managers can strongly commit by making use of joint ownership, such as mutual investments in idiosyncratic assets and joint equity stakes in common organizations (e.g., joint ventures). Joint ownership involves not only high implementation costs related to up front investments, but also high downstream costs since part of such investments may be sunk in the form of dedicated assets and organizational structures (Doz et al., 1998; Harrigan, 1988a). Nonetheless, it is precisely these high downstream costs that make exit costly and thus strongly promote commitment (Anderson et al., 1992; Weiss & Kurland, 1997; Williamson, 1985; Young-Ybarra et al., 1999). Notice that changes in formal governance that are more likely to strengthen the tie are a double-edged sword. While such formal changes more likely achieve high cooperation levels, they have two side effects. First, they tend to have high implementation costs and high downstream costs to restore autonomy whenever necessary. Second, they also tend to increase 18 the pace of tie strengthening, meaning that the relationship will likely display intermediate levels of tie strength for too little time (using Figure 2, the relationship moves away from point X too quickly). Thus, formal changes that more likely promote cooperation at the outset also increase the likelihood that the relationship will quickly achieve excessive commitment levels associated with low autonomy. Weakening Strong Ties Ties can also be excessive in their strength leading to a self-reinforcing process where high commitment reduces firms’ autonomy, which in turn further escalates commitment. To avoid this outcome, managers can pursue tie weakening through a host of changes in formal governance aimed at restoring autonomy by altering the trajectory of socialization and cospecialization. We discuss the following formal changes, arranged in an increasing order in terms of their ability to promote tie weakening and also their costs: reorganization of formal governance, bilaterally negotiated termination, and unilateral termination of the partnership.12 The least aggressive mechanism for weakening ties involves reorganization of the formal governance, by which we mean formal changes in the structure of interaction between firms. Managers can renegotiate the terms of the exchange through the reallocation of ownership stakes (Blodgett, 1992). They can also adjust contractual clauses and modify monitoring procedures (Reuer, Zollo, & Singh, 2002). To reduce the problem of socialization, managers can additionally introduce rules to constantly change the personnel at their interface. For instance, buyer representatives can be “rotated” within a firm in order to avoid dysfunctional friendship ties with seller representatives inducing favoritism and biased pricing (Jeffries et al., 2000). Such rotation practices have been evidenced in the U.S. auto industry (Humphrey et al., 2000: 719). Although these changes are not too costly, they have limited effectiveness to ameliorate the problems caused by excessive tie strength. Thus, when managers reallocate assets without By “termination” we mean the dissolution of the long-term relationship, where parties will become more autonomous than before. Notice that alliances can terminate by merger or acquisition (e.g. Dussauge et al., 2000; Kogut, 1989; Reuer, 2001), which should be interpreted as an increase, rather than a decrease, in tie strength. We do not discuss this possibility here since, for simplicity, we are focusing on interorganizational governance. 12 19 substantial changes in the extent of co-specialization or socialization, they may be mostly redistributing, rather than creating, value. In addition, changes in contract terms may not reduce interfirm commitment if social attachments largely contribute to the governance of the exchange. For instance, reducing the duration of its contracts may not grant a firm real autonomy if friendship ties still influence the choice of possible partners at the moment of contract renewal. Finally, since attitudes towards particular partnering firms tend to be collective rather than simply individually defined (Inkpen et al., 1997; Zaheer et al., 1998), rotation of buyer and seller representatives may not change socially shared norms inducing the same dysfunctional consequences emanating from personal ties.13 Alternatively, managers can bilaterally negotiate the termination of the partnership. This occurs when partners recognize that the opportunities for rent generation with the alliance are limited and jointly decide to pursue other opportunities (Doz et al., 1998). To be implemented, bilateral termination is likely to involve non-negligible negotiation costs to determine the conditions to dissolve the tie and monetary transfers to liquidate joint ownership of assets. Since parties may not agree with those terms and conditions, bilateral termination can be in some cases unfeasible (Inkpen & Ross, 2001: 143). Bilateral termination also entails the downstream costs of building new relationships, i.e., the costs of strengthening newly formed ties. Although costly, bilateral termination—if feasible—alleviates the problems caused by tie strength since it effectively severs existing social ties and provides firms with autonomy to pursue more valuable opportunities and knowledge. A more radical option is to unilaterally terminate the partnership, by which managers deliberately dissolve a long-term tie or open the relationship to alternative firms. For instance, buyers can demand that supply relations be put out for competitive bid to “test the market” (Eccles, 1981: 353). Mandating that procurement transactions will be managed through an 13 For instance, a new buyer representative may have an indirect tie with a seller representative through direct friendship ties with former buyer representatives within his or her firm. In this case, it is possible that the new buyer representative will adopt similar policies used by the former representative when dealing with the supplier. 20 Internet “business-to-business” auction may have this precise effect (Lucking-Reiley et al., 2001; Wise & Morrison, 2000). Competitive bidding may not only be used to find new suppliers, but also to create incentives for incumbent suppliers to increase the efficiency of their production processes and pursue price reductions (Eccles, 1981).14 The costs to implement unilateral termination tend to be substantial, since the lack of mutual consent implies that parties may engage in costly litigation. In addition, there are high downstream costs associated with unilateral termination. It can damage existing ties since opening up the exchange to alternative firms conveys “the risk of offending an important current partner that might compete with a business that is being courted for a future relationship” (Singh et al., 1996: 112). Perhaps more importantly, unilateral termination can also damage a firm’s reputation (Podolny & Page, 1998), as current and future partners will be unwilling to cooperate if they learn that the firm tends to destroy ties at its discretion. Thus, although unilateral termination will quickly alleviate the restrictions imposed by strong ties—e.g., firms will be able to procure products at lower prices or benefit from new technology—it will also increase the difficulty to strengthen remaining or newly formed ties. For instance, Helper (1991: 820) documents that U.S. automakers have faced “a legacy of mistrust, resulting from their years of ‘cutting the legs out from under … our suppliers,’ as one Ford executive put it.” To restore cooperation, managers had to adopt new practices in the industry such as increasing the length of contracts (Helper, 1991). But even with these changes, U.S. automakers have had difficulty triggering informal processes that strengthen ties, given their reputation of switching at will (Mudambi & Helper, 1998). In sum, similarly to the process of strengthening ties, formal changes that most effectively weaken ties are most likely to increase the costs to subsequently strengthen existing or newly formed ties. There also appears to be an asymmetry in this process: relationships can be more 14 This tends to occur when a buyer faces switching costs to transact with new suppliers due to idiosyncratic routines and relationship-specific assets. Under these conditions, incumbent suppliers will have an advantage in competitive bidding (Laffont & Tirole, 1988; Williamson, 1976)—unless, of course, a new supplier with superior technology or radical innovation emerges. 21 easily unraveled than developed, though changes that more quickly weaken ties are more costly as well (e.g., litigation expenses and downstream reputation losses). Thus, the challenge in the management of churning ties is twofold. First, managers need to employ formal changes that create as much commitment as possible whenever a certain tie becomes weak and that, downstream, guarantee as much flexibility as possible to weaken that tie whenever it becomes strong. Second, when weakening a tie, managers must employ changes that restore as much autonomy as possible but that do not severely undermine an existing tie or make it difficult to build new relationships in the future. Moderating Effects The institutional environment in which exchange relations arise will alter the probability of observing churning ties. The institutional environment of a society is the set of formal and informal “rules of the game” that constraint the behavior of individuals and firms (North, 1990). Consider formal institutions. Regulatory controls on the formation and dissolution of interorganizational relations will inhibit the emergence of churning ties by creating an “artificial inertia” in existing ties. For instance, the Japanese government has had an important role in promoting strong ties between buyers and suppliers through its active programs to “organize the subcontractors into keiretsu (channeled groups) [where they] served the same contractor(s)” (Nishiguchi, 1994: 39). There are, however, formal institutions that can actually encourage churning ties. When third-party law enforcement is effective and not costly, firms can have higher confidence that dealings will be honored, thus increasing their willingness to pursue new ties with strangers (Johnson, McMillan, & Woodruff, 2002). As a result, large trade networks where individuals transact less frequently with the same firms will tend to emerge (Greif, 1997; North, 1990). On the other hand, when third-party enforcement is costly and ineffective, firms must necessarily resort to strong ties to promote cooperation (Kali, 1999). Informal institutions at a societal level will also influence the emergence of churning ties. Certain societies rely on committed exchanges because collectivist cultural patterns (Dore, 1983; Greif, 1997; Hill, 1995) and absence of trust in strangers (Yamagishi & Yamagishi, 1994) inhibit 22 the formation of new relationships outside restricted circles. Although cooperation easily unfolds in those societies through committed relations (Dyer & Singh, 1998), the resulting strong ties are extremely difficult to disrupt. Parties will be reluctant to violate social norms and transact with firms or individuals with whom they are not familiar. As a result, churning ties are not likely to be observed when such informal institutions are present. By contrast, societies exhibiting individualistic cultural patterns and high levels of trust in strangers facilitate the establishment of churning ties since parties will tend to avidly pursue valuable opportunities even with unfamiliar firms. Another moderating variable in our model relates to a firm’s competence to form, maintain, and benefit from external relationships, which affects the cost of employing churning ties. Following previous research (Lambe et al., 2000), we use the term relational competence to describe a firm’s ability in selecting valuable and trustworthy partners, managing its alliances, and curbing its own incentives to defect so that its partners will perceive the firm as trustworthy. Firms showing relational competence “will (1) more likely choose partners who will abide by relational norms and (2) understand themselves the value following relational norms” (Lambe et al., 2000: 221). Thus, if firms exhibit relational competence then cooperation will not need to be sustained through repeated interaction, as they will follow general norms such as reciprocity or integrity. This implies that firms exhibiting relational competence will not need to use costly formal mechanisms, such as joint ownership, to strongly commit (Barney & Hansen, 1994). In addition, firms with high relational competencies are more likely to bilaterally negotiate, rather than unilaterally impose the termination of their alliances, since they will have ease in partnering with new firms. Overall, relational competencies will reduce the need for costly formal mechanisms to form and dissolve relationships; thus, the likelihood that firms will engage in new relationships and then switch to other partners will increase.15 15 It is likely that relational competencies are formed through cumulative experience with past alliances, as firms learn how to select valuable partners, and also how to manage their interorganizational relationships (Anand & Khanna, 2000; Gulati, 1999). Relational competencies may also be associated with the existence of dedicated organizational structures to identify partners and manage a firm’s ongoing alliances (Dyer, Kale, & Singh, 2001). 23 Summary of Propositions We summarize below the propositions of our model, schematically represented in Figure 3: P1. In the presence of conflicting contingencies, (a) churning ties—a pattern of periodically reinforcing and weakening interorganizational relationships—will emerge, and (b) exchanges governed by churning ties will outperform exchanges governed by either weak or strong ties permanently. P2. The following formal mechanisms can be used to strengthen ties, arranged in an increasing order in terms of their ability to promote strengthening and the extent to which they will make it more difficult to weaken the resulting ties in the future: loose contractual agreements, long-term contracts, and joint ownership. P3. The following formal mechanisms can be used to weaken ties, arranged in an increasing order in terms of their ability to promote weakening and the extent to which they will make it more difficult to strengthen the resulting ties in the future: reorganization, bilateral termination, and unilateral termination of the partnership. P4. Formal and informal institutions will moderate the effect predicted by P1, by either favoring or inhibiting the emergence of churning ties. (See our previous discussion for specific predictions.) P5. If firms have high relational competencies, they will be (a) less likely to use costly mechanisms to reinforce ties, such as joint ownership, (b) more likely to bilaterally negotiate, rather than unilaterally impose the termination of their alliance, and therefore (c) the frequency in which relationships are formed and dissolved in the context of churning ties will increase. <Figure 3 around here> CHURNING NETWORKS Our level of analysis is purely dyadic: we evaluate the dynamics of change in the strength of a single tie. Therefore, a limitation of our model is that it disregards possible managerial actions that can address the tension between weak and strong ties by “optimizing” the network within which the firm is embedded instead of adjusting ties in isolation. Thus, some authors propose that firms can form a portfolio of weak and strong ties, which supposedly combines the advantages of partnerships and arm’s-length exchanges (e.g. Faulkner, 1983; Uzzi, 1996). Indeed, casual evidence shows that firms often develop committed relationships with some partners, and engage in arm’s-length exchanges with a large number of other firms. Other authors still propose that the relevant issue in the management of networks is whether a tie is 24 redundant or not, rather than the extent of its strength (Burt, 1992). Namely, a firm should avoid forming a tie to an actor that is connected to another actor to which the firm is already tied, since this redundant contact is not likely to provide new information or opportunities.16 A strong tie can still be useful if it is nonredundant (Burt, 1992). Thus, Rangan (2000: 826) posits, “a large network of strong ties to nonredundant actors is the best sort to have.” We contend that both arguments are unsatisfactory responses to the tension between weak and strong ties. For the portfolio of tie argument, we first note that although most firms do have a portfolio of weak and strong ties, this does not necessarily mean that it is a response to conflicting contingencies. Firms may simply be using strong ties for contingencies that call for cooperation (e.g., specific assets), and weak ties for contingencies that call for autonomy (e.g., generic assets). Thus, the portfolio of weak and strong ties should be applied to the same exchange that requires both cooperation and autonomy. But indivisibilities and scale economies will make it costly to maintain more than one or a few partners simultaneously for a single exchange. Furthermore, the portfolio of tie argument neglects the dynamic forces discussed before, which alter the nature of ties over time. Thus, continuous interaction with particular partners will tend to create idiosyncratic routines and co-specialization, which will make it increasingly difficult to transact with other firms. Even if a firm tries to acquire new knowledge through weak ties, those ties will not be useful from a practical standpoint because the firm may fail to interpret external knowledge and even convince the other parties to share it. Conversely, holding weak ties with alternative firms may damage a particular strong tie since the partner may interpret those weak ties as explicit exit options and, consequently, may become reluctant to cooperate (Jones, Hesterly, Fladmoe-Lindquist, & Borgatti, 1998). Therefore, the mix of weak and strong ties applied to a particular exchange is likely to change towards few strong ties with low autonomy, or many weak ties with low cooperation. 16 For evidence supporting this argument, see McEvily and Zaheer (1999) and Gargiulo and Benassi (2000); for evidence contrary to the argument, see Zaheer and Zaheer (1997) and Ahuja (2000). 25 The redundancy argument also suffers from a similar shortcoming: it neglects endogenous changes in networks. Suppose that a firm A, which has a strong tie to another firm B, decides to form a nonredundant tie to firm C. Initially, the tie to firm C will provide firm A with novel information and opportunities. However, by transacting with firm C, firm A creates an indirect tie between B and C. Since the B and C share a common partner, they will have improved information about each other in terms of their capabilities and trustworthiness; for this reason, it is likely that firms B and C will decide to form a relationship. Empirical research shows that the formation of these “cliques” does tend to occur (Gulati, 1995b). The consequence is that the tie A-C will become redundant, thus reducing its benefits as conduit of new information and opportunities. Thus, the mere fact that a firm exploits a nonredundant tie is likely to precipitate an endogenous change—the formation of redundant ties—that undermines over time the benefits that the firm can attain from that action. Despite its beneficial effects, nonredundancy tends to be unstable (Aldrich & Whetten, 1981: 391). This discussion shows that the search for an optimal network configuration, as a static exercise, says little about how firms should dynamically adjust their positions in a context of changing networks. We contend, in particular, that network optimization requires dynamic strategies that are similar in nature to churning ties at a dyadic level. Namely, firms may need to continuously weaken the strong ties and strengthen the weak ties in their portfolio to maintain an appropriate balance of each type of tie.17 Similarly, firms may need to continuously sever redundant ties or build non-redundant ones to increase the value that they can attain from external relations. However, a more careful examination of these dynamic strategies at the network level is beyond the scope of this paper. 17 For instance, in their analysis of corporate customer-bank relationships, Eccles and Crane (1988: 89) show how certain customers “downgrade” certain banks formerly connected through a strong tie to “upgrade” other banks formerly connected through a weak tie. 26 CONCLUSIONS Given the increasing interest in models that describe how interorganizational ties contribute to firm performance (e.g. Dyer & Singh, 1998; Gulati et al., 2000), a careful analysis of the tradeoffs involved in the choice of alternative interfirm governance mechanisms is warranted. Our paper, in particular, discusses a rather pervasive dilemma: in many circumstances, firms want to combine the cooperation-based advantages of strong ties with the autonomy-based advantages of weak ties; they want at the same time commitment and flexibility. We offer a dynamic theory that explains how firms faced with this tradeoff can adjust their ties over time to increase alliance performance. Since firms cannot control informal processes that change the functionality of their relationships over time—i.e., the mix of autonomy and cooperation—they must use a limited set of formal instruments to alter the trajectory of tie strengthening or weakening. Thus, firms can dynamically combine cooperation and autonomy by periodically employing formal mechanisms that weaken ties when excessive commitment damages autonomy and formal mechanisms that reinforce ties when lack of commitment hinders cooperation. Within this perspective, our theory clearly runs against the view that relationship instability is dysfunctional. For instance, Dyer, Cho, and Wujin (1998: 73) assert that “vacillating between arm’s-length relationships and partnerships is unlikely to be a successful strategy given the longterm commitment and relation-specific investments required for strategic partnerships to be successful.” In the presence of conflicting contingencies, vacillation can be indeed a successful strategy: churning ties can deliver superior alliance performance by capturing temporarily the benefits of both weak and strong ties. Contributions to the Literature on Interorganizational Collaboration Our paper brings two main contributions to the literature. First, it adopts a completely different perspective from contingency arguments, which propose a static fit between tie strength and particular exchange or industry conditions. Our theory points out the imperative of dynamic models to predict patterns of interorganizational relations when conflicting contingencies and endogenous changes in the nature of ties render that static match impossible. In addition, by 27 highlighting the role of conflicting contingencies, our theory is consistent with dialectical models of interorganizational relations, which assert that contradictory forces are crucial to explain change (e.g. Das & Teng, 2000; Zeitz, 1980). We contribute to this literature by describing both the key sources of conflict and the resulting dynamics of change involving interorganizational ties. Second, our model potentially integrates two streams of research that have evolved rather independently: the analysis of tie dissolution (e.g. Dussauge et al., 2000; Kogut, 1989; Levinthal & Fichman, 1988; Park et al., 1996) and the analysis of tie formation (e.g. Gulati, 1995b; Kogut, Shan, & Walker, 1992; Powell et al., 1996; Stuart, 1998). Our theory clearly shows that both analyses cannot be divorced, since the decision to reorganize or terminate a tie has consequences for the formation of subsequent ties, and vice versa (Podolny et al., 1998: 70). For instance, as we discussed before, a firm that prematurely severs a tie will have difficulty in building new relationships if its reputation is damaged. Also, the decision to form relationships may actually be associated with the restructuring or termination of existing ties—as Richardson (1972: 896) puts it, “firms form partners for the dance but, when the music stops, they can change them.” Thus, the focus on isolated events of tie formation or dissolution provides only a partial picture of the dynamics and the management of interorganizational relations. Our model fills this void by discussing how different types of formal changes to restructure or sever existing ties influence the formation of new ones, and how the mechanisms employed to build new relationships affect firms’ ability to change them when necessary. Limitations and Possible Extensions An important limitation of our study is that we assume the existence of managers who monitor the performance of exchanges and adopt formal changes to restore either cooperation or autonomy. Although this allows us to focus on the main rationale of churning strategies, it ignores a real possibility: that managers themselves may be part of the processes of socialization and learning which alter the nature of ties over time, and thus may avoid any form of change. For instance, the termination of a relationship may be resisted by managers who “often staked 28 their careers on the success of the alliance project,” even when the relationship proves to be unprofitable (Inkpen et al., 2001: 144). How then to justify the impetus to churn ties? The simplest way, commonplace in economics, is to assume that firms and their managers (through reputation concerns or incentives) face market pressures to make efficient decisions. A more refined justification comes from Beckert’s (1999) model of institutional change. Namely, as a tie approaches stable, yet inferior patterns of functionality—i.e., polar levels of tie strength in the presence of conflicting contingencies—it becomes increasingly salient that change could deliver performance gains. This triggers the emergence of individuals willing to “destroy established taken-for-granted rules if they perceive such action to be profitable” (Beckert, 1999: 786). Such individuals correspond to the managers in our theory. We admit, however, that intervention mechanisms are a black box in our model, and thus deserve further development in subsequent work. Another limitation is that our model suggests that churning ties derive from firms’ willingness to increase the overall value of their relationships. However, firms may be tempted to alter the governance of their exchange to appropriate, rather than create value. For instance, Helper (1991) suggests that buyers may pursue practices associated with weak ties, such as competitive bidding, simply to increase their monopsony power and hence redistribute rents from suppliers. Thus, firms’ relative bargaining power may also influence the emergence and management of churning ties. Apart from the consideration of relational competencies, we also ignore the role of firmspecific attributes affecting the emergence of churning ties. Our model does not consider, for instance, the value that firms can attain by partnering with firms holding beneficial positions in a given network. Thus, a tie with a prestigious firm (Podolny, 1993; Stuart, 1998) may generate reputation-based benefits to other transactions even if that tie yields no direct value. Our model also ignores the role of partner asymmetries (Harrigan, 1988b) in the adoption of mechanisms to strengthen or weaken ties. For instance, firms with a large set of alternative partners will be more able to build new relationship after the termination of an exchange than firms with few 29 potential partners. Such firm-specific variables can be incorporated in our model as factors that moderate the adoption of churning ties and the use of alternative formal mechanisms of change. For instance, managers may be more willing to unilaterally sever a tie if they have a large set of alternative partners, and less willing to do so if particular attributes of their partners, such as prestige, yield positive spillovers to other transactions. Finally, our unit of analysis is a dyadic exchange and thus fails to assess interdependencies in the governance of ties among several firms. Although in the previous section we discuss how our arguments can be scaled up to a network level, other issues remain unexplored. For instance, the benefits to adopt a weak or strong tie are likely to be influenced by the extent to which other firms in the network employ weak or strong ties. Thus, when some firms adopt weak ties, the set of external opportunities available to other firms in the network increases, possibly reinforcing the trend toward weak ties (Kranton, 1996). The adoption of standardized mechanisms such as Internet procurement can be self-reinforcing for this reason: the benefits to use those mechanisms increase as the number of firms adopting the same standard increases (LuckingReiley et al., 2001). Likewise, the adoption of either strong or weak ties can follow institutionalization processes (DiMaggio & Powell, 1983), because firms may follow practices considered as legitimate by shareholders, investors, and other agents. In the presence of those externalities, we may see industry-wide cycles of tie change induced by emergent interactions between multiple agents. Although our model focuses on dyads, it can provide the initial underpinning of a theory explaining the dynamics of networks if one considers that “change always emerges at the level of dyads, where it is potentially generated, received and transmitted to other business relationships” (Halinsen, Salmi, & Havila, 1999: 784). Integrating endogenous, micro-level change occurring at dyads with emergent, macro-level change at networks will provide a major leap towards the understanding of how interorganizational relations evolve and influence firm performance. 30 REFERENCES Adler, P. S., & Kwon, S.-W. 2002. Social capital: prospects for a new concept. Academy of Management Review, 27(1): 17-40. Afuah, A. 2000. How much do your co-opetitors' capabilities matter in the face of technological change? Strategic Management Journal, 21: 387-404. Ahuja, G. 2000. Collaboration networks, structural holes, and innovation: a longitudinal study. Administrative Science Quarterly, 45: 425-455. Aldrich, H., & Whetten, D. A. 1981. Organization-sets, action-sets, and networks: making the most of simplicity. In P. C. Nystrom, & W. H. Starbuck (Eds.), Handbook of organizational design, Vol. 1. New York: Oxford University Press. Anand, B. N., & Khanna, T. 2000. Do firms learn to create value? The case of alliances. Strategic Management Journal, 21: 295-315. Anderson, E., & Weitz, B. 1992. The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research, 29: 18-34. Argote, L. 1999. Organizational learning: creating, retaining and transferring knowledge. Boston: Kluwer Academic Publishers. Argyres, N. S., & Liebeskind, J. P. 1998. Contractual commitments, bargaining, and governance inseparability. Academy of Management Review, 24: 49-63. Arrow, K. J. 1974. The limits of organization. New York: W. W. Norton & Company. Arthur, W. B. 1989. Competing technologies, increasing returns, and lock-in by historical events. The Economic Journal, 99: 116-131. Asanuma, B. 1989. Manufacturer-supplier relationships in Japan and the concept of relationspecific skill. Journal of the Japanese and International Economics, 3: 1-30. Axelrod, R. 1984. The evolution of cooperation. New York: Basic Books. Baker, G., Gibbons, R., & Murphy, K. J. 1994. Subjective performance measures in optimal incentive contracts. The Quarterly Journal of Economics, 109(4): 1125-1156. Ball, J. 2001. Daimler's new boss for Chrysler orders tough, major repairs, The Wall Street Journal: A1, A8. Barney, J. B., & Hansen, M. H. 1994. Trustworthiness as a source of competitive advantage. Strategic Management Journal, 15: 175-190. Barzel, Y. 1982. Measurement costs and the organization of markets. Journal of Law and Economics, 25: 27-48. Beckert, J. 1999. Agency, entrepreneurs, and institutional change. The role of strategic choice and institutionalized practices in organizations. Organization Studies, 20(5): 777-799. Berg, J., Dickhaut, J., & McCabe, K. A. 1995. Trust, reciprocity, and social history. Games and Economic Behavior, 10: 122-142. Blau, P. 1964. Exchange and power in social life. New York: Wiley. Blodgett, L. L. 1992. Factors in the instability of international joint ventures: an event history analysis. Strategic Management Journal, 13: 475-481. Bradach, J., & Eccles, R. G. 1989. Price, authority, and trust. American Review of Sociology, 15: 97-118. Brown, J. S., & Duguid, P. 1998. Organizing knowledge. California Management Review, 40(3): 90-111. 31 Bureth, A., Wolff, S., & Zanfei, A. 1997. The two faces of learning by cooperating: the evolution and stability of inter-firm agreements in the European electronics industry. Journal of Economic Behavior and Organization, 32: 519-537. Burt, R. S. 1992. Structural holes. Cambridge: Harvard University Press. Cohen, W. M., & Levinthal, D. A. 1990. Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 35: 128-152. Cook, K. S., & Emerson, R. M. 1978. Power, equity and commitment in exchange networks. Americal Sociological Review, 43: 721-739. Crocker, K. J., & Masten, S. E. 1991. Pretia ex machina? Prices and process in long-term contracts. Journal of Law & Economics, 34: 69-99. Das, T. K., & Teng, B.-S. 2000. Instabilities of strategic alliances: an internal tensions perspective. Organization Science, 11(1): 77-101. Dawson, C. 2001. Machete time: in a cost-cutting war with Nissan, Toyota leans on suppliers, Business Week: 42-43. DiMaggio, P., & Powell, W. W. 1983. The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48: 147160. Dore, R. 1983. Goodwill and the spirit of market capitalism. The British Journal of Sociology, 34(4): 459-482. Doz, Y. L. 1996. The evolution of cooperation in strategic alliances: Initial conditions or learning processes? Strategic Management Journal, 17: 55-83. Doz, Y. L., & Hamel, G. 1998. Alliance advantage: the art of creating value through partnering. Boston: Harvard Business School Press. Dussauge, P., Garrette, B., & Mitchell, W. 2000. Learing from competing partners: outcomes and durations of scale and link alliances in Europe, North America and Asia. Strategic Management Journal, 21: 99-126. Dutton, J. M., & Starbuck, W. H. 1979. Diffusion of an intellectual technology. In K. Krippendorff (Ed.), Communication and control in society: 489-511. New York: Gordon and Breach Science Publishers. Dyer, J. H. 1996. How Chrysler created an American keiretsu. Harvard Business Review, JulyAugust(42-56). Dyer, J. H., Cho, D. S., & Wujin, C. 1998. Strategic supplier segmentation: the next "best practice" in supply chain management. California Management Review, 40(2): 57-77. Dyer, J. H., Kale, P., & Singh, H. 2001. How to make strategic alliances work. Sloan Management Review, 42(4): 37-43. Dyer, J. H., & Nobeoka, K. 2000. Creating and managing a high-performance knowledgesharing network: the Toyota case. Strategic Management Journal, 21: 345-367. Dyer, J. H., & Singh, H. 1998. The relational view: cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review, 23: 660679. Eccles, R. G. 1981. The quasifirm in the construction industry. Journal of Economic Behavior and Organization, 2: 335-357. Eccles, R. G., & Crane, D. B. 1988. Doing deals: investment banks at work. Boston: Harvard Business School Press. Faulkner, R. R. 1983. Music on demand: composers and careers in the Hollywood film industry. New Brunswick: Transaction Books. 32 Fehr, E., & Gächter, S. 2000. Do incentive contracts crowd out voluntary cooperation?, Working paper, Institute for Empirical Research in Economics, University of Zurich. Feldman, M. P., & Audretsch, D. B. 1999. Innovation in cities: science-based diversity, specialization and localized competition. European Economic Review, 43(2): 409-429. Foss, N. J. 1996. Knowledge-based approaches to the theory of the firm: some critical comments. Organization Science, 7(5): 470-476. Galaskiewicz, J. 1985. Interorganizational relations. Annual Review of Sociology, 11: 281-304. Gargiulo, M., & Benassi, M. 2000. Trapped in your own net? Network cohesion, structural holes, and the adaptation of social capital. Organization Science, 11(2): 183-196. Geringer, J. M., & Hebert, L. 1991. Measuring performance of international joint ventures. Journal of International Business Studies(Second quarter): 249-269. Grabher, G. 1993. The weakness of strong ties: the lock-in of regional development in the Ruhr area. In G. Grabher (Ed.), The embedded firm: on the socioeconomic of industrial networks: 255-277. London and New York: Routledge. Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology, 78: 13601380. Granovetter, M. 1982. The strength of weak ties: a network theory revisited. In P. V. Marsden, & N. Lin (Eds.), Social structure and network analysis: 105-130. Beverly Hills: SAGE. Greif, A. 1997. On the interrelations and economic implications of economic, social, political, and normative factors: reflection from two late medieval societies. In J. N. Drobak, & J. V. C. Nye (Eds.), The frontiers of the new institutional economics: 57-94. San Diego: Academic Press. Gruenfeld, D. H., Mannix, E. A., Williams, K. Y., & Neale, M. A. 1996. Group composition and decision making: how member familiarity and information distribution affect process and performance. Organizational Behavior and Human Decision Processes, 67(1): 1-15. Gulati, R. 1995a. Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38: 85-112. Gulati, R. 1995b. Social structure and alliance formation pattern: a longitudinal analysis. Administrative Science Quarterly, 40: 619-652. Gulati, R. 1999. Network location and learning: the influence of network resources and firm capabilities on alliance formation. Strategic Management Journal, 20: 397-420. Gulati, R., Nohria, N., & Zaheer, A. 2000. Strategic networks. Strategic Management Journal, 21: 203-215. Hagedoorn, J., & Schakenraad, J. 1994. The effect of strategic technology alliances on company performance. Strategic Management Journal, 15: 291-309. Halinsen, A., Salmi, A., & Havila, V. 1999. From dyadic change to changing business networks: an analytical framework. Journal of Management Studies, 36(6): 779-794. Halpern, J. J. 1994. The effect of friendship on personal business transactions. Journal of Conflict Resolution, 38(4): 647-664. Hamel, G. 1991. Competition for competence and inter-partner learning within international strategic alliances. Strategic Management Journal, 12: 83-103. Hansen, M. T. 1999. The search-transfer problem: the role of weak ties in sharing knowledge across organizational subunits. Administrative Science Quarterly, 44: 82-111. Harrigan, K. R. 1988a. Joint ventures and competitive strategy. Strategic Management Journal, 9: 141-158. 33 Harrigan, K. R. 1988b. Strategic alliances and partner asymmetries. In F. J. Contractor, & P. Lorange (Eds.), Cooperative strategies in international business: joint ventures and technology partnerships between firms: 205-226. Lexington: Lexington Books. Heide, J. B., & Miner, A. S. 1992. The shadow of the future: effects of anticipated interaction and frequency of contact on buyer-seller cooperation. Academy of Management Journal, 35(2): 265-291. Helper, S. 1991. Strategy and irreversibility in supplier relations: the case of the U.S. automobile industry. Business History Review, 65: 781-824. Hennart, J.-F. 1988. A transaction costs theory of equity joint ventures. Strategic Management Journal, 9: 361-374. Hill, C. W. L. 1995. National institutional structures, transaction cost economizing and competitive advantage: the case of Japan. Organization Science, 6(1): 119-131. Holmstrom, B., & Milgrom, P. 1994. The firm as an incentive system. American Economic Review, 84: 972-991. Holmstrom, B., & Roberts, J. 1998. The boundaries of the firm revisited. Journal of Economic Perspectives, 12: 73-94. Humphrey, R. H., & Ashforth, B. E. 2000. Buyer-supplier alliances in the automobile industry: how exit-voice strategies influence interpersonal relationships. Journal of Organizational Behavior, 21: 713-730. Inkpen, A. C., & Currall, S. C. 1997. International joint venture trust: an empirical examination. In P. W. Beamish, & J. P. Killing (Eds.), Cooperative strategies: North American perspectives: 308-334. San Francisco: The New Lexington Press. Inkpen, A. C., & Ross, J. 2001. Why do some strategic alliances persist beyond their useful life? California Management Review, 44(1): 132-148. Jeffries, F. L., & Reed, R. 2000. Trust and adaptation in relational contracting. Academy of Management Review, 25(4): 873-882. Johnson, S., McMillan, J., & Woodruff, C. 2002. Courts and relational contracts. Journal of Law, Economics and Organization, 18(1): 221-277. Jones, C., Hesterly, W. S., & Borgatti, S. P. 1997. A general theory of network governance: exchange conditions and social mechanisms. Academy of Management Review, 22(4): 911-945. Jones, C., Hesterly, W. S., Fladmoe-Lindquist, K., & Borgatti, S. P. 1998. Professional service constellations: how strategies and capabilities influence collaborative stability and change. Organization Science, 9(1): 396-410. Joskow, P. L. 1988. Price adjustment in long-term contracts: the case of coal. Journal of Law & Economics, 31: 47-83. Kale, P., Singh, H., & Perlmutter, H. 2000. Learning and protection of proprietary assets in strategic alliances: building relational capital. Strategic Management Journal, 21: 217237. Kali, R. 1999. Endogeneous business networks. Journal of Law, Economics and Organization, 15(3): 615-636. Katz, R. 1982. The effects of group longevity on project communication and performance. Administrative Science Quarterly, 27: 81-104. Kern, H. 1998. Lack of trust, surfeit of trust: some causes of the innovation crisis in German industry. In C. Lane, & R. Bachmann (Eds.), Trust within and between organizations: 203-213. Oxford: Oxford University Press. 34 Kogut, B. 1988. Joint ventures: theoretical and empirical perspectives. Strategic Management Journal, 9: 319-332. Kogut, B. 1989. The stability of joint ventures: reciprocity and competitive rivalry. The Journal of Industrial Economics, 38(2): 183-198. Kogut, B. 2000. The network as knowledge: generative rules and the emergence of structure. Strategic Management Journal, 21: 405-425. Kogut, B., Shan, W., & Walker, G. 1992. The make-or-cooperate decision in the context of an industry network. In N. Nohria, & R. G. Eccles (Eds.), Networks and organizations: structure, form, and action: 348-365. Boston: Harvard Business School Press. Kogut, B., & Zander, U. 1996. What firms do? Coordination, identity, and learning. Organization Science, 7(5): 502-518. Kollock, P. 1994. The emergence of exchange structures: an experimental study of uncertainty, commitment, and trust. American Journal of Sociology, 2: 313-345. Koza, M. P., & Lewin, A. Y. 1998. The co-evolution of strategic alliances. Organization Science, 9(3): 255-264. Krackhardt, D. 1992. The strength of strong ties: the importance of philos in organizations. In N. Nohria, & R. G. Eccles (Eds.), Networks and organizations: 216-239. Boston: Harvard University Press. Kranton, R. E. 1996. Reciprocal exchange: a self-sustaining system. American Economic Review, 86: 830-851. Kranton, R. E., & Minehart, D. F. 2000. Networks versus vertical integration. RAND Journal of Economics, 31(3): 570-601. Laffont, J.-J., & Tirole, J. 1988. Repeated auctions of incentive contracts, investment, and bidding parity with an application to takeovers. RAND Journal of Economics, 19(4): 516-537. Lambe, C. J., Spekman, R., & Hunt, S. D. 2000. Interimistic relational exchange: conceptualization and propositional development. Journal of the Academy of Marketing Science, 28(2): 212-225. Larson, A. 1992. Network dyads in entrepreneurial settings: a study of the governance of exchange relationships. Administrative Science Quarterly, 37(1): 76-104. Leonard-Barton, D. 1995. Wellsprings of knowledge: building and sustaining the sources of innovation. Boston: Harvard Business School Press. Levine, J. M., & Moreland, R. L. 1991. Culture and socialization in work groups. In L. B. Resnick, J. M. Levine, & S. D. Teasley (Eds.), Perspectives on socially shared cognition: 257-279. Washington, DC: American Psychological Association. Levinthal, D. A., & Fichman, M. 1988. Dynamics of interorganizational attachments: auditorclient relationships. Administrative Science Quarterly, 33: 345-369. Levinthal, D. A., & March, J. G. 1993. The myopia of learning. Strategic Management Journal, 14: 95-112. Liebeskind, J. P., Oliver, A. L., Zucker, L. G., & Brewer, M. 1996. Social networks, learning, and flexibility: sourcing scientific knowledge in new biotechnology firms. Organization Science, 7(4): 428-443. Lucking-Reiley, D., & Spulber, D. F. 2001. Business-to-business electronic commerce. Journal of Economic Perspectives, 15(3): 55-68. Luhmann, N. 1979. Trust and power. Chichester: John Wiley and Sons. 35 Macaulay, S. 1963. Non-contractual relations in business: a preliminary study. American Sociological Review, 28: 55-70. Macneil, I. R. 1980. Values in contract: internal and external. Northwestern University Law Review, 78(2): 340-418. Madhok, A., & Tallman, S. B. 1998. Resources, transactions and rents: managing value through interfirm collaborative relationships. Organization Science, 9(3): 326-339. March, J. G. 1991. Exploration and exploitation in organizational learning. Organization Science, 2: 71-87. McEvily, B., & Zaheer, A. 1999. Bridging ties: a source of firm heterogeneity in competitive capabilities. Strategic Management Journal, 20: 1133-1156. Mudambi, R., & Helper, S. 1998. The 'close but adversarial' model of supplier relations in the U.S. auto industry. Strategic Management Journal, 19: 775-792. Nakamura, M., Shaver, J. M., & Yeung, B. 1996. An empirical investigation of joint venture dynamics: evidence from U.S.-Japan joint ventures. International Journal of Industrial Organization, 14: 521-541. Nelson, R. E. 1989. The strength of strong ties: social networks and intergroup conflict in organizations. Academy of Management Journal, 32: 377-401. Nickerson, J. A., & Zenger, T. R. forthcoming. Being efficiently fickle: a dynamic theory of organizational choice. Organization Science. Nishiguchi, T. 1994. Strategic industrial sourcing: the Japanese advantage. New York: Oxford University Press. North, D. C. 1990. Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Oliver, C. 1991. Network relations and loss of organizational autonomy. Human Relations, 44: 943-961. Oxley, J. E. 1997. Appropriability hazards and governance in strategic alliances: a transaction cost approach. Journal of Law, Economics and Organization, 13: 387-409. Park, S. H., & Russo, M. V. 1996. When competition eclipses cooperation: an event history analysis of joint venture failure. Management Science, 42(6): 875-890. Parkhe, A. 1991. Interfirm diversity, organizational learning, and longevity in global strategic alliances. Journal of International Business Studies(Fourth Quarter): 579-601. Parkhe, A. 1993. Strategic alliance structuring: a game theoretic and transaction costs examination of interfirm cooperation. Academy of Management Journal, 36: 794-829. Pfeffer, J., & Salancik, G. R. 1978. The external control of organizations. New York: Harper & Row. Pisano, G. P. 1989. Using equity participation to support exchange: evidence from the biotechnology industry. Journal of Law, Economics, and Organization, 5: 108-126. Podolny, J. M. 1993. A status-based model of market competition. American Journal of Sociology, 98: 829-872. Podolny, J. M., & Page, K. L. 1998. Network forms of organization. Annual Review of Sociology, 24: 57-76. Poppo, L., & Zenger, T. R. 1998. Testing alternative theories of the firm: transaction cost, knowledge-based, and measurement explanations for make-or-buy decisions in information services. Strategic Management Journal, 19: 853-877. Poppo, L., & Zenger, T. R. forthcoming. Substitutes or complements? Exploring the relationship between formal contracts and relational governance. Strategic Management Journal. 36 Powell, W. W., Koput, K. W., & Smith-Doerr, L. 1996. Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology. Administrative Science Quarterly, 41: 116-145. Rangan, S. 2000. The problem of search and deliberation in economic action: when social networks really matter. Academy of Management Review, 25(4): 813-828. Reuer, J. J. 2001. From hybrids to hierarchies: shareholder wealth effects of joint venture partner buyouts. Strategic Management Journal, 22: 27-44. Reuer, J. J., Zollo, M., & Singh, H. 2002. Post-formation dynamics in strategic alliances. Strategic Management Journal, 23: 135-151. Richardson, G. B. 1972. The organization of industry. The Economic Journal, 82(327): 883896. Ring, P. S., & Van de Ven, A. H. 1994. Developmental processes of cooperative interorganizational relationships. Academy of Management Review, 19: 90-118. Rowley, T., Behrens, D., & Krackhardt, D. 2000. Redundant governance structures: an analysis of structural and relational embeddedness in the steel and semiconductor industries. Strategic Management Journal, 21: 369-386. Rulke, D. L., Zaheer, S., & Anderson, M. H. 2000. Sources of managers' knowledge of organizational capabilities. Organizational Behavior and Human Decision Processes, 82(1): 134-149. Schilling, M. A., & Steensma, H. K. 2001. The use of modular organizational forms: an industrylevel analysis. Academy of Management Journal, 44(6): 1149-1168. Seabright, M. A., Levinthal, D. A., & Fichman, M. 1992. Role of individual attachments in the dissolution of interorganizational relationships. Academy of Management Journal, 35(1): 122-160. Singh, K., & Mitchell, W. 1996. Precarious collaboration: business survival after partners shut down or form new partnerships. Strategic Management Journal, 17: 99-115. Sitkin, S. B., & Roth, N. L. 1993. Explaining the limited effectiveness of legalistic "remedies" for trust/distrust. Organization Science, 4(3): 367-394. Stuart, T. E. 1998. Network positions and propensities to collaborate: an investigation of strategic alliance formation in a high-technology industry. Administrative Science Quarterly, 43: 668-698. Tushman, M. L., & Katz, R. 1980. External communication and project performance: an investigation into the role of gatekeepers. Management Science, 26(11): 1071-1085. Uzzi, B. 1996. The sources and consequences of embeddedness for the economic performance of organizations: the network effect. American Sociological Review, 61: 674-698. von Hippel, E. 1994. "Sticky information" and the locus of problem solving: implications for innovation. Management Science, 40(4): 429-439. Wegner, D. M., Erber, R., & Raymond, P. 1991. Transactive memory in close relationships. Journal of Personality and Social Psychology, 61(6): 923-929. Weiss, A. M., & Kurland, N. 1997. Holding distribution channel relationships together: the role of transaction-specific assets and length of prior relationship. Organization Science, 8(6): 612-623. Williamson, O. E. 1976. Franchising bidding for natural monopolies - in general and with respect to CATV. Bell Journal of Economics, 7(Spring): 73-104. Williamson, O. E. 1985. The economic institutions of capitalism. New York: The Free Press. 37 Wise, R., & Morrison, D. 2000. Beyond the exchange: the future of B2B. Harvard Business Review, November-December: 86-96. Yamagishi, T., Cook, K. S., & Watabe, M. 1998. Uncertainty, trust, and commitment formation in the United States and Japan. American Journal of Sociology, 104: 165-194. Yamagishi, T., & Yamagishi, M. 1994. Trust and commitment in the United States and Japan. Motivation and emotion, 18(2): 129-166. Young-Ybarra, C., & Wiersema, M. 1999. Strategic flexibility in information technology alliances: the influence of transaction cost economics and social exchange theory. Organization Science, 10(4): 439-459. Zaheer, A., McEvily, B., & Perrone, V. 1998. Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance. Organization Science, 9(2): 141-159. Zaheer, A., & Zaheer, S. 1997. Catching the wave: alertness, responsiveness, and market influence in global electronic networks. Management Science, 43(11): 1493-1509. Zajac, E. J., & Olsen, C. P. 1993. From transaction cost to transactional value analysis: implications for the study of interorganizational strategies. Journal of Management Studies, 30(1): 131-145. Zeitz, G. 1980. Interorganizational dialectics. Administrative Science Quarterly, 25: 72-88. Zenger, T. R., & Lawrence, B. S. 1989. Organizational demography: the differential effects of age and tenure distributions on technical communication. Academy of Management Journal, 32(2): 353-376. Zenger, T. R., Lazzarini, S. G., & Poppo, L. 2002. Informal and formal organization in new institutional economics. In P. Ingram, & B. S. Silverman (Eds.), The New Institutionalism in Strategic Management: 275-303: Elsevier Science. 38 TABLE 1 Fit (Alignment) Between Contingencies and Tie Strength Contingency Exchange attributes Asset specificity Difficulty to measure performance Type of knowledge Industry characteristics Technological innovation Market volatility Generic task objective 39 Weak tie (autonomy) Strong tie (cooperation) Low Low Codified High High Tacit High High Exploration Low Low Exploitation FIGURE 1 The Dynamics of Tie Change Cooperation S X W Autonomy 40 FIGURE 2 Churning Ties Cooperation X S (a) X (b) W Autonomy Managerial intervention to (a) weaken or (b) strengthen the tie Self-reinforcing processes causing further weakening or strengthening 41 FIGURE 3 The Theory of Churning Ties Loose contractual agreements Strengthening of ties when they become weak Long-term contracts Joint ownership P2 Conflicting contingencies P1 Relational competence of partners Churning ties P4 P5 P3 Unilateral termination Institutional environment Weakening of ties when they become strong Bilateral termination Reorganization Increasing ability to strengthen or weaken the tie 42