متعادل کردن تجارت کردن بین چشم اندازهای یادگیری و خطرات خارجی: الگوهای عمودی ارتباط شرکتهای تابعه MNC 'در کشورهای توسعه یافته

This paper investigates local vertical linkages of foreign subsidiaries and the dual role of such linkages as conduits for learning as well as potential channels for spillovers to competitors. On the basis of data from 97 subsidiaries, we analyze the quality of such linkages under varying levels of competition and subsidiary capabilities. Our theoretical development and the results from the analysis document a far more complex and dynamic relationship between levels of competition and MNCs’ local participation in knowledge intensive activities, i.e. learning and spillovers, than previous studies do. We find a curvilinear relationship between the extent of competitive pressure and the quality of local linkages confirming our argument of a trade-off between learning prospects and spillover risks. Furthermore, the level of subsidiary capabilities moderates this relationship.

An important consequence of foreign direct investment (FDI) lies in the phenomenon of local linkages, i.e. non-equity relationships that multinational corporation (MNC) subsidiaries develop with local firms in their host countries (Chen, Chen, & Ku, 2004). There is a substantial strand of literature that has characterized linkages’ attributes (Giroud and Scott-Kennel, 2009, Santangelo, 2009, Scott-Kennel, 2007 and Scott-Kennel and Enderwick, 2004), investigated their antecedents (Belderbos et al., 2001, Giroud and Mirza, 2006, Jindra et al., 2009 and Santangelo, 2011), and analyzed their consequences (Andersson et al., 2002, Holm et al., 2005 and Hansen et al., 2009). This literature has recently suggested that local linkages have a dual effect (Giroud & Scott-Kennel, 2009). On the one hand, local linkages act as channels through which MNC knowledge spills over to local firms (Driffield et al., 2002 and Ghauri and Buckley, 2006). On the other hand, they also act as conduits for subsidiary learning from the domestic environment (Andersson et al., 2002, Mu et al., 2007 and Giroud and Scott-Kennel, 2009). Not all local relationships have the same potential for subsidiary learning and spillovers. Building on network research (Granovetter, 1985 and Uzzi, 1996), both sub-streams of literature on learning and spillovers have suggested that these effects depend on the quality of the linkages (Andersson et al., 2002, Giroud and Scott-Kennel, 2009, Saliola and Zanfei, 2009 and Santangelo, 2009) – to some extent also referred to as linkage intensity (Giroud and Scott-Kennel, 2009 and Scott-Kennel and Enderwick, 2005) or embeddedness (Andersson et al., 2002). Linkages of high quality can be characterized by partners’ interdependence, mutual adaptation, and breadth of interaction in terms of possibilities to exchange fine-grained knowledge and information (Andersson et al., 2002, Giroud, 2003, Gulati, 1998, Uzzi, 1996 and Uzzi, 1997). As a consequence, they are more effective than arm's-length relations for information and knowledge flows in both directions. While high quality linkages offer important learning opportunities, they simultaneously expose the subsidiary's knowledge to the risk of spillover to the host-economy (Blomström and Kokko, 1998, Mudambi and Navarra, 2004 and Sanna-Randaccio and Veugelers, 2007). Despite their importance, research on the antecedents of local linkage quality is still scarce (Jindra et al., 2009 and Santangelo, 2009) for three reasons. First, while previous literature has analyzed the influence of subsidiary- and sector-specific variables (Chen et al., 2004, Holm et al., 2005, Jindra et al., 2009, Scott-Kennel and Enderwick, 2005 and Scott-Kennel, 2007), the role of the local environment and, more specifically, the role of local competitive pressure remains under-investigated despite substantial evidence suggesting that local competition is a major element influencing MNC strategy (Alcácer and Chung, 2007, Kogut and Chang, 1991 and McCann and Mudambi, 2005). Second, previous research has failed to investigate how such competition might interact with the subsidiaries’ level of capabilities. This is despite evidence in the network literature showing that the choice of building and developing linkages depends on both firm internal and external factors (Andersson et al., 2005, Frost et al., 2002, Gulati and Gargiulo, 1999 and Luo, 2003) and that these factors might interact when explaining competitive action (Blanc & Sierra, 1999). Finally, most literature on local linkages has focused on less advanced and developing economies (e.g. Hansen et al., 2009, Jindra et al., 2009 and Santangelo, 2009). While this approach increases our understanding of how such countries can benefit from foreign MNC activity (Hoekman and Javorcik, 2006 and Kugler, 2006), it neglects the linkage patterns in developed contexts, where domestic actors are likely to be highly competent, equipped with absorptive capacity, and located in competitive industries. These conditions make local firms desirable vertical partners for subsidiaries’ learning but, simultaneously, increase the risk of an erosion of competitive advantage due to spillovers. In this study we address these limitations and investigate the quality of vertical local linkages, i.e. of supply chain relationships that foreign subsidiaries build with local suppliers and customers (Giroud & Scott-Kennel, 2009). We posit that the extent to which the subsidiary perceives the local environment as highly competitive as well as the level of the subsidiary's own capabilities affect the trade-off between learning opportunities and potential spillovers. In turn, this influences the subsidiaries’ investment into their local relationships, i.e. subsidiaries adapt the quality of their linkages to these characteristics. Our results confirm our argument. We find a curvi-linear relationship between perceived local competitive pressure and the quality of linkages. In addition, the level of the subsidiary's capabilities negatively moderates this curvi-linear relationship. Our study has several contributions. First, we contribute to the recent stream of research on local linkages of MNC subsidiaries (Chen et al., 2004, Jindra et al., 2009, Santangelo, 2009 and Saliola and Zanfei, 2009). We confirm literature that has argued that local competitive pressure is an important influencing factor on MNC strategy (e.g. Holm et al., 2005). Furthermore, we show that in developed countries, increasing local competitive pressure can be positively or negatively related to the quality of local linkages because of spillover risks and learning opportunities: it depends on the initial level of competition in the host country. This study extends previous literature that suggested a more simple effect of local competitive conditions on MNC strategic behavior (e.g. Alcácer, 2006 and Alcácer and Chung, 2007). It also adds to our understanding of the circumstances under which host countries might profit most from the presence of foreign firms (Marin & Bell, 2006). Second, our findings support literature that has argued that both firm internal and external factors need to be integrated in studies on linkages (e.g. Giroud & Scott-Kennel, 2009), since they might interact with each other (Alcácer & Chung, 2007). In our study, subsidiary capabilities have an important role as they moderate the effect of increasing competition. The building, development and adaptation of MNC host country linkages is apparently a highly complex process. Third, based on our findings, we suggest that studying linkages in developed countries is important as learning opportunities and spillover risks increase in such environments, thus leading to strong reactions by subsidiaries. This provides a complement to studies on emerging/developing countries (e.g. Jindra et al., 2009). Finally, we argue that FDI phenomena, such as local linkages, can be better explained by complementing traditional economic reasoning with findings from network theory. To this end, we confirm that studying the quality of linkages is important (Giroud and Scott-Kennel, 2009 and Scott-Kennel, 2007). We also add that perceptions of environmental conditions are strong drivers of subsidiary behavior. This is an important dimension to study because the network-based literature states that the context of business relationships is socially constructed (e.g. Anderson, Håkansson, & Johanson, 1994). Firms react to their perceived environment, rather than simply adapting to constraints exerted by an “intractable externality” ( Astley and Fombrun, 1983, p. 576). The remainder of the paper is organized as follows. In the next section, we review the existing research on local linkages, spillovers and subsidiary learning, and recall the relevance of quality linkages. We then elaborate on the “trade-off” between local learning and spillover associated with quality linkages. Subsequently, we develop and test our model. We conclude with a discussion of our empirical results, the study's limitations, and practical implications.

We analyze the relationship between environmental conditions and the strategic behavior of MNCs by focusing on the impact of perceived local competitive pressure on the foreign subsidiary behavior regarding adjustments in the quality of linkages to vertical local partners. We show that subsidiaries in developed countries adapt the quality of their most important local relationships to the perceived level of competition in the host environment and to their level of capabilities. Our findings reveal a curvilinear relationship between local competitive pressure and the quality of subsidiaries’ vertical linkages and a negative moderating effect of the subsidiary capability level on this relationship. This paper offers three major contributions. First, we contribute to the recent stream of literature on subsidiary local linkages (Chen et al., 2004, Jindra et al., 2009, Saliola and Zanfei, 2009 and Santangelo, 2009). Previous research has paid scant attention to the environmental antecedents of the quality of linkages and, in particular, to the role of local competitive conditions. We suggest that subsidiaries adapt their local vertical linkages to the changing competitive conditions in the host-country in an attempt to manage the trade-off between learning opportunities and spillover threats. Our results are consistent with previous work arguing that local competition is one of the most critical factors of the host-country influencing the strategic behavior of foreign firms (Blomström et al., 1992, Holm et al., 2005, Kogut and Chang, 1991 and McCann and Mudambi, 2005). Our finding of a curvilinear relationship between local competition and the quality of vertical linkages also contributes to the stream of literature suggesting a rather simple relationship between host country competition and firm strategic behavior. Previous studies (Alcácer, 2006, Baum and Haveman, 1997, Cantwell and Santangelo, 2002 and Sanna-Randaccio and Veugelers, 2007) have shown that firms tend to avoid co-locating or starting knowledge-intensive activities in regions characterized by high competition (Alcácer and Chung, 2007 and Sanna-Randaccio and Veugelers, 2007), i.e. competition inhibits MNCs’ local participation. In contrast, we show that local competitive pressure does not always deter subsidiaries’ interaction with local agents. Instead it has a double role as the marginal effect of a one-unit increase in competitive pressure does depend on the initial level of competition. Whereas very high competitive pressure discourages subsidiaries from increasing the quality of their local relationships because of spillover risks, lower levels of competitive pressure make spillover risks less critical. As a consequence, additional learning opportunities weigh stronger and encourage the subsidiary to increase the quality of local linkages. Firms’ strategic reaction to local competition is not univocal as suggested by previous research (Baum & Haveman, 1997), but rather varies depending on the perceived level of competitive pressure. Furthermore, we highlight that the reaction to varying levels of competition also depends on the extent of subsidiaries’ capabilities. In fact, when subsidiaries have very low levels of capabilities they will not reduce the quality of their linkages even in presence of very high competition: they have nothing to lose, but still a lot to learn. By showing that the complex relationship between competition and quality linkages is further moderated by subsidiary capabilities, our study is also in line with recent literature arguing that firms’ endowment of resources and capabilities has to be integrated in analyses on spillovers and learning (Alcácer and Chung, 2007 and Santangelo, 2011). Previous research has shown that host-competition and firm capabilities interact in determining MNCs’ location choices (Alcácer & Chung, 2007). We add to this literature by showing that not only the relationship between competition and location choice is moderated by firm capabilities, but also the relationship between competition and inter-organizational strategy. We also extend previous literature by using a developed country context. We find that especially in developed countries the interaction of local competition and subsidiary capabilities has a strong impact on subsidiary's networking behavior as subsidiaries attempt to manage the bi-directional flow of knowledge. This substantially extends the work on MNC host-country linkages that, to-date, has primary focused on vertical linkages in less developed or developing host countries (Hansen et al., 2009, Jindra et al., 2009 and Santangelo, 2009). Second, our study provides support to recent work attempting to enrich the traditional literature on FDI (Buckley and Casson, 1976, Dunning, 1980 and Hymer, 1976) with findings from network research (Andersson et al., 2002 and Ghauri et al., 2005). The former has compiled substantial evidence on the phenomenon of MNC host country linkages, the reasons for firms to invest into such relationships, and their potential consequences. We complement traditional FDI literature by studying the quality dimension of local linkages and emphasizing the role of perceptions subsidiary managers have of environmental threats and opportunities. To this end, we support previous literature that has argued for including the quality dimension of local linkages as a much better proxy for bi-directional knowledge flows that are so critical for the net effect of FDI ( Giroud & Scott-Kennel, 2009). Furthermore, we adopt from network theory a much more socially enriched understanding of economic exchange. In particular, network research has established the idea that perceptions of environmental conditions drive the behavior of firms ( Anderson et al., 1994). In our model, the subsidiaries’ perceptions of local competitive pressure and their own level of capabilities help explain their inter-organizational strategy. Thus, we argue that network theory offers perspectives that may complement traditional FDI theory, and help gaining a more realistic understanding of how foreign firms manage the trade-off between potential learning and spillovers. Third, our findings have a bearing on the literature on FDI spillovers to host-markets (Castellani and Zanfei, 2006 and Santangelo, 2009). This literature has so far reported that more capable foreign subsidiaries typically establish high quality linkages and, as a result, are sources of potential spillovers to the host markets (Jindra et al., 2009, Marin and Bell, 2006 and Santangelo, 2009). Shifting the context of analysis to more developed countries, our study shows that the potential spillovers associated to more capable subsidiaries critically depend on local market competitive conditions to the extent that spillovers to the host market may fail to materialize under strong local competition. Highly capable subsidiaries decrease the quality of linkages in presence of strong local competition to protect their valuable assets, thus reducing the potential for spillover. We show that medium levels of competition might provide an optimum level for host countries to profit from the presence of foreign MNCs. This result is consistent with previous research suggesting that an industry's innovative output is maximized in presence of moderate levels of competition (Scherer, 1965). The mirror consequence of this reasoning provides a major insight to the literature on subsidiary learning (Almeida, 1996 and Almeida and Phene, 2004). This literature has regarded the host market primarily as a source of learning for the foreign subsidiary (Cantwell, 1989, Mudambi and Navarra, 2004 and Phene and Almeida, 2008). Our analysis confirms this argument only under the conditions of low perceived competitive pressure. We find that some subsidiaries might consciously forgo the opportunity to learn from high-quality relationships with local partners in developed countries. This is because the relative detrimental effects of spillovers outweigh the learning benefits. 5.1. Limitations Our study suffers from some limitations. First, we do not cover the effects of other types of “spillover-controlling” mechanisms, such as formal protection strategies (De Faria & Sofka, 2010), beyond the adjustment of local linkages quality. Future research could investigate other mechanisms that foreign subsidiaries adopt to defend their knowledge, and how these are used to manage the environmental threats. Second, only linkages to customers and suppliers are included in our analysis. Yet, despite a certain level of underestimation of spillovers and learning from the local context this approach might cause, both effects are arguably affected in the same way thus not skewing our analysis. Third, while our focus on the six most important relationships of subsidiaries assures that the relationships do bear some level of importance and are thus managed with care, we have limited knowledge on the rationale underlying the assessment of this importance. However, this problem is not uncommon in other disciplines. For instance, ego-centered network studies often ground their empirical analysis on the identification (made by the informants) of a number of personal acquaintances or colleagues in order to collect information on network density and centrality. The criteria through which these acquaintances are selected are rather general and do not seem to be an issue (Ibarra, 1993 and Morrison, 2002). Finally, our measure of subsidiary capabilities is based on subsidiaries’ skills in terms of purchasing and sales activities. Although this is consistent with our focus on vertical linkages, it does not include technological skills. This is not a strong concern as both purchasing as well as marketing/sales capabilities can be important sources of competitive advantage (e.g. Heide, 1994). There is also evidence suggesting that technical and marketing capabilities are correlated, as the effective management of upstream and downstream markets represents a strong input for firms’ technical development processes (Calantone and Di Benedetto, 1988 and Moorman and Slotegraaf, 1999). Nevertheless, future research could validate our findings in the specific context of technological capabilities. 5.2. Practical implications This study bears interesting practical implications. For subsidiary managers, our results imply that adaption processes regarding inter-organizational strategies of subsidiaries are immensely complex undertakings. Subsidiary managers need to embrace such complexity and be aware that inter-organizational strategies encompass both gains and costs, which are driven by a multitude of factors. Inter-organizational relationships need to be managed with care, especially in developed market contexts. This requires subsidiary managers to be able to scrutinize carefully local conditions and relate that to firm-internal capabilities. For headquarters managers, our findings suggest that subsidiaries may be able to balance learning opportunities with spillover risks. Thus, avoiding FDI in a risky location altogether might be too strong of a reaction. For policy makers, it is important to notice adverse selection logics in subsidiaries’ linkage behavior. More skilled subsidiaries will shy away from high quality local linkages in presence of strong competition and only the less capable subsidiaries will be eager to develop close and interdependent linkages with local firms. In this case, the net outcome of subsidiaries’ linkages for domestic firms might be lower than expected as the domestic firms might face situations in which they share more valuable knowledge with foreign subsidiaries than they receive back from them.