14Why Do MNCs Divest or Retain Foreign Subsidiaries?Approaches from Dependency and Redundancy in Subsidiary Networks Naoki YasudaDISCUSSION AND CONCLUSION This study sheds light on subsidiary networks, extending the literature on divestment of foreign subsidiaries by analyzing the data of Japanese affiliates in manufacturing industries from 1995 to 2013. By applying the concept of dependency, results show that hub functions and authority functions discourage divestment. Findings suggest that if other subsidiaries in subsidiary network are highly dependent on a given subsidiary, MNCs are less likely to divest that subsidiary. However, hypotheses 3a and 3b, derived from the concept of redundancy, are not supported since the results show effects contrary to the predictions of these hypotheses. This is likely because there are two different explanations for the same function of a given subsidiary: (1) reducing redundancy and (2) economies of density. According to the economies of density, cost effectiveness rises if there are many similar subsidiaries in a country or a region (e.g., Caves, Christensen, & Tretheway, 1984). Based on this argument, MNCs are less likely to divest subsidiaries to maintain cost effectiveness. This is the polar opposite of the reducing-redundancy argument. Results suggest that the economies of scale explanation are supported by the estimations. This study advances the understanding of existing literature in several ways. First, it extends the literature on FDI theory by classifying subsidiary functions, such as “networked FDIs.” Second, it contributes to the divestment literature. Unlike previous studies, which view divestment in light of the characteristics of the participants, this study invokes concepts of dependency and redundancy to identify foreign subsidiaries that have a stronger likelihood of being divested. These findings have managerial implications; for example, managers may be able to predict which competitors are likely to divest which foreign subsidiaries. Aside from the findings, this study identifies several important areas for future research. First, regarding the third hypotheses, there are two competing arguments on divestment of subsidiaries when a focal subsidiary duplicates the function of surrounding subsidiaries. Finding a resolution to these competing explanations is an opportunity for future research. Second, this study focuses on divestment decisions by MNCs. However, a concept of reconfiguration refers to the management of divested resources (Karim, 2006). MNCs redeploy the divested resources to other subsidiaries or headquarters. Therefore, considering the resource management process following subsidiary divestments by MNCs is a fruitful opportunity for future research. I acknowledge that this study has several limitations. First, it applies a logic of dependency and redundancy. However, theoretically, these concepts originate in resource dependence theory and social network theory (Bouquet & Birkinshaw, 2008). According to resource dependence theory, dependency is a source of power for actors (Pfeffer & Salancik, 1978). The concept of power argues that actor A’s dependence on actor B implies that B holds power over A (Emerson, 1962). Structural holes in social network theory show benefits of non-redundancy among network contacts (Burt, 1992: 20). Non-redundancy is valuable because it offers access to non-redundant information and the power to control unconnected parties (e.g., Burt, 1992). More efficient networks have more
元のページ ../index.html#18