اثرات ذخایر دانش جهانی در مورد بهره وری در شرکت های چند ملیتی: نقش عمق و وسعت بین المللی

This study rests upon the premise that differences in the productivity performance of multinational enterprises (MNEs) stem from variations in their ability to access and combine globally distributed knowledge reservoirs within one organization. Its contribution lies in demonstrating that this important source of variation is determined by (a) the idiosyncratic manner in which the MNE's network of subsidiaries is structured, (b) the international breadth and depth of this network and (c) its location choices in the global landscape. We find that when multinationals spread their operations across many geographical markets, they benefit from knowledge externalities more than when they concentrate their activities in few countries. We further show that the ability to exploit spatially distant knowledge depends not only on idiosyncrasies specific to the MNE, but also on exogenous forces associated with international variations in appropriability regimes and industry-specific technological opportunities. As our study considers how the subsidiaries of the MNE collectively influence the productivity of the entire group, it captures complementarities and synergies within the group, and deepens understanding of how MNE-specific and location bound factors jointly shape performance outcomes.

A prevalent theoretical avenue for explaining interfirm variations in productivity rests upon the effects of external knowledge (Argyres and Silverman, 2004, McGahan and Silverman, 2006 and Miller et al., 2007). Notwithstanding the valuable contributions of prior studies, understanding of the forces shaping the ability of multinational enterprises (MNEs) to exploit spatially distant knowledge from various nations remains incomplete. This significantly limits theory development about the role of knowledge externalities in the global economy. Our study addresses this lack of understanding. It centres on the premise that the R&D conducted by other organizations around the world leads to the creation of global knowledge reservoirs (i.e. globally dispersed pools of ideas and specialist knowledge regarding scientific advances and technological developments). Building on this concept, it analyzes the factors influencing the generation and diffusion of knowledge, the ability of MNEs to access knowledge and, thus, the effects of global knowledge reservoirs on the productivity performance of MNEs. The current study extends prior research in three important ways. It is theoretically accepted that each subsidiary is embedded in an interactive and integrated network of subsidiaries (Cantwell and Piscitello, 2000). Studies that approach the MNE as a global whole offer valuable insights into how firms invest abroad (Ito and Wakasugi, 2007 and Belderbos and Zou, 2009), create patents (Penner-Hahn and Shaver, 2005 and Singh, 2008) and transfer knowledge across units (Fisch, 2003 and Kurokawa et al., 2007). However, these contributions overlook the impact on performance arising from the knowledge reservoirs resident in the entirety of the host countries in which the MNE operates. As a result, scholarly understanding of the role of such resources is developing through the accumulation of anecdotes rather than through systematic empirical research (Frost, 2001). We address this weakness by examining whether, and under what conditions, the network of MNE subsidiaries enables the entire group as a whole to become more productive through combining globally dispersed knowledge within one organization. Our analysis relies on a novel mapping approach that matches the location choices of MNEs to the knowledge reservoirs residing in those locations. Rather than focusing on knowledge flows within just one or two nations (Almeida and Kogut, 1999 and Iwasa and Odagiri, 2004), we model productivity as a function of the knowledge originating from 18 countries and 28 industries. Since this approach captures not only the firm's entire network but also most of the world's research efforts, it offers a more complete account of the synergistic function of the MNE and deepens our understanding of how its global operations interact with country-specific resources to determine performance outcomes. Our second contribution lies in demonstrating that variations in MNEs’ ability to benefit from global knowledge reservoirs can be explained by the structure of their portfolios of subsidiaries. To capture the fact that, while some multinational groups spread their operations across a large number of foreign nations (Quintas et al., 2008 and Lahiri, 2010), others concentrate their efforts in a few countries and are more deeply embedded there (Allen and Pantzalis, 1996 and Tang and Tikoo, 1999), we develop the constructs of international depth (the extent of business operations and investment in host countries) and international breadth (the number of countries in which the MNE operates). Building on these multidimensional constructs, we propose and test a set of hypotheses explaining how differences in depth and breadth impact the ability of MNEs to exploit knowledge reservoirs. Since our analysis identifies the factors determining the success of international knowledge sourcing, it has important implications for firm strategy. Third, despite the importance of the supply side of technological change, little research has examined the role of exogenous factors in explaining international variations in knowledge externalities. We complement studies that focus on firms’ own attributes by showing that the benefits of external knowledge are contingent upon (1) the set of technological opportunities in a given industry, and (2) international variations in intellectual property rights (IPR) protection. By combining MNE-, industry- and location-specific factors, our integrated framework enables us to demonstrate how the generation, evolution and diffusion of knowledge may vary across industries and countries; the relative power of these factors in explaining performance asymmetries; how such forces jointly shape productivity outcomes.

Our study rests upon the proposition that interfirm variations in productivity stem from differences in the ability of MNEs to access and combine global knowledge reservoirs within one organization. Its contribution lies in demonstrating that this important source of variation depends on (1) the idiosyncratic manner in which the MNE's network of subsidiaries is structured; on exogenously determined factors pertaining to (2) international differences in IPR protection, and (3) industry-specific technological opportunities. Prior research on knowledge flows focuses on a single country and therefore overlooks the impact on performance arising from the knowledge reservoirs resident in the entirety of the host countries in which the MNE operates. Our framework overcomes this limitation by incorporating not only the MNE's entire network of subsidiaries, but also most of the world's research effort. By examining how the subsidiaries of the MNE collectively impact productivity, the analysis captures complementarities and synergies within the MNE, and deepens our understanding of how MNE-specific and location bound factors jointly shape performance. The construct of knowledge reservoirs developed in this study has proven to have significant power in explaining interfirm productivity differences, thus highlighting the importance of integrating insights from knowledge externalities and internationalization research in theoretical and empirical modelling. Previous studies argue that coordination across MNE units is often weak (Singh, 2008). Despite the difficulty of combining location bound resources from several countries and subsidiaries (Gupta and Govindarajan, 2000), we find that the knowledge accumulated by a subsidiary increases the performance of the entire group. This result suggests that MNEs can successfully integrate globally distributed knowledge reservoirs to create value above the aggregate of the value generated by each individual unit in the network. It also implies that the network of MNEs enables them to turn external knowledge reservoirs – by definition non-proprietary – into proprietary sources of competitive advantage (Frost, 2001). Our study therefore documents how MNE-specific advantages originate from (or are reinforced by) host-country environments, and points to the importance of looking beyond organizational boundaries to explain interfirm productivity variations. In incorporating the role of geographically distributed knowledge, our framework enriches prior work on knowledge sourcing and on the localization of spillover effects (Almeida and Kogut, 1999, Rosenkopf and Almeida, 2003, Chung and Alcácer, 2002 and Alcácer and Chung, 2007). Consistent with previous research, we find the generation of spillovers to be a location bound phenomenon that is determined exogenously. However, we also add to this literature by showing that the ability to pursue externalities is endogenously determined by the internationalization decisions and location choices of the MNE, and the breadth and depth of its network of subsidiaries. Hence, our study not only supports conceptualizations that underscore the role of knowledge in unlocking the economic potential of firms (Kogut and Zander, 1993 and Grant, 1996), but also contributes to these theories by explaining why some MNEs excel at exploiting external knowledge in the global economy, while others are less successful in doing so. Our findings support taking such a contingent view, indicating that knowledge reservoirs are not equally beneficial to all MNEs. Thus they highlight the need to ask not whether overseas subsidiaries tap into location bound knowledge ( Frost, 2001), but under what conditions do they? A deeper understanding of these contingencies may help managers evaluate the strategic implications of their location choices, and encourage firms that are in the habit of centralizing their operations at home, to reshape their strategy and access frontier technology in other countries as effectively as possible. According to our findings, international expansion without careful consideration of the trade-off between depth and breadth may reduce productivity benefits. By creating a link between the structure of international networks and the performance effects of knowledge originating in different locations, we also enrich research on firms’ internal organization (e.g. Argyres and Silverman, 2004). Even if two MNEs have the same level of foreign assets, sales and number of subsidiaries, their international depth and breadth and location choices may differ considerably. These choices, in turn, may influence how far each MNE benefits from global knowledge reservoirs, and so result in different productivity outcomes for each firm. In this respect, our findings indicate that, from the point of view of knowledge, entering new markets is likely to be more effective in raising productivity performance than deepening foreign presence in markets in which the MNE already operates. Even so, as the optimal balance between depth and breadth varies depending on firm- and industry-specific idiosyncrasies, the decision as to which of these two dimensions to prioritize should be left to managers. Our results also have implications for understanding the interaction between internal and external knowledge resources. According to the absorptive capacity hypothesis (Cohen and Levinthal, 1990 and Escribano et al., 2009), firms must possess an internal knowledge stock to benefit from externalities. Although our findings do not necessarily contradict this premise, they indicate that the cumulative or additive effect of knowledge reservoirs is greater for MNEs with a smaller knowledge stock. This finding extends research on knowledge localization by showing that there is a trade-off between building internal knowledge and leveraging external knowledge resources from a variety of locations. It implies that the knowledge benefits of collocation can partially substitute for the knowledge base of the multinational group. The finding that proximity to knowledge reservoirs is more advantageous to multinationals that have deficiencies in internal resources is surprising and counterintuitive given the weight of research suggesting that tapping into foreign knowledge is a not a panacea for firms that are lagging in technological capabilities (Penner-Hahn and Shaver, 2005). Rather, it furnishes support for the view that the overlap between internal knowledge and external resources decreases the likelihood of identifying complementarities and, thus, the overall value of collocation. In doing so, it is congruent with managerial prescriptions advocating that firms undertaking little R&D may nevertheless succeed in securing profitable ideas from outside (Chesbrough, 2003 and Alcácer and Chung, 2007). It is also in line with the widely debated asset-seeking view, namely that cross-border expansion may be driven by the need to gain knowledge from abroad. A similarly intriguing story emerges for IPR protection. Prior studies suggest that weak appropriability regimes decrease the returns to a firm's own innovation. Notwithstanding these negative consequences, we find evidence that weak IPR regimes increase the performance of MNEs by enabling them to exploit external ideas. By showing that advantageous regimes are not necessarily strong (Pisano, 2006), our findings challenge the prevalent view that environments with inefficient protection are detrimental to innovation and performance. Although these regimes may not be desirable for single-unit organizations, it appears that they work in favour of multi-location MNEs that may be particularly adept at using co-specialized resources to protect their innovations. The evidence that MNEs actually profit from weak appropriability regimes supports the view that multinationals use internal organization to compensate for ineffective external institutions. Our findings therefore offer an explanation for the paradox of why MNEs can find it attractive to conduct R&D in territories where IPR protection is inadequate (Zhao, 2006). Our study rests upon the proposition that interfirm variations in productivity stem from differences in the ability of MNEs to access and combine global knowledge reservoirs within one organization. Its contribution lies in demonstrating that this important source of variation depends on (1) the idiosyncratic manner in which the MNE's network of subsidiaries is structured; on exogenously determined factors pertaining to (2) international differences in IPR protection, and (3) industry-specific technological opportunities. Prior research on knowledge flows focuses on a single country and therefore overlooks the impact on performance arising from the knowledge reservoirs resident in the entirety of the host countries in which the MNE operates. Our framework overcomes this limitation by incorporating not only the MNE's entire network of subsidiaries, but also most of the world's research effort. By examining how the subsidiaries of the MNE collectively impact productivity, the analysis captures complementarities and synergies within the MNE, and deepens our understanding of how MNE-specific and location bound factors jointly shape performance. The construct of knowledge reservoirs developed in this study has proven to have significant power in explaining interfirm productivity differences, thus highlighting the importance of integrating insights from knowledge externalities and internationalization research in theoretical and empirical modelling. Previous studies argue that coordination across MNE units is often weak (Singh, 2008). Despite the difficulty of combining location bound resources from several countries and subsidiaries (Gupta and Govindarajan, 2000), we find that the knowledge accumulated by a subsidiary increases the performance of the entire group. This result suggests that MNEs can successfully integrate globally distributed knowledge reservoirs to create value above the aggregate of the value generated by each individual unit in the network. It also implies that the network of MNEs enables them to turn external knowledge reservoirs – by definition non-proprietary – into proprietary sources of competitive advantage (Frost, 2001). Our study therefore documents how MNE-specific advantages originate from (or are reinforced by) host-country environments, and points to the importance of looking beyond organizational boundaries to explain interfirm productivity variations. In incorporating the role of geographically distributed knowledge, our framework enriches prior work on knowledge sourcing and on the localization of spillover effects (Almeida and Kogut, 1999, Rosenkopf and Almeida, 2003, Chung and Alcácer, 2002 and Alcácer and Chung, 2007). Consistent with previous research, we find the generation of spillovers to be a location bound phenomenon that is determined exogenously. However, we also add to this literature by showing that the ability to pursue externalities is endogenously determined by the internationalization decisions and location choices of the MNE, and the breadth and depth of its network of subsidiaries. Hence, our study not only supports conceptualizations that underscore the role of knowledge in unlocking the economic potential of firms (Kogut and Zander, 1993 and Grant, 1996), but also contributes to these theories by explaining why some MNEs excel at exploiting external knowledge in the global economy, while others are less successful in doing so. Our findings support taking such a contingent view, indicating that knowledge reservoirs are not equally beneficial to all MNEs. Thus they highlight the need to ask not whether overseas subsidiaries tap into location bound knowledge ( Frost, 2001), but under what conditions do they? A deeper understanding of these contingencies may help managers evaluate the strategic implications of their location choices, and encourage firms that are in the habit of centralizing their operations at home, to reshape their strategy and access frontier technology in other countries as effectively as possible. According to our findings, international expansion without careful consideration of the trade-off between depth and breadth may reduce productivity benefits. By creating a link between the structure of international networks and the performance effects of knowledge originating in different locations, we also enrich research on firms’ internal organization (e.g. Argyres and Silverman, 2004). Even if two MNEs have the same level of foreign assets, sales and number of subsidiaries, their international depth and breadth and location choices may differ considerably. These choices, in turn, may influence how far each MNE benefits from global knowledge reservoirs, and so result in different productivity outcomes for each firm. In this respect, our findings indicate that, from the point of view of knowledge, entering new markets is likely to be more effective in raising productivity performance than deepening foreign presence in markets in which the MNE already operates. Even so, as the optimal balance between depth and breadth varies depending on firm- and industry-specific idiosyncrasies, the decision as to which of these two dimensions to prioritize should be left to managers. Our results also have implications for understanding the interaction between internal and external knowledge resources. According to the absorptive capacity hypothesis (Cohen and Levinthal, 1990 and Escribano et al., 2009), firms must possess an internal knowledge stock to benefit from externalities. Although our findings do not necessarily contradict this premise, they indicate that the cumulative or additive effect of knowledge reservoirs is greater for MNEs with a smaller knowledge stock. This finding extends research on knowledge localization by showing that there is a trade-off between building internal knowledge and leveraging external knowledge resources from a variety of locations. It implies that the knowledge benefits of collocation can partially substitute for the knowledge base of the multinational group. The finding that proximity to knowledge reservoirs is more advantageous to multinationals that have deficiencies in internal resources is surprising and counterintuitive given the weight of research suggesting that tapping into foreign knowledge is a not a panacea for firms that are lagging in technological capabilities (Penner-Hahn and Shaver, 2005). Rather, it furnishes support for the view that the overlap between internal knowledge and external resources decreases the likelihood of identifying complementarities and, thus, the overall value of collocation. In doing so, it is congruent with managerial prescriptions advocating that firms undertaking little R&D may nevertheless succeed in securing profitable ideas from outside (Chesbrough, 2003 and Alcácer and Chung, 2007). It is also in line with the widely debated asset-seeking view, namely that cross-border expansion may be driven by the need to gain knowledge from abroad. A similarly intriguing story emerges for IPR protection. Prior studies suggest that weak appropriability regimes decrease the returns to a firm's own innovation. Notwithstanding these negative consequences, we find evidence that weak IPR regimes increase the performance of MNEs by enabling them to exploit external ideas. By showing that advantageous regimes are not necessarily strong (Pisano, 2006), our findings challenge the prevalent view that environments with inefficient protection are detrimental to innovation and performance. Although these regimes may not be desirable for single-unit organizations, it appears that they work in favour of multi-location MNEs that may be particularly adept at using co-specialized resources to protect their innovations. The evidence that MNEs actually profit from weak appropriability regimes supports the view that multinationals use internal organization to compensate for ineffective external institutions. Our findings therefore offer an explanation for the paradox of why MNEs can find it attractive to conduct R&D in territories where IPR protection is inadequate (Zhao, 2006). Finally, another contribution of our study lies in demonstrating that the rewards multinationals reap from knowledge reservoirs are contingent upon the set of technological opportunities in a given industry. We find that, irrespective of their location choices, MNEs benefit more from global knowledge reservoirs when they participate in industries in which technical understanding grows rapidly. This result demonstrates that differences in performance can arise from variations in MNEs’ technological positioning, and points to the significance of exogenous factors in conceptualizing the effects of spillovers. The practical ramifications of this is that when managers reorganize their product offerings or enter new product categories, they should carefully evaluate how the possibilities for technological advance in a given field may either impose important constraints or present valuable opportunities (Rosenberg, 1974). Although this factor is beyond the control of the firm, managers should respond by directing some of their efforts towards regions of technology where the possibilities for new understanding are higher. Indeed, as investing in dynamic technological domains while expanding their operations abroad is a fruitful mechanism for benefiting from external knowledge, careful consideration of these joint moderating effects should be a central part of their strategy.