What determines knowledge sourcing from host locations of overseas R&D operations? We investigate factors that influence the extent to which overseas R&D laboratories source knowledge from host locations. Drawing on both the capabilities perspective and the embeddedness perspective, we have developed a conceptual model and then examined it empirically focusing on overseas R&D labs of Japanese multinationals. Statistical findings from negative binomial regressions show that both technological capabilities of the lab and external embeddedness in the local scientific and engineering communities matter.
The globalization of research and development (R&D) is an important component of the ongoing trend towards globalization of the economy (Guellec and van Pottelsberghe de la Potterie, 2001). According to Gerybadze and Reger (1999), the degree of globalization of R&D measured by various indicators such as the proportion of R&D expenditure has increased substantially since the 1990s in most of the large R&D-intensive multinational corporations (MNCs). Zander (1994) found that in 1990, 40% of all technological activities of a sample of Swedish MNCs were carried out abroad, as compared to only 30% in 1980. Kuemmerle (1999) also found that 32 large MNCs in the pharmaceutical and electronics industries in his survey undertook 25.8% of their R&D efforts outside of their home countries’ boundaries in 1995, compared to only 6.2% in 1965.
Recently, the nature of global R&D activities has evolved substantially in many leading MNCs from traditional “home-base exploiting (HBE)” ones to “home-base augmenting (HBA)” ones (Kuemmerle, 1999). In other words, an increasing number of overseas R&D labs have begun to explore new knowledge from host locations and even globally beyond their traditional roles, by exploiting and extending their existing technologies originally developed in their home countries (Birkinshaw et al., 1998 and Cantwell and Mudambi, 2005). By acquiring new knowledge abroad, these home-base augmenting labs help MNCs develop technologies and products to serve not only the host market but also the home and the global markets (Ambos et al., 2006).
This trend is salient even among Japanese MNCs, which are latecomers to R&D globalization compared to U.S. and European firms (Asakawa, 2001a). Although the proportion of R&D internationalization by Japanese firms is minimal (Pearce, 1989 and Cantwell and Zhang, 2006) and the share of foreign affiliates in industrial R&D remains at less than 5% for Japan (OECD, 2007), Japan's overseas R&D ratio (local R&D expense/domestic R&D expense) increased from 2.9% in 1997 to 4.1% in 2002 (METI, 2003 and Nomura Research Institute, 2005).
As MNCs increase their global R&D efforts and expand the roles of their overseas R&D labs, globalization of R&D has drawn growing attention from both academic scholars and practitioners. However, existing studies in this stream have focused largely on economic and political aspects of R&D globalization, and thus they offer little insight into how to manage overseas R&D activities. Penner-Hahn and Shaver (2005) contend that, despite the burgeoning literature that enjoins firms to globalize their R&D in order to access new technologies, we know little about the conditions that induce MNCs to do so. In addition, recent research on global R&D activities has largely “missed the opportunity for theoretical advancement that might arise from drawing upon more general theories of innovation and technological progress in organizations” (Frost, 2001: 101). Few studies have investigated mechanisms affecting knowledge acquisition, development, and transfer in global R&D activities (Ambos et al., 2006 and Frost and Zhou, 2005).
In this paper, we seek to advance the study of global R&D activities by proposing and testing a model of how overseas R&D labs of MNCs source knowledge from host locations. To take a more theoretical and balanced perspective, we draw on both the capability perspective from evolutionary economics and the embeddedness perspective from organizational theory. Based on this multi-disciplinary framework, we develop hypotheses regarding overseas lab-level characteristics that influence the sourcing of knowledge from host locations.
Focusing on the HBA type of overseas R&D labs of Japanese multinationals, we collect data on these lab characteristics from both lab-level surveys and U.S. patent (citation) data, which we use to trace knowledge flows from host locations to overseas labs. We then employ negative binomial regressions to investigate factors influencing the level of knowledge sourced from host locations. Statistical results support our main hypotheses regarding both lab capabilities and external embeddedness in host locations.
