The creation, integration and retrieval of knowledge across borders

The geographic landscape of innovation has changed in recent years, embodying a larger number of countries that are dispersed across the globe. With this change, multinational corporations (MNCs) began to more aggressively expand their global presence in order to access complementary, and sometimes even superior, resources that are context-specific, and embedded in distant nations. In this thesis, I focus on the activities of subsidiaries located outside home-country borders to examine how the MNC creates, integrates and retrieves knowledge across borders. Four empirical studies constitute this thesis, where each follows on from the previous study in a sequential manner. In all four studies, the data used consists of a sample of 26,708 patents published by the United States Patent and Trademark Office (USPTO) to 1,022 foreign subsidiaries belonging to one of 238 semiconductor MNCs headquartered in the United States. These data span the time period 1980-2005, and are assigned to subsidiaries located in 42 countries, of which 15 are considered to be emerging market economies. The first study explores why cross-country collaboration on R&D with home-country inventors enhances the value of innovations that are conducted offshore. The results show that, in terms of the overall value, emerging economy subsidiaries benefit from increasing home-country involvement. In advanced economies, however, patents with the highest value are attained when the host-country and home-country inventors contribute equally to the innovation. As for the value that these R&D endeavours create for the MNC, the most valuable patents are attained when the home-country and the host-country inventors contribute equally, both in advanced and in emerging economy subsidiaries. The second study breaks down the value of innovations into two components, the value they create internally, within the MNC, and the value that they create to external firms in the host-country. The results show that innovations developed entirely in the emerging economy subsidiaries are less likely to diffuse in the MNC network, and they are also less likely to diffuse externally, in the host-country. However, for innovations which feature cross-country collaboration, the barriers which prevent internal diffusion can be overcome while, simultaneously, reducing knowledge diffusion in the host-country. The third study explores the long-term effect of cross-country collaboration on the performance of foreign inventors, and shows that it does indeed have a long-term effect, allowing them to generate high-impact patents and explore new technologies on their own. These positive effects are even more pronounced for inventors from emerging economy subsidiaries. However, the new knowledge that is generated in this way is unlikely to be integrated by the offshore subsidiaries to form new capabilities. Possible explanations are provided in the corresponding chapter. The fourth study examines two aspects of knowledge spillovers: how they can be mitigated and, if they do occur, how lost knowledge can be retrieved. Firms deploy certain strategies to prevent knowledge from spilling over. Some strategies – although they can be successful at curbing spillovers – can also make it increasingly more difficult for MNCs to retrieve knowledge after spillovers. The results show that at moderate levels of knowledge complexity coupled with wide geographic dispersion of its underlying elements can maximize the gap between internal and external appropriation. However, the relationship between the two is delicate, as minor variations could increase the rates of spillovers and decrease the rates of knowledge retrieval