Attempts to measure the spillover effects of multinational enterprises on host countries have generally been cross-sectional and limited to labour productivity in manufacturing for a single country. Recent work in growth theory has measured the extent to which growth in total factor productivity in a country depends not only on domestic R&D capital stocks but also on foreign R&D capital stocks. This paper extends such work by adding foreign direct investment stocks to foreign trade as a channel linking total factor productivity levels between countries. This is done by considering the role of trade and FDI as diffusion channels for G6 R&D to the OECD countries. There are three main results: the coefficient estimates for FDI are higher than those for trade in the standard model; the importance of the trade channel is much reduced once FDI is included; and the overall spillovers increase significantly with the inclusion of FDI.

Changing patterns of foreign direct investment (FDI) stock have raised important questions about their impact on domestic economies. For example, it is often thought that increased inward FDI contributes to domestic capital formation, whereas increased outward FDI reduces it. We demonstrate that such generalizations are inappropriate. We develop hypotheses linking the impact of FDI to the underlying motivation for investment. These hypotheses are tested using available Canadian industry-level data. The implication of our results is that rapid growth in outward FDI, relative to inward growth, should not be considered as a negative development, and may reflect success. Journal of International Business Studies (2003) 34, 282–289. doi:10.1057/palgrave.jibs.8400030

No abstract

and Key Results■ Several recent papers by Rugman and his colleagues have improved our understanding of the regional concentration in the activities of the world's largest 500 MNEs. The current paper extends this literature in two dimensions. First, a formal statistical analysis is undertaken to test whether patterns of US MNE assets, sales, income, and employment are consistent with a transactions cost interpretation. Second, this paper allows for the national dimension, defined as activities inside the home country, to be a possible explanation of regional concentrations of MNE activity.