Abstract: This paper aims to identify genuine technological spillovers from multinational firms (MNEs). To this end, we use data on R&D from MNEs to measure spillovers, while most of the existing literature uses output to measure the foreign presence in an industry (what we call output-based spillovers). In line with the existing literature, we distinguish between horizontal spillovers (i.e., intra-industry linkages) and vertical spillovers (i.e., backward ?or downstream? and forward ?or upstream? inter-industry linkages). Our results show that the three types of technological spillovers from MNEs are positive, with the horizontal spillovers the larger ones, followed by
backward spillovers. The effect of forward spillovers is much smaller in magnitude. Moreover, we find that not controlling for industry size (i.e., technological spillovers from all firms in an industry) leads to underestimating both horizontal and backward spillovers from MNEs, and to overestimating forward spillovers from MNEs. Finally, we find that the distinction between technological and output-based spillovers is of great relevance. The size of backward technological spillovers is approximately 44% of the size of output-based backward spillovers, while for horizontal spillovers both types of spillovers are quite similar. Importantly, output-based
forward spillovers are negative while technological forward spillovers are positive.
