This paper explores the spatial patterns of Italian high-tech affiliates in the U.S. using a unique international data source at the firm level. The paper uses new statistical tools for spatial analysis - local spatial correlation indicators (LISA) - and compares these with traditional geographical concentration and global spatial correlation indicators. LISA indicators enable us to spot heterogeneity in location patterns, as well as different agglomeration profiles at the local level, in the location of Italian affiliates in the United States depending upon the technology intensity of the industry or service sectors. Bivariate LISA analysis is also used to assess whether Italian high-tech affiliates in the U.S. tend to agglomerate around world-leading American universities that specialise in advanced technologies. Further, the empirical results are related to the relevant literature on the location of R&D activities and, in particular, to knowledge-sourcing FDI motivation.
Keywords: Multinational enterprise, firm location choice, high-technology industries and services, knowledge sourcing
Jel Code: R3, F23, R12, C21