We examine how the ways in which firms geographically configure their global portfolios of R&D units influence the effectiveness of firms’ own R&D investments and of external technical knowledge in enhancing firm performance. Our analysis indicates that the strength of these effects depends on the extent to which firms spread their R&D units across countries (geographic dispersion of R&D) and the extent to which firms establish multiple R&D units within each country (co-location of R&D). We show that geographic dispersion and co-location are associated with distinct value creation and value capture mechanisms and in turn lead to different performance outcomes. Although geographic dispersion enhances the effects of a firm's own R&D on its performance, R&D co-location limits such effects. These relationships are reversed when we consider the effects of external technical knowledge on firm performance. R&D co-location, rather than geographic dispersion, is what renders the exploitation of external knowledge more effective in enhancing firm performance. Our results suggest that future research should shift its focus from the degree of R&D globalization to how a portfolio is globalized and geographically structured.