This paper analyses the role of the composition of the regional stock of knowledge in explaining innovation performance. The paper provides three main contributions. First, it investigates the relevance of Jacobs knowledge externalities in characterizing the technological capabilities at the regional level. Second, it applies the Hidalgo-Hausmann
(HH) methodology to analyze knowledge composition by looking at patent data of 214
regions, located in 27 state members of the European Union (EU) during the years 1994-
2008. Third, it econometrically assesses the role of knowledge base composition in a
knowledge generation function. The results of the empirical analysis confirm that the
characterization of regional knowledge base through the HH indicators provides interesting information to understanding its composition and to qualify it as a provider of the Jacobs knowledge externalities that account for the dynamics of regional innovative performance