A class of spatial econometric methods in the empirical analysis of clusters of firms in the space

In this paper we aim at identifying stylized facts in order to suggest adequate models for the co-agglomeration of industries in space. We describe a class of spatial statistical methods for the empirical analysis of spatial clusters. The main innovation of the paper consists in considering clustering for bivariate (rather than univariate) distributions. This allows uncovering co-agglomeration and repulsion phenomena between the different sectors. Furthermore we present empirical evidence on the pair-wise intra-sectoral spatial distribution of patents in Italy in 1990s. We identify some distinctive joint patterns of location between different sectors and we propose some possible economic interpretations. A previous version of this paper was presented at the Workshop on Spatial Econometrics and Statistics, held in Rome 25–27 May 2006. We wish to thank the participants for the useful comments received. The comments received by two anonymous referees are also gratefully acknowledged. They improved substantially the quality of our work.