Proximity and Innovation in Italian SMEs

Abstract: In this paper we assess the relevance of both knowledge creation and diffusion processes in affecting Italian SMEs’ propensity to innovate. In doing so a knowledge production function (KPF) is estimated for a representative sample of small and medium manufacturing firms over the period 1998-2003. To account for endogeneity of R&D effort in the KPF, we estimate a Heckman selection model on R&D decisions and obtain two main results. First, we do not find the probability of being engaged in intramural R&D activities to be significantly related to firm size. Second, for those firms engaged in R&D activities, the intensity of R&D effort increases with firm size. Then, the KPF is estimated for three different samples of firms using a standard probit where the probability that SMEs will innovate depends upon intramural R&D effort, regional and industrial spillovers and a vector of interaction and control variables. The main results obtained from this second set of regressions are the following: first, we find the probability to innovate to be positively related to sectoral spillovers, the magnitude of such impact being decreasing in firms’ size. Second, knowledge diffusion via geographical proximity enhances the probability of the recipient firm to innovate only if it has an appropriate endowment of human capital.