Innovation and productivity in a S&T intensive sector: the case of Information industries in Spain

This paper adds to the empirical literature on the relationships between R&D, innovation and productivity at the firm level. The focus is on Spanish enterprises in Information industries, which are acknowledged to be at the forefront for both innovative activity and R&D performance. The analysis is performed on ca. 1800 enterprises included in the PITEC database (the Spanish source of the EU Community Innovation Survey) for the period 2004-2013. Using a three-stage "CDM" model we consider: (i) factors affecting the decision to conduct R&D, including the role of perceived importance of innovation on firm's R&D performance, (ii) the impact of the predicted R&D effort on companies' effective undertaking of product, process, organisational and marketing innovations, as well as their simultaneous occurrence and (iii) whether and to what extent such innovations boost productivity. In the specific context of this R&D intensive array of industries, the decision of undertaking R&D appears to be strongly influenced by the importance attributed to enhancing existing products or creating new ones, as well as by the size of the company, the fact of being young and local, and the availability public funding. These elements also greatly impact on enterprises' R&D effort, thus providing some arguments in favour of R&D promotion policies, in particularly addressed to start-ups. Expected R&D performance, in turn, appears to be strongly related to the actual achievement of such innovations, including non-technical ones. By focusing on innovation patterns, it was possible to ascertain a strong complementarity between different families of innovation (as expected, given these industries' specificities), as well as to qualify existing evidence on the innovation-productivity conundrum. Indeed, we show that results depend on the way innovation types are modelled and combined. Controlling for complementarities, enterprises performing focused non-technical innovations and joint technical and non-technical ones (mixed-mode innovators) are likely to be more productive (in terms of sales per capita) than their peers, while stand-alone technical innovations give inconclusive results.