The strength of science and technology drivers for SME innovation Mario Davide Parrilli • Aitziber Elola Accepted: 20 January 2011 / Published online: 15 March 2011 Ó Springer Science+Business Media, LLC. 2011 Abstract This analysis feeds into the academic debate on the most proficient innovation mode across firms, placing special emphasis on the characteristic case of small and medium enterprises (SMEs). Of the three main approaches considered, the first stresses the importance of innovation based on science and technology (STI) drivers, such as research and devel- opment (R&D) and human capital, whereas the second approach emphasises innovation based on learning by doing, by using and by interacting (DUI); the third, more recent approach is conceived as a combination of the former two (STI ? DUI). In this paper, the three models are tested on a sample of 409 SMEs that have been supported by a public programme for innovation promotion developed by the Basque Government in Spain. The result is quite different from what is expected, yet it is insightful and potentially useful for both academics and policy-makers. Contrarily to what one would expect, SME innovation output is in fact more sensitive to STI drivers than to DUI drivers. Keywords Small firms Á Innovation modes Á R&D JEL Classifications M21 Á O32 Á L26 1 Introduction The debate on competing through a ‘high road of development’ has been going on for about 20 years (Pyke and Sengenberger 1992), and many countries and regions across Europe and beyond have been designing and promoting important innovation poli- cies (OECD 1992; European Union 1998). This paper focusses on the innovation modes utilised by small and medium enterprises (SMEs) (Cooke and Wills 1999; North et al. 2001; Nauwelaers and Wintjies 2002; Audretsch 2003; Jensen et al. 2007; Lundvall and Lorenz 2010), bringing our own contribution to the debate on the most relevant innovation drivers. In this work we benefit from 409 questionnaires filled in by firms that have been supported by the Enterprise Innovates programme of the Basque Government. These questionnaires correspond to firms from the four most well-represented sectors within the set of firms supported by the programme: machine tools (NACE 29), metal products (NACE 28), paper and graphic arts (NACE 22) and consulting, engineering and informa- tion technology (NACE 72–74). This analysis, and the related research question, feeds into the theoretical debate on the most proficient innovation profile for firms and their systems that stress, on the one hand, the option of innovation based on science and technology (STI) drivers such as R&D expenditure, infrastructure and human capital (Griliches 1979; Romer 1994; Greunz 2005) and, on the other, innovation based on learning by doing, by using and by interacting (DUI) M. D. Parrilli (&) Á A. Elola Basque Institute of Competitiveness, University of Deusto, Camino Mundaiz, 50, 20012 San Sebastia ´n, Spain e-mail: m.d.parrilli@orkestra.deusto.es 123 Small Bus Econ (2012) 39:897–907 DOI 10.1007/s11187-011-9319-6