The Effects of National and International Interaction on Innovation: Evidence from the Irish CIS: 2004–06

Abstract

This paper analyses the importance of decisions to interact nationally and internationally for the likelihood of process and product innovation in a sample of Irish firms. The key contribution is to provide an empirical test of the relative importance of geographically proximate versus distant interaction, using a two-step procedure to remove potential endogeneity in interaction decisions. In doing so it finds that only national and only international interaction have the expected positive effects on the probability of innovation, while engaging in both national and international interaction has no effect. The findings support hypotheses on the importance of both geographically proximate and distant interaction for innovation, though the lack of significance for both national and international interaction means there is no evidence to support the proposition that these forms of interaction are complementary.

Keywords:

• Geography
• interaction
• innovation

Notes

¹ By definition NUTS2 regions must contain a population of between 800,000 and 3,000,000 people (EuroStat, 2010a). The combined population of Ireland's two NUTS2 regions is approximately 4,200,000 in 2006, only slightly in excess of this upper limit (Central Statistics Office, 2010). Further to this, the combined area of Ireland is approximately 6,900 square kilometers, which is only 100 square kilometers greater than that of the Cumbria NUTS2 region in the UK (EuroStat, 2010b). This evidence justifies the identification of national interaction as being geographically proximate.

² Product and process innovation are not the sole forms of innovation a firm can implement. The Oslo Manual identifies four types of innovation: product, process, marketing and organizational innovation (OECD, 2005). While each of these kinds of innovation is important for firm performance the Irish CIS only relates R&D and external interaction to product and process innovation. This limits this analysis to the consideration of these two forms of innovation.

³ Further to this, interaction between firms and other group companies is excluded as this form of interaction is exclusive to non-Irish owned enterprises and is predominately international in scope.

⁴ The NACE Rev 1 codes selected are: High-Tech Manufacturing (24, 29, 30–35); All Other Manufacturing (10–14; 15–37 excluding high-tech, 40–41), Wholesale, Transport, Storage and Communication (51, 60–64), Financial Intermediation (65–67) and Computer, Architecture and Engineering Services (72, 74.2, 74.3). The definition of high-technology is taken from the OECD classification (European Commission, 2003). Forfás consider Computer, Architecture and Engineering Services as part of Ireland's ICT sector (National Competitiveness Council, 2009). For a fuller discussion see Jordan and O'Leary (2005, 2008).