Diversity and intensity of information and communication technologies use and product innovation: evidence from Chilean micro-data

Abstract

This study uses data from two waves of the Encuesta Longitudinal de Empresas (ELE) to examine the relationship between information and communication technologies (ICT) use and product innovation in Chilean firms. Our findings sustain the hypothesis that ICT act as enablers of innovation. However, the impacts of ICT on product innovation depend on the type of application considered. In particular, we find positive and significant association between production-integrating ICT, i.e. administrative and industry-specific software, and product innovation, while this is not the case for market-oriented ICT such as e-commerce or client relationship manager software. Finally, the results show that not every ICT combination is beneficial for innovation: firms that show a basic use of ICT are not associated with a better likelihood of introducing innovation, while firms with an advanced use of ICT are those with the more likelihood of innovating.

Keywords:

• innovation
• ICT
• micro-data
• probit model

JEL Classification:

• D22
• O30
• C26

Acknowledgements

I am infinitely indebted to Giovanni Stumpo and Omar Bello for research guidance and precious discussions. I would also like to thank Leandro Cabello, Tommaso Ciarli, Sebastián Fleitas, Rodolfo Lauterbach and two anonymous referees for helpful commentaries. Thanks are also due to the participants of MEIDE Conference in Santiago de Chile in November 2013 and of the KID Summer School in Nice in July 2014. This work was part of a research project funded by the Production, Productivity and Management Division, Economic Commission for Latin America and the Caribbean (ECLAC). The contents are sole responsibilities of the author.

Notes

1. It corresponds approximately to 36,000 USD (1 45 USD). Firm sizes are stratified according to the following segmentation: micro (800–2400 UF), small (2400–25,000 UF), medium (25,000–100,000 UF) and large (100,000 or more).

2. Variable definitions and summary statistics for our sample are reported, respectively, in Tables A1 and A2 in the appendix.

3. To check the robustness of our results we run the regressions without using the prior innovation variable. The results regarding the positive and significant association between production-integrating ICT are confirmed, although the marginal effects are higher when not controlling for past innovation. In the case of market-oriented ICT, the only application showing positive and significant results is CRM. This result implies that the software was adopted only because the firm is already innovative. Tables of the regressions performed as robustness check are available upon requests.