Technological Learning and Innovation Gestation Lags at the Frontier of Science: from CERN Procurement to Patent

This paper contributes to the literature on the impact of Big Science Centres on technological innovation. We exploit a unique dataset with information on CERN's procurement orders to study the collaborative innovation process between CERN and its industrial partners. After a qualitative discussion of case studies, survival and count data models are estimated; the impact of CERN procurement on suppliers' innovation is captured by the number of patent applications. The fact that firms in our sample received their first order over a long time span (1995-2008) delivers a natural partition of industrial partners into "suppliers" and "not yet suppliers". This allows estimating the impact of CERN on the hazard to file a patent for the first time and on the number of patent applications, as well as the time needed for these effects to show up. We find that a "CERN effect" does exist: being an industrial partner of CERN is associated with an increase in the hazard to file a patent for the first time and in the number of patent applications. These effects require a significant "gestation lag" in the range of five to eight years, pointing to a relatively slow process of absorption of new ideas.