Abstract:

Universities contribute to economic development and shape innovation system. This paper investigates how the cooperation between firm and university affects innovation output of firms. Previous research highlights the importance of science–industry interaction and show that cooperation increases the chances of the firms to innovate. The investigation better qualifies this relationship showing that cooperation increase innovativeness of firm enhancing radical innovation (i.e. new-for-the-market and new-for-the-world) rather than imitative innovation (new-for-the-firm). Small and Medium Enterprises benefit more than large firms from the cooperation with universities. The research was conducted using the Manheim Innovation Panel (MIP) that is the German contribution to the European Community Innovation Survey (CIS). Unlike previous studies that are generally based on cross-section data, my investigation uses panel data techniques and instrumental variable to account for unobserved time-invariant heterogeneity and potential endogeneity. The use of secrecy as strategy to protect innovation is used to instrument the cooperation variable. The rationale is that secrecy influences cooperation propensity but not systematically innovation success. The endogeneity tests based on instrumental variable suggest that endogeneity is not a major issue in the data. Results in cross section and in panels do not differ substantially. As additional check, the implementation of synthetic (or pseudo) panel (Deaton,1985) for innovation firm level studies is explored. The methods requires to group observation according to characteristics that do not vary along time. In individuals’ studies, usually the grouping is done on the basis of sex and age. In the case of firms, only the economic sector can be used reducing considerably the sample size and the stability of estimates.