Threshold E ects From Absorptive Capacity and the E ectiveness of Innovation Policy

In this paper, absorptive capacity e ects of human capital on innovation are introduced into the endogenous growth model. In order to empirically test their relevance, a spatial endogenous sample splitting estimator is proposed. With the help of the proposed model and estimation strategy, we are able to empirically demonstrate the relevance of absorptive capacity e ects on innovation controlling for the direct impact of human capital. Accounting for threshold e ects sheds lights on relevant absorptive capacity-induced di erences in R&D productivity patterns across Europe which would be masked by standard estimation procedures. Moreover, absorptive capacity signi cantly a ects the geographical impact and the relative magnitude of the spatial e ects from regional R&D policy. These results suggest consequences for the economic viability of R&D in lagging areas and the spatial con guration of relative regional development