R&D, innovation activity, and the use of external numerical flexibility

We address theoretically and empirically the impact of R&D and innovation activity (IA) on the use of external numerical flexibility (ENF). We build a firm-sided model showing that a first-order stochastic dominance shift in the productivity distribution function decreases the probability of hiring workers with temporary contracts, while a second-order shift has ambiguous effects. Next, using a dataset based on a survey of Italian manufacturing firms, we find that R&D and IA increase the extensive and intensive margins of employing workers with temporary contracts. Moreover, we disentangle the impact of different types of R&D and IA, finding that extra muros R&D always has a positive effect, while the effect of intra muros R&D is generally null. Also the effect of IA changes according to the type of activity: positive with product innovation, null with process innovation.