In this paper we evaluate the effectiveness of r&d tax incentives in quebec, using manufacturing firm data from 1997 to 2003 originating from r&d surveys, annual surveys of manufactures and administrative data. The estimated price elasticity of r&d is –0.10 in the short run and –0.14 in the long run, with slightly higher elasticities for small firms than for large firms. We show that there is a deadweight loss associated with level-based r&d tax incentives that is particularly acute for large firms. For small firms it is not sizeable enough to suppress the r&d additionality, at least not for quite a number of years after the initial tax change. Incremental r&d tax credits do not suffer from this deadweight loss and are from that perspective preferable to level-based tax incentives.