Matching methods for impact evaluation of public subsidies to business R&D: Measuring heterogeneous effects

The objective of this paper is to offer a broad profile of firms with publicly supported R&D projects, which allows us to explain their different degrees of additionality. With this objective, in a first step we use standard Propensity Score Matching techniques to estimate treatment effects at the firm level, and then we explore the determinants of the heterogeneity in these individual effects through the estimation of an equation for their determinants. For our analysis, we use information from a sample of 8,168 Spanish firms for the period 2007-2014. We report three main results. First, firms with multiple program participation show higher additionality. However, individual treatment effects, which are positive for firms with low support intensities, go sharply below the average for firms with very high support intensities. Second, the degree of additionality is positively related to firm characteristics denoting a more innovative nature, while it is negatively associated with features present in firms involved in more market-oriented R&D projects. Third, firm size has a positive relation to the probability of full additionality, but a negative association with the degree of additionality in terms of net R&D intensity. These results can provide public agencies with some tools for adjusting their selection procedures