Selection or Self-Rejection? Applications into a Treatment Program : The Case of R&D Subsidies

We build a structural model of application and selection into a treatment program to study the effects of R&D subsidies. The investments of the participant and the public agency running the program affect the treatment outcome. Using project level data we find that larger firms have higher marginal profitability of R&D. Rates of return on R&D are high and skewly distributed. Median expected profit from applicants' R&D projects is 500,000 euros, which increased by 12,000 by expected subsidies. Firms with SME status receive higher subsidies. Conditioning on SME status, subsidies are increasing in firm size, and in technical challenge of the project. The median increase in the public agency's utility not appropriated by the applicant is 16,000. Non-applicants have lower expected subsidies and higher application costs. Application costs decrease in the number of previous applications and increase with the profitability shock. Ignoring application costs severely biases the estimated rates of return upwards