5. Discussion and conclusion
5.1. Empirical contributions to literature This work provides empirical evidence on the effect of a funding programtargeting academic–industrypartnerships onfirminnovative performance. To our surprise, we found no significant positive effect of funding on the full sample of qualitatively similar firms, especially for patents, as Kaiser and Kuhn (2012) documented an increase in filed patents in a similar program. We posit that the discrepancy in findings may be due to differences in how patents are measured as well as the counterfactual sample of comparison. In our design, we used patents filed up to 5 years after funding grants, whereas Kaiser and Kuhn use applications of patents that may or may not have been issued. Unless all filed patents are granted (which is relatively uncommon), firm’s filed patents will be greater than granted patents. Moreover, we used in our counterfactual sample firms that were denied funding but still submitted a comprehensive proposal, whereas Kaiser and Kuhn employed a matched nearest neighbor sample of firms similar to funded ones based on observables. Our points of comparison are different, since we compare our funded firms to firms that already had the intention of pursuing the proposed project, while Kaiser and Kuhn compare their funded firms that may or may not have had the intention to pursue an R&D project.