Page 42 - IDEA Study 8 2017 Direct subsidies and R&D output in firms

bold research plans due to a fear of not obtaining the grant of a patent in time to fulfil the formal project evaluation requirements. It should be emphasized that the results for ALFA and in part for TIP are preliminary, because the data on applications for IP protection is incomplete due to the delay in their publishing process. In addition, the fourth and last call of ALFA could not be included in the analysis at all, because of the limited availability of IP protection data in recent years and because some of the supported projects are still running. Hence, it will only be possible to estimate the overall output additionality effects of ALFA using this methodology in several years’ time. Moreover, so far we have only focused on the programmes’ immediate short-term impacts, while what truly matters is whether the programmes deliver lasting long-term impacts on the participating firms’ innovation behaviour. As already noted, furthermore, the data limitations mean that we are not able to distinguish whether these R&D subsidies stimulated the generation of new technology that would not otherwise have been generated or whether they only stimulated firms to use IP instruments to protect existing or partly existing technology that would not otherwise have been formally protected. Either way, the subsidies had a “crowd-in” effect that should be seen as a positive impact, since there is evidence that formal instruments of IP protection remain underused in the Czech Republic (Heilemann 2014 and TA CR 2015). One positive effect of formal IP protection is that inventions are coded and disclosed to the market and can be accessed by others. As a result, those inventions are added to the knowledge pool and stimulate subsequent innovations. Even if the programmes only stimulated IP protection, which might be perhaps seen as an unintended positive impact, thelearning involved has the potential to improve the efficiency of the firms’ innovation activities in the future. Finally, a major limitation of our analysis is that it only considers one type of R&D output: instruments of formal IP protection. In practical terms, using this data has major analytical advantages, because the instruments are clearly defined, based on external review criteria implemented by patent offices, and because harmonized data is available both for the subsidy recipients and a control group. Admittedly, however, there are other types of R&D outputs, such as non-patented inventions, new products introduced to the market and process innovations implemented in practice, which 40