Page 32 - IDEA Studie 07 2023 TACR

ARE SUBSIDIES TO BUSINESS R&D EFFECTIVE? REGRESSION
DISCONTINUITY EVIDENCE FROM THE TA CR ALFA PROGRAMME IDEA 2023
with the additionality (or motivation) effect of the funding, as sometimes happens with stake holders and popular commentators.
Moreover, the devil is in the details. Our results show that understanding of the pro- gramme effects significantly deepens when analysis is done not only of the total R&D expenditure, which none of the previous RD studies on this topic has been able to do, but also on their detailed sub-categories by the source of R&D funding and the types of R&D costs. Studies that rely solely on indirect proxies of R&D inputs and/or measures of general economic performance to evaluate R&D subsidy programmes might miss the most salient part of the picture. For example, this study would not find any significant results in the absence of information on R&D expenditure. Studies of the effects of R&D subsidies should exploit the full advantage of data from R&D surveys, which is readily collected in fine detail in most advanced countries.
From a methodological perspective, the RD approach should be applied to other R&D subsidy programmes, whenever data and the underlying design of the instrument allows for it, and should become a standard in programme evaluation in the coming years. Admittedly, a major hindrance to this line of enquiry has been limited access to administrative microdata from evaluation of projects proposals from within the funding providers – the evaluation points and/or ranking of the proposals – which is necessary for RD analysis. In the Czech Republic, TA CR should be praised for beginning to provide this data on a regular basis, although other funders, most notably the Ministry of Industry and Trade, are yet to follow suit. Czech policymakers can derive important lessons from rigorous quantitative evaluations of these programmes to facilitate policy learning and future policy design.
A major limitation of this analysis that should be reiterated is that it only considers direct effects of subsidies on recipient firms. Assessments of subsidy programmes should also consider their spillover effects on unsupported firms. After all, the government should not subsidize business R&D with the narrow aim to help only that particular firm, but rather with an eye to maximize the desired social impacts, and to achieve positive externalities through wider knowledge spillovers. Until we manage to pin down the extent of knowledge spillovers from subsidised projects, which are indeed notoriously difficult to measure, the jury is still out on whether a subsidy programme contributed to the development of the economy as a whole.
30