This study aims at assessing the level of efficiency of public R&D spending and public R&D support for private R&D and to compare efficiency scores among OECD countries, in particular EU Member states over the past two decades. The analysis rests on the concept of efficiency which is based on the relationship between public R&D spending and the additional R&D in the business sector induced by such measures. The differences observed in efficiency performance across countries is then explained in relation to control variables, e.g. exogenous framework conditions. Given the different specifications considered and the two methods implemented, i.e. Stochastic Frontier and non-parametric Data Envelopment Analysis, results inevitably diverge and in some cases to a large extent. Some results are nevertheless robust and consistent through the different methods and model specifications. For the efficiency scores, three groups of countries emerge. The first group refers to the most efficient countries in terms of R&D public support, namely Australia, Canada, Finland, Germany, Japan, Netherlands, New Zealand, Singapore, Switzerland, and the USA. The second group is composed by France, Hungary, Italy, Korea, Norway, Sweden, Spain and the United Kingdom. The third group of the less efficient countries in terms of R&D public spending includes China, Croatia, Czech Republic, Israel, Latvia, Poland, Romania, Russia, and the Slovak Republic. In terms of the determinants explaining these performances, conditional efficiency scores are found to be higher in the most industrialized countries worldwide. In terms of GDP per capita, countries in an intermediate position are characterized by lower efficiency scores of their public R&D support funding. Unsurprisingly countries with the best performance in terms of innovative activities are also the ones that exhibit the highest efficiencies of their public R&D support, while higher government expenditures in percentage of total consumption are associated with lower performance in terms of efficiency. In terms of industrial structure, economies with a higher share of high-tech manufacturing sectors in the total manufacturing value added benefit from higher efficiency performance of their public R&D supporting policy instruments while economic growth does not seem to affect efficiency scores. In terms of regulatory and macroeconomic conditions, a positive and significant impact is observed on the efficiency scores of a more favourable tax regime to international trade, of more deregulation in the labour and business markets as well as of the strength of the Intellectual Property Right System. On the other hand, access to sound money does not appear to positively affect public R&D efficiency. The share of the general government consumption in percentage of total consumption negatively affects the efficiency scores and as expected countries with high inflation rates appear to be less efficient. Finally, there seems to be no unique public strategy that determines high efficiency levels and more general conclusions about "optimal" policy mixes with respect to public R&D would require to go beyond the pure macroeconomic level.
