The role of research and development (R&D) as an engine for technological progress is well documented. For small and open economies spillover effects of foreign R&D appear to be the dominant source of productivity growth (Coe and Helpman; 1995 and Keller; 2004). There is, however, evidence that domestic R&D might be crucial for the capacity of firms to absorb the knowledge developed abroad, i.e. own R&D has a second face (Griffith et al., 2004; Keller, 2004; Kinoshita, 2000). Along with R&D investments, a positive productivity effect of human capital investments is also empirically supported. Benhabib and Spiegel (1994) find more explanatory power in the human capital level than in its change. The human capital level serves, they argue, as a facilitator of technological dispersion. This implies that not only accumulation of R&D knowledge, but also that of human capital, has a second face. This hypothesis is supported by later studies that include both R&D knowledge and human capital as absorptive capacity determinants (Griffith et al., 2004; Crespo et al., 2004). This applied endogenous growth analysis explores how a small and open economy should form its policy strategies regarding education, innovation, and trade for stimulating long-term productivity growth. We analyse the impact of three different policy instruments, whilst simultaneously taking into account the two faces of R&D as well as education. The first is a general subsidy to R&D, which generates external standing-on-shoulders and productivity effects as in Romer (1990) and increases the absorptive capacity of industries using R&D-intensive technology. We compare such a stimulus with a similar amount of resources devoted to stimulate trade in R&D-intensive technology. This will also affect both domestic innovation and knowledge externalities, as well as spillovers from abroad. The third policy experiment is to devote the similar amount to public investments in education that increase the number of high skilled and the level of human capital. Besides enhancing labour productivity directly, productivity is improved through strengthened absorptive capacity. We use an endogenous growth model calibrated for a small, open economy. Growth processes take place through domestic innovation and R&D externalities as in Romer (1990), as well as through industry-specific absorption of knowledge spillovers abroad, which is endogenously affected by each industry’s trade (imports and exports), use of human capital, and R&D intensity. The contribution of the present paper is to study a large number of empirically evident mechanisms in interplay. In contrast, several earlier contributions study endogenous growth through domestic R&D, but leave no role for international knowledge absorption, which is an important growth channel for small, open economies. Diao et al. (2006) model productivity growth through international trading, and Hübler (2010) also includes absorptive capacity effects of human capital, but neither allow for domestic innovation. Diao et al. (1999) and Ghosh (2007) adapt the Romer model to the small open economy case by accounting for both innovation and absorption of international spillovers, but only through the R&D industry. Bye et al. (2008) come closest to the present model, by modelling industry-specific absorptive capacities, but the two-fold role of human capital is not modelled. In the present analysis, industries vary according to their specific trade, skill, and R&D intensities. We include import and export activities as indicative for the trade openness. Most studies focus on the import channel, only, but new empirical evidence supports that exports has a similar role. Most studies focus on the import channel, only. However, new empirical evidence also supports exports as a determinant. A CGE approach can simultaneously grasp the many direct and indirect interdependencies of R&D, education, and trade. It will also account for these activities’ interplays with other market imperfections and existing government interventions. Our preliminary results indicate approximately similar welfare effects of all the policy alternatives. But the industrial pattern and the relative roles of imitation and innovation processes for productivity growth and welfare differ. An R&D subsidy causes the R&D activity to expand considerably. This brings about positive externalities both through internal knowledge spillovers and productivity gains for users of R&D based capital. Absorption of spillovers from international technological development is also facilitated by the domestic dispersion of R&D based capital. Many final goods industries reduce their use of high skills as a result of the expansion of the R&D industry, thus, their absorptive capacity through human capital deteriorates. The export promotion policy has the intended outcome of increasing absorption of knowledge across borders in the R&D industry more than does the R&D policy. However, this comes at the expense of absorption in most other industries. Education policies have a direct positive effect on R&D activity because this sector is the far most human capital intensive. The R&D expansion comes less at the expense of other human capital intensive industries compared to the other policy alternatives.
