Modeling the R&D effects on the Czech economy in a CGE framework incorporating Romer’s theory of endogenous growth

In line with the EU strategy on smart and sustainable growth, there has been an increasing attention directed to Research and development activities in the Czech Republic. Gross expenditures on R&D have doubled between 2000 and 2008 and the private R&D sector remains the biggest contributor to this expansion. Furthermore, the support of the private R&D sector from public resources is one of the highest within other European countries. Based on these observations, the private R&D sector potentially represents an important source of growth and innovation in the Czech economy. Precisely due to its strategic role in the economy, it is necessary to properly quantify its effects on economic growth and to assess the efficiency of governmental support directed to private R&D sector. Following this necessity, the objective of the paper is to evaluate the impact of private R&D sector on the Czech economy by incorporating Romer’s concept of endogenous growth modeling into the existing CGE model (Romer, 1990). In concrete, the CGE model captures the effects stemming from the production of capital varieties in an imperfectly competitive market with forward looking agents. The results of this paper are part of the post-doc research grant of the Czech Science Foundation “Evaluation of Research and Development Effects on the Economic Growth of the Czech Republic with the Use of a Computable General Equilibrium model“. The assumptions of Romer’s model are highly stylized. Examples include i) the assumed existence of an intermediate capital goods sector which easily converts homogenous capital into varieties, ii) the assumption that each firm in the capital goods sector converts exactly one design (patent) into a variety, or iii) the existence of a unique R&D sector that is engaged in producing new ideas. Furthermore, Romer’s patent-based approach to valuate R&D results is not in line with the current way of R&D representation in national accounts which is based on the indicator of gross R&D expenditures. Despite these challenges, various authors have attempted to translate the endogenous growth model into the CGE framework. Perhaps the earliest contribution can be found in the work of Diao, Roe and Yeldan (1999) on Japan, which considers monopolistic competition in the sector of variety capital, and the effect of international spillovers on the productivity of the R&D sector. A more recent version of Diao, Roe and Yeldan´s approach is presented by Madanmohan Ghosh (2007) who studies the R&D effects on Canadian economy. The most recent applications in the CGE framework can be found in Bye, Fæhn and Heggedal (2009) and Bye, Jacobsen (2011) from the Norwegian statistical office. A detailed documentation to the model and its calibration is presented in Bye, Fæhn, Heggedal, Jacobsen, Strøm (2008). The methodological approach in this paper is based on the Romer’s theory of endogenous growth and follows the approaches in the recent literature. The paper builds on a recursively-dynamic CGE model that incorporates the effects of R&D investments by accumulating knowledge in the economy, developed previously by the author . In this research, following the Romer’s model of endogenous growth, the CGE model is further extended to incorporate monopolistic competition in the sector of private R&D which produces variety capital. In addition, the dynamisation of the CGE model is modified so that the agents follow a forward looking behavior characteristic for the intertemporal CGE models. In line with Romer’s concept, it is assumed that the private R&D sector represents research efforts of private businesses to produce new designs. However, as opposed to the original setting, there is no explicit distinction between the private R&D sector and the variety-capital goods sector. Following the Dixit-Stiglitz approach of modeling the production of varieties (such as Bye et al, cited above) it is assumed, that the companies involved in the private R&D operate in a monopolistic competition environment – each R&D firm produces a different design and therefore different capital variety. All firms face certain amounts of fixed costs stemming from the research efforts and they maximize their profits under a perceived elasticity of demand for varieties. The elasticity of demand is also the elasticity of substitution between different capital varieties following the Dixit-Stiglitz functional form. The public R&D sector is not involved in the production of capital varieties, but it produces general knowledge that consequently enters the production process of both public and private R&D as a specific production factor. Thus, public R&D activities directly increase total factor productivity of the public R&D sector, and they also provide positive spillovers to the private R&D sector. Besides the two R&D sectors, there are 17 final production sectors, all of which employ the new ideas produced by private R&D sector converted in new varieties. The higher the number of varieties, the higher is the capital stock and the total productivity of the final goods sector. Because the new ideas bear a non-rival feature, all production sectors employ all available capital varieties in the economy. The CGE model replicates the economy of the Czech Republic in 2008, which is formalized in the Social Accounting Matrix (SAM). The SAM was built with the use of three data sources: Czech National Accounts, publically available R&D data from Frascati surveys and (purchased) micro-level data of private R&D companies, also from the Frascati survey. The final size of the SAM is a matrix of 56x56 size. The CGE model is applied in three scenarios. The first – baseline - scenario provides the growth rates of the economy under stable conditions, which refers mainly to the governmental support policy of the private R&D sector. The second scenario analyzes the effect of an increased subsidy rate applied to the private R&D sector on the dynamics of economic growth. In concrete, the results enable to assess changes in the number of new R&D firms, the total R&D production and the GDP. In the third scenario, the efficiency of governmental support is assessed by comparing effects of various subsidy rate options between different sectors of the national economy. Consequently, the results obtained from the intertemporal dynamic model are compared to the model following recursive dynamisation and the differences in economic growth are analysed. Furthermore, the results of sensitivity tests are reported, which asses the effects of different substitution elasticities between capital varieties. The model is also tested concerning different forms of investment utility functions and their effect on R&D and economic growth.