Optimal R&D Intensity: Welfare Analysis

Not only has spending on research in Korea been growing in real terms, but its relative importance compared to other economic activities has been steadily increasing as well in Korea. In this paper, we introduce the socially optimal share of R&D in the economic growth model. We consider both laissez-faire(market equilibrium) steady-state growth rate and the social optimum. Because individuals and individual firms do not internalize the effect of individual knowledge accumulation when optimizing on consumption and capital accumulation, the equilibrium growth rate may be less than the socially optimal rate of growth. We develop the model(Jones and Williams, 2000) that find the decentralized economy typically underinvests in R&D relative to what is socially optimum. We also examine how far deviates the steady-state growth rate from social optimum in Korean economy. We focus here on the positive external effects of R&D. According to ordinary LS analysis, the estimate of X (the elasticity of knowledge production to R&D personnel) is 0.1559. So, we can see the optimal R&D share in Korea is 3.32%. This implies that the ratio of optimal share to market share of R&D in Korea is 1.8. According to Johansen(1991)'s cointegration analysis, the estimate of X ranges from 0.294 to 0.365. So, we can see that the lower bound for the optimal share of R&D in Korea, is 3.74%. In brief, from all cases of the form for the knowledge production function, we could decide that both the intertemporal-spillover and appropriability effects in Korean economy tend to make the average growth rate less than optimal. Because the business-stealing effect that tends to make it greater than optimal conflicts and is dominated by the former two effects, the laissez-faire average growth rate in Korea may be less than optimal.