Following Zellner and Revankar (1970), we estimated generalized Cobb-Douglas production function. The model allows economies of scale to vary and to increase and then decrease as output rises. The reason is that we should not overestimate the contribu...
Following Zellner and Revankar (1970), we estimated generalized Cobb-Douglas production function. The model allows economies of scale to vary and to increase and then decrease as output rises. The reason is that we should not overestimate the contribution of technical progress in growth accounting exercises.
We can summarize the results of this study. First, though they are relatively small in magnitude, it is apparent that there has been economies of scale in Korean economy, and it is smoothly increasing. Second, we notice that there are long-run equlibrium relationship between scale economies, productivity(level and growth), and research expenditures(input). This implies that we have to consider returns to scale in growth accounting method. Third, we estimated VEC(vector error correction) model, and the results show that the error correction coefficient in TFP adjustment equation is of the appropriate positive sign. But, overall results show that vector error correction system is not stable. Finally, we performed growth accounting to decompose the effect of scale economies from the (estimated) Solow residual. We could misleadingly overestimate the contribution of technological progress, if we omit the externalities from economies of scale.