A better estimate to the contribution rate of education on economic growth in China from 1999 to 2003

摘要

The traditional methods of estimating economic contribution rate of education (ECRE) are based on hard computing such as statistical methods, which ignore both the long-term effect of education and the lagged effect of education on economy growth. This paper proposes the fusion method of neural networks, fuzzy systems and genetic algorithms made in the realm of soft computing to estimate the ECRE. Firstly, a target system (a country or a region) is categorized softly according to the level of Science and Technology (S&T) progress. Secondly, potential human capital stock and actual human capital stock in the same cluster are calculated and set up the internal correlation between them (fuzzy mapping). Thirdly, we conceptualize actual human capital as one production factor, joined with the other two production factors, land and fixed asset, to set up the fuzzy mapping to economic growth. Finally, we obtain the ECRE through two marginal rates, namely marginal economic growth to actual human capital stock, and marginal actual human capital to potential human capital. This method greatly reduces the bias in the ECRE that results from the indirect and lagged effects of education. It therefore identifies the effect of education on economic growth more explicitly. Based on the level of S&T progress, 31 provinces in China could be classified into three clusters. The first cluster (developed S&T) has an ECRE of 11.60%, and contains two provinces; the second cluster (developing S&T) has an ECRE of 8.82%, and contains 11 provinces; the third cluster (underdeveloped S&T) has an ECRE of 1.49% and contains 18 provinces.