運用隨機共同成本邊界函數聯合估計中國銀行業之成本效率及市場競爭度

JOINTLY ESTIMATING THE METAFRONTIER COST FUNCTION AND PRICE EQUATION TO EXAMINE THE CHINESE BANKING INDUSTRY S COST EFFICIENCY AND MARKET COMPETITIO

本文探討中國銀行業在2006年至2016年間的成本效率以及市場競爭程度，將整體銀行業分為二個群組，第一類為國有銀行，第二類為城市商業銀行。第一階段採用Huang et al. 2018）提出的方法，運用關聯結構法針對兩群組樣本，個別聯立估計隨機成本邊界函數和價格迴歸方程式。得以分別計算兩群組銀行的成本效率值和市場競爭度指標；第二階段採用Huang et al. 2014）提出之方法，合併兩群組資料後，再度聯立估計隨機共同成本邊界與共同邊際成本函數，進而計算技術缺口比率、共同邊界效率值與潛在Lerner指數。實證分析發現城市商業銀行的成本效率較國有銀行高，在競爭度方面，潛在Lerner指數與傳統Lerner指數都指向城市商業銀行有較強市場力，而與結構法支持國有銀行市場力較強的結論相反。

This paper applies the stochastic metafrontier cost function to jointly estimate the market competition and cost efficiency of the Chinese banking industry using data spanning 2006–2016. We divide Chinese banks into two groups, i.e., state-owned banks and city commercial banks. The model that is developed by Huang et al. (2018) is first applied to estimate the cost frontier and price equation in the context of copula methods. Then the metafrontier model of Huang et al. (2014) is employed to jointly estimate the stochastic metafrontier cost function and marginal cost functions in order to compute the technology gap ratios, metafrontier cost efficiency, and the potential Lerner index. The empirical results show that city commercial banks are more efficient than state-owned banks. Since the average value of the Lerner index for city commercial banks exceeds that of state-owned banks, the former are found to have a higher market power than the latter.