本文的目的是探討開放新銀行後的1992-2000年間台灣新、舊銀行的表現及其決定因素。我們採用資料包絡法計算銀行的技術效率和Malmquist生產力變動指數來衡量銀行表現。主要實證結果有以下兩個。第一、開放新銀行後,新、舊銀行的技術效率差不多,但都比開放新銀行前(1987-1991年)的舊銀行顯著的低。隨機效果Tobit長期追蹤模型迴歸結果發現銀行競爭和資產規模皆不影響新、舊銀行的技術效率。專業及投資均有利於新銀行。對新、舊銀行分別而言,放款品質惡化是造成新銀行的技術效率低落的主因。由於規模效率下滑,亞洲金融風暴後舊銀行的技術效率顯著下跌。第二、根據生產力變動指數,開放新銀行後,舊銀行表現比開放新銀行前好,甚至超越同期的新銀行。動態長期追蹤模型迴歸結果發現生產力成長率有負持續性的現象。舊銀行的資產規模與生產力的關係脊在最適規模。銀行競爭其實有利於新銀行的生產力成長。新銀行的投資比舊銀行的投資更具競爭力。雖然新銀行在很多方面比舊銀行佔不少優勢,放款品質變差及亞洲金融風暴造成新銀行表現不如舊銀行。
Taiwan's banking industry has experienced dramatic structural changes since the granting of new banking licenses in 1991. The purpose of this paper is to examine bank efficiency and productivity growth over 1992-2000 that are evaluated using data envelopment analysis (DEA). Our findings are as follows. (1) Technical efficiency for old banks has fallen after 1992, and there is no significant difference between new and old bank technical efficiency over 1992-2000. The random-effects Tobit panel regression results indicate that specialization and investment would benefit new banks. In addition, rising non-performing loans and the Asian financial crisis resulted in a decrease in bank technical efficiency for both new and old banks. (2) In terms of the Malmquist productivity indexes, productivity growth of old banks is higher than that of new banks over 1992-2000. The dynamic panel data regression results indicate that productivity growth exhibits significant negative persistence for both new and old banks. Banking competition and investment benefit new banks. Moreover, non-performing loans and the Asian financial crisis can explain why new banks have a lower productivity growth rate than that of old banks over 1992-2000.