東亞銀行危機預警模型研究

Prediction of Bank Failure in East Asia： A Survival Analysis Approach

彭思蓉(Szu-Jung Peng)

指導教授 ： 許振明

國立臺灣大學/社會科學院/經濟學研究所/碩士(2008年)

https://doi.org/10.6342/NTU.2008.01574

摘要

過去數十年，銀行危機的發生越來越頻繁。而在全球化的趨勢之下，擁有類似金融背景並彼此經濟相依的東亞各國更應共同維持區域內銀行體系的穩定，建立風險預警機制，以避免銀行危機擴散，使得東亞金融危機再次發生。
本研究便以東亞十國—中國大陸、香港、印尼、日本、韓國、馬來西亞、菲律賓、新加坡、台灣與泰國共計253家當地商業銀行(其中62家為危機銀行，191家為正常銀行)為研究對象，應用參數存活模型，分析2001年至2006年影響東亞各國銀行存活率與危機率的關鍵因素。
從本研究的實證結果可知，銀行的存活期間以Log-logistic模型配適度最佳，即銀行的危機率從0開始隨著時間增加而遞增，達到極大值後遞減。此外，亦可發現營運因素、資產品質因素、廣義的貨幣供給(M2)佔外匯準備比率、貸款利率對存款利率比與民間放款佔GDP比率對於東亞各國銀行的存活期間有顯著的負向影響。

關鍵字

東亞 ； 銀行失敗 ； 銀行危機 ； 因素分析 ； 存活分析

並列摘要

During the last decade, episodes of banking crises have become more frequent. In these countries suffered from banking crises, not only hurt financial sector and economic development, but also result in inestimable cost to society. Thus, establishing an early warning system to prevent bank failure and stabilize regional banking system is very important.
This paper will analyze the determinants that may result to the bank failure between 2001 and 2006, using a dataset of 253commercial banks operating in China, Hong Kong, Indonesia, Japan, Korea, Malaysia, Philippines, Singapore, Taiwan and Thailand (of a sample of 253 banks,62experience distress). The analysis will extract the common factors by means of principle component analysis out of factor analysis, and further apply the parametric survival model.
According to the empirical findings, we find that the survival time best follow the Log-logistic distribution. Based on the characteristics of hazard function under Log-logistic distribution, it can be inferred that the hazard rate of banks shows a pattern of increase followed by decrease as the survival time increases. Furthermore, it can be found out that operative factor, asset quality factor, M2 to foreign exchange ratio, lending rate to deposit rate ratio and credit claim to private sector to GDP ratio is significantly affect the duration and hazard rate of banks.

並列關鍵字

East Asia ； bank failure ； banking crises ； factor analysis ； survival analysis

參考文獻

余士迪、張呈徽和紀志毅 (2007)，「利用混合機率分配探討農會信用部發生擠兌的危險率」，2007年台灣經濟計量學術研討會，台北:中央研究院經濟研究所。

王瑜琳與洪嘉聲 (2004)，「農業信用部擠兌與經營狀態之探討」，《農業經濟叢刊》，10，77-100。

Altman, E. I. (1968), “Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy ,” The Journal of Finance, 23, 589-609.

Barros, F. and J. Williams (2007), “Analysing the Determinants of Performance of Best and Worst European Banks: A Mixed Logit Approach,” Journal of Banking and Finance, 31, 2189-2203.

Braga, M. J., V. G. F. Bressan, E. A. Colosimo and A. A. Bressan (2006), “Investigating the Solvency of Brazilian Credit Unions Using a Proportional Hazard Model,” Annals of Public and Cooperative Economics, 77, 83-106.