國有銀行、政治干預與銀行民營化

Government Banks, Political Interference and Bank Privatization

林智勇(Chih-Yung Lin)

指導教授 ： 沈中華

國立臺灣大學/管理學院/財務金融學研究所/博士(2010年)

https://doi.org/10.6342/NTU.2010.02049

摘要

本論文包含三篇關於國有銀行、政治干預與銀行民營化的文章。第一篇文章為“反民營化:國有銀行是否有較好的績效？從政治觀點的角度分析之”。 第二篇文章為“銀行民營化、國家治理、投資人保護與銀行管制”。 第三篇文章為“銀行民營化是否能改進其績效表現?暫時或持久效果”。 最後，我們在最後一節說明論文的結論。

一、反民營化:國有銀行是否有較好的績效？從政治觀點的角度分析之
過去有許多的研究都指出國有銀行的績效表現比民營銀行來的差。但是卻沒有實證研究去研究此現象存在的原因為何?本研究企圖利用三十七個國家從1993到2007的資料去補足此文獻上的缺口。為了簡潔的理由，我們將國有銀行比較差的表現定義為“GOB 效果”。兩個配對方法 (Nearest-Neighbor Matching 和 Mahalanobis Metric Matching) 被用來選取該國特徵變數相似的配對樣本。我們定義政治干預的過程為「當國有銀行被要求去併購倒閉銀行的過程」。接者我們發現國有銀行在併購一家倒閉銀行之後，其績效有明顯的下降。此證據顯示政治干預的過程可能是GOB 效果存在的原因。因此我們接著去研究沒有受到政治干預的國有銀行表現，我們發現這些銀行的績效並沒有比民營銀行來的差。所以我們認為政治干預的過程會造成國有銀行的績效變差，以及傳統上我們所認為的GOB 效果是源自於政治干預影響的結果。

二、銀行民營化、國家治理、投資人保護與銀行管制
過去有許多的研究都指出國有銀行的民營化可以提升其績效表現。為了簡潔的理由，我們將銀行民營化後的績效提升定義為“民營化效果”。 但是卻沒有實證研究去研究為何民營化效果存在於某些國家但卻不存在於其他國家。本研究企圖利用43個國家從1992到2007的民營化資料去補足此文獻上的缺口。兩個配對方法 (Nearest-Neighbor Matching 和 Mahalanobis Metric Matching) 被用來選取該國特徵變數相似的配對樣本。我們提出了四個假說並發現民營化效果在國家治理差、投資人保護差以及銀行活動管制較鬆的國家會比較有效。

三、銀行民營化是否能改進其績效表現?暫時或持久效果
本研究企圖利用43個國家從1992到2007的民營化資料去研究銀行民營化是否能改進其績效表現?四個配對方法 (Nearest-Neighbor Matching、1-to-2 Nearest Neighbor Matching、Mahalanobis Metric Matching和 1-to-2 Mahalanobis Metric Matching) 被用來選取該國特徵變數相似的配對樣本。實證結果如下：首先，銀行民營化可以提升其績效表現在於股東權益報酬、淨利息收益與不良資產放款等衡量指標，但是卻無法提升銀行的資產報酬。次之，在大部分的案例當中，完全民營化比部分民營化來的有效。第三點，資產出售的民營化方式相對於公開釋股的民營化方式有著比較大的績效改進。最後，民營化的績效在長期相對於短期有著比較大的改進效果。

關鍵字

國有銀行 ； 政治干預 ； 配對理論 ； 倒閉銀行 ； 銀行民營化和國家治理

並列摘要

This dissertation consists of three essays on Government Banks, Political Interference and Bank Privatization. The first essay is “De-Privatization: Do Government Banks Perform Better? A Political Interference View.” The second essay is “Bank Privatization, Country Governance, Investor Protection and Bank Regulation.” The third essay is “When Does the Privatization of Banks Improve Performance? Transient or Persistent Effect.” Last, we present the conclusions in last section.

Essay 1:
De-Privatization: Do Government Banks Perform Better? A Political Interference View
Many studies report that government-owned banks underperform compared with private-owned banks. However, no empirical studies have examined the reasons for this evidence. This study addresses this gap using banks from 37 countries from the period 1993 to 2007. For simplicity’s sake, the underperformance of government-owned banks (GOB) is referred to as the “GOB effect.” Two matching methods, the Nearest-Neighbor Matching and Mahalanobis Metric Matching, are adopted to seek matched banks (e.g. sharing similar characteristic variables) located in the same country. We define political interference as the process by which government-owned banks are mandated to merge with a distressed bank. We will then show that government-owned banks, after merging with a distressed bank, suffer adverse performance. This evidence indicates that political interference may be the cause of underperformance. Furthermore, we observed that government-owned banks that did not experience any political interference generally perform as well as private-owned banks. Therefore, we reiterate that political interference negatively affects government banks’ performance, and the GOB effect in stereotypical image is a consequence of political influences.

Essay 2:
Bank Privatization, Country Governance, Investor Protection and Bank Regulation
Many studies report that the privatization of government-owned banks (GOBs) can enhance their performance. For simplicity’s sake, the improvement of bank privatization referred to as the “privatization effect”. However, no empirical studies have examined why privatization effect exists in some countries, but not in others. This study addresses this gap using bank privatization from 43 countries during 1992~2007. Two matching methods, the Nearest-Neighbor Matching and Mahalanobis Metric Matching, are adopted to seek matched banks (e.g. sharing similar characteristic variables) located in the same country. We propose four hypotheses and find that the privatization effect are enhancing when countries with worse country governance, weak investor protection and fewer bank activity regulation.

Essay 3:
When Does the Privatization of Banks Improve Performance? Transient or Persistent Effect
This study compares the performances of privatized banks from 43 countries during 1992-2007 by using four matching theories, Nearest-Neighbor Matching, 1-to-2 Nearest Neighbor Matching, Mahalanobis Metric Matching, and 1-to-2 Mahalanobis Metric Matching. The evidence demonstrates the following: first, the privatized banks outperform non-privatized banks in terms of return on equity (ROE), net interest margin (NIM) and non-performance loan (NPL) but are tied in terms of return on asset (ROA); second, in most cases, full privatization is more effective than partial privatization in improving bank performance ; third, the results demonstrate that privatization through asset sales yield a better performance; and finally, the results demonstrate that improved performance, measured in terms of profitability, is exhibited more in the long-term while with respect to improved performance in terms of asset quality, the improvements are more prevalent in the short-term.

並列關鍵字

Government-owned banks ； political interference ； matching methods ； distressed bank ； bank privatization ； and country governance

參考文獻

Beck, T., Cull R., Jerome A., 2005. Bank privatization and performance: Empirical evidence from Nigeria. Journal of Banking and Finance 29, 2355-2379.

Bonin, J., Hasan I., Wachtel P., 2005. Privatization matters: Bank efficiency in transition countries. Journal of Banking and Finance 29, 2155-2178.

Bos, J.W.B., Kool C.J.M., 2006.Bank efficiency: The role of bank strategy and local market conditions. Journal of Banking and Finance 30, 1953-1974.

Boubakri, N., Cosset J.C., Fischer K., Guedhami O., 2005. Privatization and bank performance in developing countries. Journal of Banking and Finance 29, 2015-2041.

Boubakri, M., J. C. Cosset, and W. Saffar, 2008, Political Connections of Newly Privatized Firms, Journal of Corporate Finance 14, 654-673.