留學生作業(yè)代寫：銀行倒閉與危機
The Bank Failures and Crisis

在文獻中提出的大量的研究中，關(guān)于銀行倒閉的預測進行了研究。這是特別重要的，，因為銀行倒閉是對經(jīng)濟會產(chǎn)生直接威脅。然而，銀行破產(chǎn)并不是一個新話題；他們自上世紀70年代末以來變得越來越普遍，各種方法已被用于構(gòu)建預測模型，為建立未來銀行倒閉的可能性。研究人員開發(fā)了適當?shù)墓ぞ�，這將使他們能夠從過去的經(jīng)驗檢測到銀行的問題。本文提供了一個審查，這樣的方法來找到一個銀行危機的指標，并評估他們的表現(xiàn)。

The prediction of bank failures has been researched in a large number of studies presented in literature. It is especially crucial because bank failures are a direct threat to the economy. However, bank insolvencies are not a new topic; they have become increasingly common since the late 1970s. Various methods have been used to construct prediction models in order to establish the probability of future bank failures. Researchers developed appropriate tools which would have enabled them to detect bank problems from past experience. This paper provides a review of such methodologies employed to find the indicators of a banking crisis and assesses their performance.

Firstly, the relatively more popular approach is to use logit or probit models. A representative study that uses the logit framework is by Logan (2001). This study investigates the balance sheet characteristics of the small and medium-sized UK banks in two time periods, specifically in the quarter prior to the announcement of the closure of Bank of Credit and Commerce International (BCCI) and before the recession of the early 1990s. It deals with finding the short-term predictors and the longer-term leading indicators of bank failures. The dependent variable in the logistic regression is a discrete variable that takes the value 1 for failure and the value 0 for non failure. According to Logan (2001) the results from logit and probit estimation techniques are fairly close.

As stated in Logan (2001) analysis, explanatory variables such as rapid credit growth, high levels of provisions as a percentage of total assets and high ratio of risk-weighted assets to unweighted assets constitute measures of credit risk and would be potential leading indicators of bank failure. Furthermore, illiquidity creates additional losses and it is proxied by the following variables: the non-marketable loans as a percentage of total assets, the proportion of bank's deposits placed by other UK banks and the liquidity mismatch between short term assets and liabilities. Moreover the excessive exposure to real estate industry or the high dependence on Net Interest Income increases the bank vulnerability. Another helpful variable in predicting failure is the size of the bank with the purpose to capture diversification opportunities. Also, capital adequacy and some earning variables like income to cost ratio, profit as a percentage of total assets, provisions and profitability net of tax or profitability pre tax display the ability of the bank to overcome unexpected losses. The main measure of capital is identified by the leverage ratio which expresses the assets as a percentage of the total net capital.

Interpreting the results of this academic work, low loan growth, low profitability, low short-term assets relative to liabilities, high dependence on net interest income and low leverage are found to be significant short term predictors of banking crises. As Logan (2001:24) points out, "banks that went to fail were already showing signs of fragility." In the short-term analysis, it is lower rather than higher credit growth which is associated with failure because vulnerable banks need to write off bad loans. Additionally, the more short-term liabilities exceed short-term assets, the greater the possibility of bankruptcy because the cost of serving the liabilities is bigger than the revenue from the assets. Furthermore, the earnings from traditional lending are more volatile. Consequently, when the dependence on net interest rate is high, the likelihood of failure will be increased. As expected, lower probability is also presented to be positively related to failure. Moreover, leverage ratio has a negative effect on failure, indicating that the probability of failure will be increased if the leverage ratio is decreased. It could be explained by the fact that the weakened banks are forced to hold high capital in relation to assets.

In contrast, the longer-term leading indicators of future failure would be more useful as they can be used by regulators or central bank in order to develop action for prevention policy. For this reason, data have been examined from the pre-recession period, before banks actually weakened. The most statistically significant variable from the earlier period regression is loan growth. In this estimation period, high credit growth is positively correlated with failure and it is proved to be a significant precursor of the banking crisis. The rapid expansion of bank loans, at the peak of the previous boom, reflects loan quality problems in the next recession period because of a poor selection of creditworthy customers.

The predictive performance of indicators based on the regression with short time horizon has been encouragingly accurate in identifying subsequent failures. Appropriate variables have been selected in order to minimize the likelihood of multicollinearity and therefore, more accurate results have been exacted. However, the predictive ability of the model deteriorates when a longer-term period is explored.

The estimation of a multivariate logit econometric model, based on a large sample of developed and developing countries for 1980-94, indicates that banking crises tend to occur when the growth rate of GDP is low and inflation is high. Also, high interest rates and balance of payments deficits create deterioration in the bank condition. High exposure to foreign exchange risk tends to increase the likelihood of a banking crisis. Similarly, the bank profitability is negatively affected by an unexpected depreciation of the domestic currency, particularly when banks borrow in foreign currency and lend in domestic currency. The ratio of M2 money supply to the central bank's holdings of foreign-exchange reserves is introduced as an explanatory variable because vulnerability in the banking sector appears to be associated with sudden capital outflows. Moreover, the presence of deposit insurance scheme creates incentives for excessive risk-taking and therefore more fragile banking sector. Other factors significantly associated with increased vulnerability in the banking sector are low liquidity, a high share of credit to the private sector and past credit growth. Finally, countries with weak institutions are more likely to experience crises.

Secondly, the signaling approach is another major predictive method that has been adopted in the literature. Kaminsky (1998) and Borio and Lowe (2002) are examples of studies based on the signaling approach. Signal extraction model compares the behavior of single variables in periods of tranquility with periods of banking crisis. An indicator provides a signal of future crisis when it exceeds a predetermined threshold value which distinguishes between normal and abnormal behaviour for each variable. Nevertheless, there are two types of signalling errors, a type I error represents failure to identify a crisis and error type II represents a false prediction of failure.

The performance of signaling approach depends on the ability to accurately call crises and non-crises episodes. In common with the results from logit regression, the signalling approach reports that high real interest rates, rapid domestic credit growth, low output growth and falls in the terms of trade divided by real exchange rate are associated with banking sector problems. The most suitable approach for a global early warning model is the logit model whereas for a country specific the best early warning model is the signal extraction.

Estrella et al. (2000) conducted a study for evaluating the effectiveness of capital ratios in identifying bank failure. It has been estimated a logit model under cross-sectional data rather than panel in order to avoid time dependency. Hazard analysis is another interesting method which has been used by Estrella et al. (2000) to forecast the likelihood of failure over a longer period, for instance when the data are time series. It is concluded that the leverage ratio and the gross revenue ratio are better in predicting bank failure over a short time horizon whereas the risk weighted ratios are more effective predictor of failure over long time horizons.

An analysis of U.S bank failures was carried out by Thomson (1991) using a single-equation logit model in order to discriminate samples of failed and non-failed banks over the 1984-1989 period. Previous studies had to pool bank failures across years to obtain an adequate sample. However, in this paper each year was examined separately because of the large number of failed banks in the sample period. The author incorporated measures of economic conditions in the failure prediction equation, as well as the traditional balance sheet risk measures.

Most of the proxy variables in the model were based on the CAMELS criteria, which regulators use for rating bank institutions. Capital adequacy, Asset quality, Management quality, Earnings, Liquidity and Sensitivity to market risk constitute the measures of banking rating system. Early warning system variables which are used as proxies for the asset quality, capital adequacy and liquidity risk, are similarly indentified with the analysis of Logan (2001). In addition, loans to insiders as a percentage to the total assets and overhead expenses divided by total assets reflect the quality of board and management. The coefficients of the above variables are positive and significant which indicate that the management risk and insider abuse are positively related to failure. Furthermore, the unemployment rate, the growth in personal income, the business failure rate and a measure of economic diversification were included in the model as economic condition proxies. The majority of these variables are significantly related to bank failure as highlighted four years previously when an institution actually failed. Hence, the model could be used as an early warning model of bank failure because the out-of-sample forecasting is very accurate.

Similar empirical study was undertaken by Bongini et al. (2002) to compare accounting data, stock market prices and credit ratings as indicators of bank distress. The volatility of equity as measured by the annualized monthly standard deviations of daily stock market returns, the historical default rate associated with each credit rating, the deposit insurance premium and the ratio of the market value of equity to book value of liabilities were variables included in this analysis. The authors suggested that credit ratings have limited predictive power in forecasting bank closure. It was further found that measures of bank fragility based on stock market information are a better predictor of bank failures than measures of bank fragility based on accounting information and ratings of credit risk agencies.

The empirical research by Kolari et al (2000) investigated early warning systems to determine the risk of failure of large U.S commercial banks from the period 1989 to 1992. The sample was divided into an original sample and a holdout sample which is useful for testing the forecasting accuracy of models. It was employed a parametric logit model and a nonparametric trait recognition approach to predict failures. The advantage of the trait recognition method is that it detects valuable information about the bank risk from complex interactions variables. Therefore, it is concluded that the trait recognition model outperforms the logit model.

In conclusion, failure prediction models that have been developed in the literature could have helped the prediction of financial crisis 2007-8. It might be not possible to predict absolutely the degree of crisis and the enormous dimensions that it has acquired internationally but at least, it could have been partly foreseen. Researchers have employed a number of empirical models that evaluate the leading indicators of bank crisis and they deduced that some indicators provide substantial predictive power in forecasting bank failure.

Moreover, the numerous case studies indicate that there are similarities among all banking crises even if the experience varies quite substantially across countries and over time. In particular, asset bubble, real growth rate of equity market price indexes and large deficits of current account as a percentage of gross domestic product are similar pre-crisis macro indicators. Therefore, the financial crisis 2007-8 was a natural consequence of all the above and not a completely unexpected event. However, it is important to point out that the choice of estimation models makes a difference to predictive efficiency of indicators so it should be consider the policy maker's objectives when constructing predictive models. Finally, failure prediction models are a necessary but not efficient tool for predicting bank crisis so they should be applied together with the country-specific macroprudential surveillance.

References 參考文獻
Bell J and Pain D (2000) "Leading indicator models of banking crises - a critical review" Financial Stability Review, Bank of England, issue 9, article 3, December pp. 113-29.
Bongini P, Laeven L and Majnoni G (2002) "How good is the market at assessing bank fragility? A horse race between different indicators" Journal of Banking and Finance, vol. 26, no.5, pp. 1011-1028.
Borio C and Lowe P (2002) "Assessing the risk of banking crises" BIS Quarterly Review, December, pp 43-54.
Caprio G and Klingbiel D (1998) "Bank Insolvency: Bad luck, Bad Policy or Bad banking" World Bank web site 
Demirguc-Kunt A and Detragiache E (1998) "The Determinants of Banking Crises in Developing and Developed Countries" IMF staff papers, vol.45 No.1 pp.81-109.
Estrella A, Park S and Peristiani S (2000) "Capital ratios as predictors of Bank failure", Federal Reserve Bank of New York Economic Policy Review, vol.6, no.2, pp33-52.
Kaminsky G (1998) "Currency and Banking Crises: The Early Warnings of Distress" Board of Governors of the Federal Reserve System, International Finance Discussion Paper, No. 629.
Kolari J, Glennon D, Shin H and Caputo M (2000) "Predicting Large U.S Commercial bank failures" Working Paper 2000-1, Economic and Policy Analysis, January.
Logan A (2001) "The UK's small banks crisis of the early 1990s: what were leading indicators of failure? Bank of England Working paper, No 139.
Thomson J (1991), "Predicting Bank Failures in the 1980s," Federal Reserve Bank of Cleveland Economic Review 27, pp. 9-20.