房屋貸款授信風險評估研究－以ｃ銀行為例

A Study on Evaluating the Credit Risk of Home Loan in the Case of C Bank

黃佐民(Tsu-Ming Huang)

指導教授 ： 李正文

中原大學/商學院/國際貿易研究所/碩士(2010年)

https://doi.org/10.6840/cycu201000251

摘要

自從次級房貸風暴在2007年年中浮現之後，2008年7月美國金融業就出現嚴重問題，除了美國眾多的銀行倒閉外，二房（房利美（Fannie Mae）和房地美（Freddie Mac））也遭受政府接管，造成了繼30年代經濟大蕭條後的金融大海嘯。金融海嘯正襲擊各國經濟，我們面臨了空前未有的危機，全球經濟大蕭條成為揮之不去的陰霾。
在現代社會中，銀行扮演金融中介的角色，其經營績效之良窳，一直是大眾所關切的課題。而授信是銀行收益之主要來源，故房屋貸款對銀行而言是非常重要且核心的業務。
行政院金融監督管理委員會和財政部都規定本國銀行將實施新巴塞爾協定，這對國內金融業者風險控管能力將造成相當衝擊。為因應新版「巴賽爾資本協定」，各金融機構應建立一套明確化、科學化且簡要化特性的房貸授信審核模式，以降低房屋授信風險。
本研究係以國內某家金融機構，2005年1月到2009年12月底貸放之房屋貸款案件中，抽取正常戶318戶、逾期戶36戶，共354戶作為樣本，利用Probit模型及Logit模型分析貸款戶授信品質之優劣，希望能找出影響房屋貸款逾期的主要因素，以建構最佳的授信評量模式。
而研究結果發現，經Probit迴歸分析後在所投入15項自變數中，得出四項與假設相符分別為：年齡、貸款金額 、寬限期、估價值。而經 Logit 迴歸分析後在所投入15項自變數中，得出五項與假設相符分別為：年齡、貸款金額、寬限期、估價值、職位。
研究發現，本研究與以往研究列為顯著之變數有不顯著的情形，是因為受到總體經濟的影響。故 本研究建議金融機構授信人員必須透過即時的統計資訊，適時修正審查的標準。

關鍵字

不良放款 ； 信用風險 ； 房屋貸款

並列摘要

Since the Subprime Mortgage storm in 2007 year, there had been serious problems in US financial industry in July, 2008. Numerous banks were closed down; meanwhile, Fannie Mae and Freddie Mac were also taken over by the US government. It had resulted to a financial tsunami, which had never happened after the Great Depression. The financial tsunami had been diminishing the economical growth in many countries; which now became a hovering haze worldwide. We had an unprecedented crisis for the global economical depression.

In the modern society, the bank is an important financial agency. The effectiveness is always a concern for the public. The credit is a main resource of bank’s income. Obviously, the house mortgage is a crucial business to the bank.

The Financial Supervisory Commission under the Executive Yuan and Ministry of Finance both stipulate that the banks in our country will implement the new Basel II. This will create a certain impact to the risk control ability of the domestic financial industry. In response of the Basel II, every financial institute should construct a set of clear, scientific, and simple pattern to the house mortgage credit verification so that it can lower down the risk of house credit.

This research is based on the mortgage loan cases of a financial organization from January in 2005 to the end of December in 2009. From those cases, 354 cases were drawn as samples, in which there were 318 normal cases and 36 overdue cases. Giving the strength of Logit mode and Probit mode on the analysis of mortgage credit quality, the main influence of mortgage overdue is expected to be found in order to structure the best credit evaluation pattern.

The research has discovered that among the 15 variables, there are 4 variables consistent to the hypothesis analyzed by the Probit regression mode—age, mortgage amount, grace period, appraisal value. On the other hand, among the 15 variables, there are 5 variables consistent to the hypothesis analyzed by the Logit regression mode—age, mortgage amount, grace period, appraisal value, and position.

The research has found that those significant variables in other researches become insignificant. It is due to the effect of the macroeconomics. Therefore, this research suggests financial credit person should adjust the audit standards in good time by means of instant statistic data.

並列關鍵字

Bad Loans ； Credit Risk ； Mortgage

參考文獻

呂美慧（2000），金融機構房貸客戶授信評量模式分析-Logistic 迴歸之應用，政治大學金融研究所碩士論文。

臺灣金融研訓院編輯委員會（2002），銀行授信實務。

林勉今（2003），消費性貸款授信風險評估之研究-以 X 銀行為例，大同大學事業經營研究所碩士論文。

李彥仕 (2006) ，影響現金卡違約之因素分析，國立中央大學資訊管理研究所碩士論文。

Edward, C. L.、Smith, L.D. (1992),“An Analysis of Default Risk in Mobile Home Credit,” Journal of Banking and Finance,16,299~312。