本研究以國內某大商業銀行為研究對象,以91~96年最近五年中之消費性擔保貸款客戶為研究樣本,萃取該銀行北部八家分行,正常消費者擔保貸款客戶150件,及北縣市債權(逾放)區域中心,抽取逾期戶156件,共306件樣本,從借款戶所提供個人資料表及該行徵信人員向票據交換所查詢,最近一年內是否有退票紀錄?並查詢聨合徵信中心,了解借款戶貸款額度,繳息情形、現金卡、金融卡額度使用情形,共選出本研究16個風險變數,研究方法則採用統計學之敘述統計、卡方分析、羅吉斯迴歸分析(Logit),進行實證分析,研究發現16個風險變數中,影響借款戶違約之顯著變數有:職業、職稱、教育程度、婚姻、留通訊資料、服務現職、服務年資、信用卡及現金卡有效張數、年收入、放款成數等9項因素。 本研究結果期望能提供商業銀行授信人員、債權中心人員及銀行同業,日後設定處理不良債權管理機制的參考,使不良資產管理,得以在短期時間內,提早發現,並做最有效的處理,達到銀行增加利潤盈餘之目標,確保銀行經營管理競爭優勢。
This research will target at 150 normal collateralized consumer loans from 8 braches of a famous commercial bank located in the Northern District and another 156 overdue collateralized loans listed in the Taipei City and Taipei County Overdue Debt Management Centers. A five year time range from 2002-2007 of these total samples of 306 cases will be studied which content includes personal data sheet filled by individuals and records shown in the Bill Exchange Commission such as rejected checks records. In addition, 16 risk variables such as: United Credit Bureau of consumers’ loan amount, interest payment schedules, cash cards, financial cards, amount and usage conditions are also investigated. The descriptive statistical, Chi-squared test, Logistic regression, and differentiation analysis are applied as the research methods. The empirical results and contributions are as follows: Significant variables affecting branch of contracts:Occupation, job title, education background, marital status, contact information, years of current job experience, current effective numbers of credit cards and debit cards, annual salary, ratios of loans in total of 9 items . The purpose of this paper is to provide significant references to bank credit release personnel, debt centers and banking personnel while facing bad debt problems resolutions. Early discoveries of bad assets in a shorter period of time can help more efficient resolution conducts and the goal of increasing bank profit and assurance of bank competitive advantages can be achieved.