外部監控、代理問題與信用風險模型之實證研究

本文利用Logistic迴歸建構非集團業信用風險模型(Credit Risk Models)，考慮代理問題，找出影響公司績效之顯著變數，期能在公司未發生財務風險(Financial Risk)前，預測出可能發生之機率，達事前預警與風險管理(Risk Management)之效。實證分析上，採用K-S檢定、M-U檢定與Logistic迴歸分析。實證結果發現，當董事長兼任總經理，降低代理問題提高預測準確率，可以降低發生財務風險的機率，值得提供投資人決策參考；不論董事長是否兼任總經理，非集團企業宜降低每股現金流量、重視會計師保留意見、避免會計師(CPAs)更換，外部人會計師之監控能力增強，可提高公司治理能力。因此本文之實證價值與管理意涵為非集團業重視公司治理，可降低信用風險發生的可能性。

關鍵字

外部監控；公司治理；信用風險模型；財務風險管理

並列摘要

A primary objective of this research is to use statistical methods to construct a logistic credit risk model on non-group companies through discussions of agency problem that is associated with issues surrounding the separation of management and ownership. The credit risk model would make it possible to predict in advance the probability of a financial risk attacking non-group companies. To construct such a credit risk model,this study employs statistical procedures such as K-S test,M-U test and logistic regression model.Empirical findings are: in the case where ownership is combined with management, no agency problems were present, and the degree of predictive accuracy was increased.However, non-group companies with combined ownership and management should reduce the cash floe per share,heed auditing advices accounting opinions from a CPA,refrain from replacing CPAs,and institute an outsider monitoring and corporate governance mechanism. In short,non-groups companies that value corporate governance would reduce credit risk.