Predicting the Default Risk of Firms: A Model with Safety Covenants

Abstract

有鑑於企業違約預警對於經濟體系的重要性，本文應用障礙選擇權理論，建構一個較符合實際經濟社會違約過程的違約預警模型。實證結果顯示，相較於傳統BSM（Black and Scholes, 1973與Merton, 1974）利用市場資訊所估算之結構式模型，以障礙選擇權理論建立的DOC(down-and-out call option)模型，其違約預警能力相較BSM模型而言有提升之效果。另外，透過censored Tobit迴歸模式觀察影響兩種結構式模型表現的相關因素，亦可以發現，DOC模型於建構過程中，較傳統BSM模型多考慮、企業獲利性層面因素，故更能有效地偵測企業違約之發生。因此，本文認為DOC模型亦可作為另一個判斷企業違約風險的預警工具。

This study uses barrier option theory to establish a credit risk model with greater relevance to the process of default by firms in the real world. As compared to the traditional Black-Scholes-Merton (BSM) structural model, which makes use of market information along with the results of the empirical testing of default predicting performance, we suggest that our down-and-out call option (DOC) model, established on the basis of barrier option theory, provides superior performance. When factoring in profitability, and when using the censored Tobit regression model to observe the characteristics of these two structural models, we find that the DOC model is more effective at predicting default events; we therefore conclude that the DOC model is another appropriate model for the measurement of the credit risk of firms.