A critique of credit risk models with evidence from mid-cap firms

The significant problems experienced by banks during the global financial crisis (GFC) have highlighted the critical importance of measuring and providing for credit risk. This chapter will examine three popular categories of credit risk measurement models and provide an analysis of the relative shortcomings and advantages of each method. The categories include ratings based models, financial statement analysis models, and the Merton / KMV structural model. Each model assesses different criteria, and responds differently to dynamic economic circumstances. As part of our assessment of these models, we provide a US-based empirical analysis, using a mid-cap data set that spans a 10-year time frame, and that includes the pre-GFC, GFC, and post-GFC periods. Outputs for each of the three model categories are benchmarked against actual impaired assets and defaults. We find the key advantages of the ratings and accounting based models to be the wide range of factors included in the initial analysis, but that they are very slow to respond to changing economic conditions. Structural models, on the other hand, while factoring in lesser factors into the initial analysis, are very responsive to dynamic conditions. An understanding of the merits and disadvantages of the various models can assist banks and other credit modelers in choosing between the available credit modeling techniques.