Loan Portfolio Loss Models With More Flexible Asymmetry and Tails for Korean Banks and a Comparison of Their Regional Concentrations

This article extends the Vasicek model for the Gaussian single-factor portfolio loss distribution to a skew-elliptical multifactor model and proposes loss models with more flexible asymmetry and fat tails for credit portfolios. By comparing the sensitivities of portfolio loss value-at-risk with respect to the model parameters, we show that the portfolio loss distributions depend not only on the model parameters but also on the multivariate nature of the factor model. With the empirical tests based on the Korean banks' nonperforming loan rates, our proposed models outperform the Vasicek model and provide more appropriate asset correlation and a systematic risk measure reflecting the characteristics of bank-level nonperforming loan rates. In addition, we measure the regional concentration of all the Korean banks using the Herfindahl-Hirschman index. We find that the regional concentration of a bank is strongly related to bank-level systematic risk, and the multivariate nature of asset return has a huge effect on bank-level systematic risk for the given regional concentration of a bank.