Macro stress test for credit risk

Purpose – The purpose of this study is to analyze the benchmark model and offer a practical implementation of the macro stress test. The emergence of complicated instruments such as securitized products has rendered the risk management methodologies used in non-crisis periods insufficient. The macro stress test has become prominent as both an internal risk management tool for financial institutions and a way for supervisory authorities to maintain financial stability. However, no practical model is available for transforming macro stress scenarios into the risk parameters of an institution’s internal model. Design/methodology/approach – This study presents a model for assessing a company’s default risk through a multi-level regression based on simultaneous estimates of the impacts of company-specific, macroeconomic and sector-specific risk factors using panel and time series data. Findings – Equity capital, EBITDA, the current ratio and the fixed assets to fixed liability ratio are selected as the company-specific factors, while the CPI core rate, overall unemployment rate, overnight call rate and JGB yield to subscribers are selected as the macroeconomic factors. The correlation coefficients among the latent sector factors are significant at 5 per cent. In addition, the accuracy ratio values prove that the presented model has more default prediction power than do models without them. Originality/value – This study is the first to provide a benchmark model for incorporating macroeconomic variables into a credit risk model for use in a bottom-up macro stress test.