Vojtěch Siuda
This paper provides a framework for conducting simulations and stress testing in the non-financial corporations sector. It relies on national accounting and uses a set of input-output tables to track the propagation of shocks between parts of the sector while staying fully consistent with the big picture framed by the core forecasting model and the underlying scenario. The simulation framework allows standard macroeconomic developments to be captured, but one-off measures such as government wage and salary compensation and loan moratoria can also be easily implemented. The main output of the simulation is a set of industry-level performance and profitability variables. These variables can be used for various types of analysis, such as credit risk modelling and profitability and liquidity analysis. Some of them – such as the forecasting of portfolio default rates via learning process – are shown in the paper. The historical default rate estimates obtained are accurate and economically sensible for the majority of industries and exhibit a high degree of reliability even under very severe economic conditions. Given its national accounting framework and its level of detail, the model can be used to support decision-making processes and to evaluate the effects of existing or planned economic policies. Two different scenarios are considered to demonstrate the benefits of the proposed approach.
JEL codes: G01, G32, H63
Keywords: Credit default, default rate forecast, economic shock propagation, input-output tables
Issued: December 2020
Download: CNB WP No. 12/2020 (pdf, 1.5 MB)