The validation of different systemic risk measurement models

Hu Wang and Shuyang Jiang

Abstract

This paper improves on the DebtRank model by incorporating a capital buffer. We use data from China’s banking industry to compare the systemic risk measured by the original, improved and differential DebtRank models. Our results show that the relationship between the systemic risk and initial shock varies over the three models. When the initial shock is small, the systemic risk measured by the improved Debt- Rank is the smallest, followed by the original DebtRank, while the systemic risk measured by the differential DebtRank is the largest. When the initial shock exceeds a certain threshold, the systemic risk measured by the original DebtRank is the smallest, followed by the improved DebtRank, while the systemic risk measured by the differential DebtRank is the largest. This shows that, when the risk shock is small, the existence of a capital buffer inhibits risk contagion and thus reduces systemic risk. However, as the risk shock increases, the role of the capital buffer in reducing risk contagion reduces. The improved model measures systemic risk more accurately than both the original and differential DebtRank models.