Do some Value-at-Risk Models Provoke Financial Market Destabilization? - An Agent-Based Financial Market Perspective

The aim of this paper is to explore the impact of different Value-at-Risk (VaR) models on the stability of financial markets. Based on a numerical analysis, we test how simple and more sophisticated Value-at-Risk models affect financial market stability. This is important to implement effective regulations to prevent financial market instability in advance. The Basel Committee of Banking and Supervision (BCBS) does not stipulate that banks use a special type of VaR model. Therefore, in practice, many banks use methods with quite simple assumptions. Testing the efficiency and reliability of different VaR models with different underlying assumptions was and still is an important strand of research. Here, mostly the accuracy of estimating a special quantile is exploited. In this study, we explore the adequacy of VaR models from another perspective. We adjust a heterogeneous agent model by integrating regulations of Basel II concerning market risk. For this purpose, we use the financial market model by Lux/Marchesi (1999/2000) that can replicate stylized facts of financial markets quite well. First results indicate that the formula for calculating the level of regulatory capital for market risk prescribed by the BCBS prohibits more market stability by the use of more sophisticated models.