| Résumé | System-wide financial stress testing has gained prominence in the academic and policy literature as a crucial framework for assessing the resilience of the financial system. It involves evaluating the capability of the financial system to absorb shocks while maintaining its critical functions. Given the increasingly intricate and complex financial system in Luxembourg, it is important to step up the oversight of the financial sector in Luxembourg from a system-wide perspective. Such a system-wide perspective would also be appropriate for other complex financial systems abroad.
This paper develops a structural framework for system-wide financial stress testing with multiple interacting contagion and amplification effects acting through a dual channel of liquidity and solvency risk. The framework allows us to identify vulnerabilities arising from banks and investment funds in Luxembourg. It follows the system-wide stress testing framework of the ECB by incorporating the flow-performance relationship and the liquidity management with both leverage and cash targets for investment funds. By capturing the market risk, credit risk and liquidity risk simultaneously, the framework highlights liquidity and solvency interactions, allowing for dynamic balance sheet adjustments, an advanced integration of regulatory constraints and endogenous market price formation.
This study considers almost all banks (excluding branches) and three main types of investment funds, i.e., Equity Funds, Bond Funds, and Mixed Funds in Luxembourg. The exogenous shocks stem from hypothetical adverse scenarios that are similar to the Great Financial Crisis, the sovereign debt crisis, and the recent COVID-19 pandemic. Several stylized facts are documented for banks and the three types of investment funds during 2020-2023. First, the simulated shocks have significant first-round and higher-order effects on investment funds, in particular on Equity Funds. Moreover, Bond Funds display a stronger amplification factor than other types of investment funds. Second, the impact on Luxembourg banks is substantially muted. The overall bank capital depletion, measured by the total risk exposure amount, is low even in view of the tail risk metrics, which reflects the strong resilience of the Luxembourg banking sector as a whole. Finally, for both investment funds and banks, their vulnerabilities still reflect the procyclicality of the financial system. In view of these results, the joint modelling of banks and non-banks delivers clear benefits to the analytical capabilities of central banks and informs policymakers in developing the non-bank macroprudential toolkit of the future. |
