Dynamics of systemic risk in European gas and oil markets under the Russia–Ukraine conflict: A quantile regression neural network approach

Abstract

The Russia–Ukraine conflict (RUC) has triggered a serious natural gas (hereafter referred to as gas) and oil supply crisis in Europe. We propose a quantile regression neural network model to capture the non-linear evolution of systemic risk in the European gas and oil markets. We find that the RUC significantly increased systemic risk in both the European gas and the oil markets. The systemic risk is higher, and rises much more quickly and falls much more slowly in the gas market than in the oil market. The dynamics of systemic risk are closely linked to major events during the RUC. The US-dollar-to-ruble exchange rate contributes most to this systemic risk, followed by Europe’s gas stocks and gas imports from Russia. In terms of risk exposure, the gas market is more vulnerable than the oil market. We propose an elasticity coefficient of systemic risk to evaluate its sensitivity to stress scenarios. Our study provides important insights into managing the systemic risk in European gas and oil markets after the RUC.

Keywords

• Russia–Ukraine conflict
• European energy markets
• Quantile regression neural network
• Systemic risk
• Sensitivity analysis