Risk Measurement and Actuarial Logic Analysis of ESG Investment Portfolios under the Exponential Model: Evidence from Green Industries

Environmental, Social and Governance( ESG) investment has attracted more and more attention in the field of green finance. However, The nonlinear and heavy tail nature of the risk associated with ESG assets is not well represented by traditional asset risk measures such as mean–variance model. The paper fills an important void in current research regarding the reason for underestimating the risk of ESG assets using standard models. A new composite model based on exponential finance theory and actuarial analysis is developed here.We achieve this by combining the methods of dynamic sensitivity analysis and probabilistic, time-consistent valuation. Methodologically, we extend a single-index model with an explicit ESG risk factor, and we construct an ESG Adjusted Conditional Value at Risk (ESG-CVaR) metric. Using historical data of constituents of the S&P100 and green industry classifications, our formulation is empirically validated using a portfolio optimization with realistic constraints.The limits are the ESG score and leverage ratio. Empirical results reveal that the ESG risk factor is significant on some levels. The green stocks have negative skewness and fat tail, which can not be captured in the classic model. The CVaR with ESG always has larger value of risks than the classic VaR.stressing on the materiality of extreme ESG events.The optimized portfolio built with respect to the proposed model is more resilient, having a smaller maximum draw down in case of market stress related to the sustainability issue. To conclude, Applying actuarial methods to financial risk modelling provides us with an intuitive, powerful way of addressing the issue of ESG risk.