Volatility Modeling and Dependence Structure of ESG and Conventional Investments

Abstract

The question of whether environmental, social, and governance investments outperform or underperform other conventional financial investments has been debated in the literature. In this study, we compare the volatility of rates of return of selected ESG indices and conventional ones and investigate dependence between them. Analysis of tail dependence is important to evaluate the diversification benefits between conventional investments and ESG investments, which is necessary in constructing optimal portfolios. It allows investors to diversify the risk of the portfolio and positively impact the environment by investing in environmentally friendly companies. Examples of institutions that are paying attention to ESG issues are banks, which are increasingly including products that support sustainability goals in their offers. This analysis could be also important for policymakers. The European Banking Authority (EBA) has admitted that ESG factors can contribute to risk. Therefore, it is important to model and quantify it. The conditional volatility models from the GARCH family and tail-dependence coefficients from the copula-based approach are applied. The analysis period covered 2007 until 2019. The period of the COVID-19 pandemic has not been analyzed due to the relatively short time series regarding data requirements from models’ perspective. Results of the research confirm the higher dependence of extreme values in the crisis period (e.g., tail-dependence values in 2009–2014 range from 0.4820/0.4933 to 0.7039/0.6083, and from 0.5002/0.5369 to 0.7296/0.6623), and low dependence of extreme values in stabilization periods (e.g., tail-dependence values in 2017–2019 range from 0.1650 until 0.6283/0.4832, and from 0.1357 until 0.6586/0.5002). Diversification benefits vary in time, and there is a need to separately analyze crisis and stabilization periods.

Saudi Green Banks and Stock Return Volatility: GLE Algorithm and Neural Network Models

“…Furthermore, Górka and Kuziak (2022) explored the dependency between ESG indices and the conventional indices' volatilities using the GARCH models family and tail dependence coefficients. Moreover, Reber et al (2021) showed that the ESG disclosure reduced idiosyncratic volatility and downside the tail risk, and that higher ESG scores decreased volatility and risk.…”

Vine Copula Based Dependence Modeling in Sustainable Finance

“…The estimated parameters are given in Table D. 13. Next standardized residuals are used to form the pseudo copula data.…”

“…timber and forestry stocks. Also, G órka and Kuziak (2022) 13 included copulas to model volatilities and find diversification benefits. Others have focused on explaining how ESG scores are combined using a Random Forest algorithm.…”

“…For tree T 2 , all possible edges allowed by the proximity condition are considered. On them, we find a spanning tree maximizing the absolute Kendall's τ computed from the pseudo data as defined in (13). We continue that way until all trees, pair copula families and its parameters are selected.…”

ESG Indices Efficiency in Five MENA Countries: Application of the Hurst Exponent

“…As a result, with this emergence of socially responsible stock indices, some researchers have been focused on analyzing the performance of these indices compared to their conventional counterparts (Fowler & Hope, 2007;Jain et al, 2019;La Torre et al, 2020), and other researches analyze the volatility and risk of ESG indices (Sudha, 2015;Sabbaghi, 2022;Górka & Kuziak, 2022). Yet other studies test the relationship between redefinitions of socially responsible indices and financial performance (Capelle-Blancard & Couderc, 2009;Jain et al, 2019).…”