1 - Anderson Rocha de Jesus Fernandes CENTRO FEDERAL DE EDUCAÇÃO TECNOLÓGICA DE MINAS GERAIS (CEFET/MG) - Belo Horizonte
2 - Sabrina Espinele da Silva UNIVERSIDADE FEDERAL DE MINAS GERAIS (UFMG) - CEPEAD
3 - Simone Evangelista Fonseca UNIVERSIDADE FEDERAL DE OURO PRETO (UFOP) - DECAD/ICSA
Reumo
Socially Responsible Investments can be defined as an integrated organizational strategy that combine economic and financial interests with social and environmental issues. It is constituted by a bunch of companies that along with their normal activities (production of goods and services), also act on improving the community where they are located. In a broader perspective, one argues that the SRIs are important to a long-lasting economic and social wellbeing, what is discussed in the so-called Sustainable Development Goals, created by the United Nations.
We try to understand how Brazilian ESG/SRI indices (ISE and IGCT) differ in terms of risk and return from the broad indices (Ibovespa and IBrX), considering the mean-variance domain, which is the base for the portfolio theory. This paper aims to examine, using unadjusted and risk-adjusted methods, the homogeneity in terms of mean (return) and variance (risk) of the ESG/SRI indices comparing it with broad indices under the following hypotheses: i) ESG/SRI indices have higher returns than the market in general; and ii) ESG/SRI indices have lower risk than the market in general.
There is an extensive literature on social responsibility. In what regard corporate finance, the main approach is related to the companies’ ability to become more valuable while taking ESG/SRI practices such as a higher level of disclosure and acting in the development of local communities, schools, places, forests, etc. (Liang & Renneboog, 2020). Another possible perspective regards the way markets price ESG activities, a phenomenon that differs by country, production sector, ownership characteristics, and other factors (Gillan, Koch & Starks, 2021; Meng-Tao et al., 2023).
Our data consists of indices’ daily prices collected from B3. We selected two ESG/SRI indices, ISE and IGCT, and two broad indices, Ibovespa and IBrX. First, we did an unadjusted analysis using a parametric t-Student test and a non-parametric Wilcoxon test for equal means and performed a parametric F test and a non-parametric homogeneity Levene test for the variance of the returns. We also performed a risk-adjusted analysis represented by a trend regression model and a regression based on the Capital Asset Pricing Model (CAPM).
There is no evidence that the mean returns of ESG/SRI indices are statistically different from the returns of broad indices. Returns of the broad indices have higher variance. Even though all indices exhibit an increasing behavior, the ESG/SRI indices do not follow the same rhythm of Ibovespa and IBrX. ISE and IGCT, do not show a significant Jensen’s alpha, which implies that they do not show persistent abnormal returns with Ibovespa or IBrX as the market portfolio. There is evidence that the systematic risks are statistically different from 1, i.e., the ESG/SRI indices are less risky.
ESG/SRI indices do not have statistically different returns from the broad indices, which could imply that those good practices do not produce higher values for firms or portfolios that is composed by such firms. However, ESG/SRI returns have lower variance and lower systematic risks that is, they have low levels of risk, perhaps due to the higher degree of transparency. There is no evidence of abnormal returns.
Gillan, S.L, Koch, A., & Starks, L.T. (2021). Firms and social responsibility: a review of ESG and CSR research in corporate finance. Journal of Corporate Finance, 66, 101889.
Liang, H., & Renneboog, L. (2020). Corporate Social Responsibility and Sustainable Finance: a review of the literature, ECGI Working Paper Series in Finance. Working Paper nº 701.
Meng-Tao, C., Da-Peng, Y., Wei-Qi, Z., & Qi-Jun, W. (2023). How does ESG disclosure improve stock liquidity for enterprises – empirical evidence from China. Environmental Impact Assessment Review, 98, 106926.