Managing ESG Ratings Disagreement in Sustainable Portfolio Selection

Sustainable Investing identies the approach of an investor whose aim is twofold: on the one hand, she wants to achieve the best compromise be- tween portfolio risk and return, but she also wants to take into account the sustainability of her investment, assessed through some Environmental, So- cial and Governance (ESG) criteria. According to the relevant literature on this topic, the inclusion of sustainable goals in the portfolio selection process has an actual impact on nancial performances (see, e.g., [3]). ESG indices provided by the rating agencies are generally considered as good proxies for the sustainability performance of an investment, as well as, appropriate mea- sures of the Socially Responsible Investments (SRI) in the market. Despite this, the lack of alignment among ratings provided by dierent agencies is a crucial issue that inevitably undermines the robustness and reliability of these measures (see [1, 2]). Indeed, the ESG rating disagreement could result in conicting information, implying a diculty for the investor in the ESG evaluation of her portfolio. In fact, this may cause some underestimation or overestimation of the market opportunities for a sustainable investment. In this paper, we deal with a multi-criteria portfolio selection problem taking into account risk, return, and ESG criteria as objectives. For the securities in the market we consider more than one agency and propose a new approach to overcome the problem of managing the disagreement of the dierent ESG ratings. We exploit the k-sum optimization strategy provided in [5, 4] by which we are able to re-formulate the original three-objective nonlinear op- timization model as a convex quadratic programming model. An extensive empirical analysis of this model is performed on real-world data sets taken from main stock markets.