Non-contour efficient fronts for identifying most preferred portfolios in sustainability investing

Keywords Multiple objective programming; ESG in investing; Efficient frontiers; ESG integration; Tri-criterion portfolio selection Highlights * The paper provides a framework for solving the three-objective Markowitz problem where sustainability in the form of ESG is the third objective. * With no clear definition of ESG integration in the literature, we provide a crisp definition of what it is and is not. * A new technique called non-contour efficient fronts is presented for searching tri-criterion efficient surfaces for most preferred portfolios. * While current methods leave ESG on the table, the non-contour methodology leads to portfolios with significantly greater ESG content. * We develop an ESG integration index that is able to assess the degree to which ESG has been integrated into a portfolio. Abstract The paper focuses on investors whose strength of interest in sustainability issues (such as environmental, social, and governance) causes ESG to become a third criterion alongside risk and return in portfolio selection. This causes the efficient frontier to become an efficient surface. This means that an investor's optimal portfolio is no longer the point of most preferred risk/return tradeoff on the mean-variance (M-V) efficient frontier, but is the point of most preferred risk/return/ESG tradeoff on the investor's M-V-ESG efficient surface. However, to find such a point requires non-trivial ESG integration which is the name given to the process of integrating ESG into the portfolio construction process after screening. With the third objective transporting the problem into 3D-space, it is difficult to search the efficient surface in any kind of comprehensive fashion using M-V based or other bi-criterion techniques as this is akin to a 2-dimensional being trying to view a 3-dimensional object. To remedy the situation, the paper proposes a tri-criterion approach that computes efficient surfaces and special non-contour curves (called NC-efficient fronts in the paper) that are stretched across the efficient surface so as to dragnet it for the points of best ESG integration within it. Using the methodology and data from the S&P500, the paper conducts computational tests on problems with up to 500 securities and under different constraint conditions so as to know what to expect from the new approach over a range of situations. Author Affiliation: (a) University of Georgia, Department of Finance, Athens, Georgia 30602, USA (b) University of Augsburg, Faculty of Business Administration and Economics and Centre for Climate Resilience, Universitaetsstrasse 12, 86159 Augsburg, Germany * Corresponding author. Article History: Received 30 November 2021; Accepted 5 August 2022 Byline: Ralph E. Steuer [rsteuer@uga.edu] (*,a), Sebastian Utz [sebastian.utz@uni-a.de] (b)