A0212
Title: ESG compliant optimal portfolios: optimizing after screening or screening while optimizing
Authors: Costanza Torricelli - University of Modena and Reggio Emilia (Italy) [presenting]
Beatrice Bertelli - University of Modena and Reggio Emilia (Italy)
Abstract: Accounting for environmental, social, and governance (ESG) dimensions in optimal portfolios is of uttermost importance for the financial industry. Given the limited literature and the lack of consensus on the best-performing ESG strategies, the aim is to assess the impact of ESG on optimal portfolios according to approaches that differ in terms of ESG compliance level. The risk-adjusted performance of three main approaches is compared: the first consists of optimizing on an ESG-screened sample, the second optimizes on an unscreened sample but adds to the optimization problem a portfolio ESG-score constraint, the third combines features of both by optimizing with an ESG constraint over a (slightly) screened sample. The optimization follows a recent study in minimizing portfolio residual risk with a desired level of portfolio average systemic risk. The sample based is on the 586 stocks of the EURO STOXX Index (2007 - 2022) and Bloomberg ESG scores. Optimization after screening (1st approach) implies a risk-adjusted performance superior only with heavy screening (>50\%). Accounting for ESG while optimizing (2nd approach), returns portfolios with a performance that, for each level of systemic risk, worsens as the target ESG level increases, whereby portfolios with the highest ESG scores perform better during bullish periods. Optimizing with an ESG constraint after screening (3rd approach) combines the advantages of both when the screening threshold is low (20\%).
