ESG-Aware Asset Selection and Portfolio Optimization: A Comparative Study of Multi-Criteria Decision-Making Methods

Besides the main goal of investment, which is to obtain profit, investors are also paying more attention to the broader impact of their choices. This awareness is reflected in portfolio construction, where environmental, social, and governance (ESG) scores are included alongside financial considerations. In this study, we propose a two-stage portfolio construction framework that uses ESG factors while separating asset selection from portfolio optimization, thereby allowing the effects of sustainability factors to be examined at both levels. Using a universe of 418 U.S. equities with ESG scores and sector classifications, assets are screened and ranked using three multi-criteria decision-making methods (AHP, TOPSIS, and VIKOR), based on ESG score, expected return, and volatility, while also ensuring sector balance. In the second stage, the selected asset universes are optimized in a mean-variance framework, producing classical, green (high ESG), and brown (low ESG) efficient frontiers. The findings indicate that ESG constraints have only a limited effect on portfolio efficiency. Green portfolios show similar risk-adjusted performance to classical portfolios. The asset selection methods mainly influence the composition and risk of the portfolios. VIKOR-based portfolios are more stable, TOPSIS-based portfolios are more volatile, and AHP portfolios are between these two cases. Overall, the results suggest that portfolio outcomes depend on both the asset selection method and the way sustainability constraints are introduced during optimization.