ABSTRACT Sustainable governance depends on the joint functioning of institutional quality, fiscal discipline, environmental performance, and socioeconomic inclusion. However, many composite indicators rely on subjective weighting schemes and leave the structural role of governance underspecified. This study addresses these limitations by developing a theory‐guided and validation‐aware Multicriteria Decision‐Making (MCDM) framework to evaluate sustainable governance within the 30‐country Sustainable Governance Indicators (SGI) 2024 sample. The empirical design uses 28 SGI indicators organized into four balanced thematic blocks. The primary specification combines Entropy, CRITIC, and MEREC average weights with the Combined Compromise Solution (CoCoSO) ranking method, while Fuzzy Decision by Opinion Score (FDOS), Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (Fuzzy TOPSIS), and Borda–Copeland aggregation are retained as robustness layers. Theme‐level sensitivity results show that weighting schemes are not identical, but country rankings remain strongly concordant across alternative ranking procedures. Benchmark comparisons with the corresponding SGI top‐level dimensions further indicate that the retained primary model is substantively credible. The findings support the interpretation of governance as a coordinating institutional architecture linking fiscal, environmental, and socioeconomic performance. Critically, the study positions governance not merely as a policy output but as the institutional scaffolding within which firms pursue eco‐innovation, plan environmental investments, and navigate sustainability transitions. Countries exhibiting strong governance–fiscal–environmental alignment offer more predictable regulatory contexts and lower transition friction for corporate environmental strategy. Rather than claiming algorithmic novelty, the study contributes a transparent and reproducible composite‐design strategy that combines balanced indicator architecture, explicit robustness reporting, and benchmark‐based validation.
Rençber et al. (Fri,) studied this question.