Algorithmic Decomposition of ESG Scores: A Sustainability Governance Analysis of Indian Listed Companies

Environmental, Social, and Governance (ESG) scores have become widely used indicators of corporate sustainability performance. However, the internal structure and informational content of these composite scores remain insufficiently examined, particularly in emerging markets. This study investigates the construction and explanatory structure of ESG scores for Indian listed companies using firm-level ESG data obtained from CRISIL for the year 2024. A diagnostic analytical framework is applied to decompose the composite ESG score into its underlying Environmental (E), Social (S), and Governance (G) components. Using descriptive statistics, correlation analysis, ordinary least squares (OLS) regression, residual diagnostics, and clustering techniques, the study evaluates the extent to which the aggregate ESG score is explained by its components and identifies structural patterns in firm-level ESG performance. Results show that the composite ESG score is largely an aggregation of its underlying components, with Governance exhibiting the highest marginal contribution to the final ESG score. Residual analysis reveals minor deviations between predicted and observed ESG values, suggesting limited but observable rating inefficiencies. Cluster analysis further identifies distinct sustainability profiles among firms rather than a uniform ESG progression. To enhance methodological rigor, robustness checks using Principal Component Analysis (PCA), sector-interaction regression models, and bootstrap validation are proposed to examine stability of coefficients and sectoral heterogeneity. The findings indicate that while ESG scores provide a convenient summary indicator, they primarily represent structured combinations of underlying sustainability indicators rather than independent measures of sustainability performance. The study contributes to ESG analytics by providing a structural decomposition framework that improves interpretability and transparency in sustainability ratings.