Institutional Quality, ESG Performance, and Aggressive Tax Planning in Developing Countries

Abstract

Aggressive corporate tax avoidance represents a significant fiscal and governance challenge in developing economies, where public revenues are critical for sustainable development and enforcement capacity is often uneven. This study examines whether environmental, social, and governance (ESG) performance constrains corporate tax avoidance and whether this relationship is conditioned by national institutional quality. Using a multi-country panel of 2464 publicly listed non-financial firms from 14 developing economies over the period 2015–2023, the analysis employs fixed-effects estimation, dynamic System GMM, and instrumental-variable (2SLS) techniques to address unobserved heterogeneity and endogeneity concerns. The results indicate that stronger ESG performance is associated with significantly lower levels of tax avoidance; however, this effect is highly contingent on institutional quality. ESG exerts a substantive disciplining role primarily in governance-strong environments characterized by effective regulation and credible enforcement. Heterogeneity analyses further reveal that the ESG–tax avoidance relationship is driven mainly by the governance and environmental pillars, is more pronounced among large firms, varies across regions, and strengthens over time as ESG frameworks mature. In contrast, the social ESG dimension and smaller firms exhibit weaker or insignificant effects, consistent with symbolic compliance in low-enforcement settings. By integrating stakeholder, legitimacy, agency, and institutional theories, this study advances a context-sensitive understanding of ESG effectiveness and helps reconcile mixed findings in the existing literature. The findings offer policy-relevant insights for regulators and tax authorities seeking to strengthen fiscal discipline and development financing in developing economies.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.