Gender-diverse boards curb corporate tax avoidance, and firms that deploy AI see a larger governance effect; AI capability amplifies the tax‑compliance role of diverse boards in developing‑country firms.

Main Finding

Board gender diversity (BGD) is associated with higher effective tax rates (i.e., less corporate tax avoidance) in firms from developing economies, and this governance effect is significantly strengthened in firms with greater AI capability (using a lagged AI measure). Results are robust to multiple alternative measures and econometric corrections for selection and endogeneity.

Key Points

• Sample: 1,586 non-financial firms from developing economies, 2009–2023.
• Core result: Greater representation of women on boards correlates with reduced tax avoidance (higher effective tax rates).
• Moderation by AI: Firms with higher AI capability show a stronger positive relationship between BGD and tax compliance—AI amplifies the governance role of gender-diverse boards.
• Robustness: Findings hold under alternative tax-avoidance metrics and BGD specifications, heterogeneity checks, and correction methods including Heckman selection, propensity score matching (PSM), and instrumental-variable 2SLS.
• Estimation strategies address key threats: firm fixed effects for unobserved heterogeneity and dynamic two-step System GMM to handle persistence in tax behavior and endogeneity.

Data & Methods

• Data: Balanced panel of 1,586 non-financial firms from developing countries, 2009–2023.
• Dependent variable: Effective tax rate(s) and alternative tax-avoidance measures (noted but unspecified here).
• Key independent variables: Board gender diversity (BGD) measures; AI capability proxied/constructed and entered with a lag.
• Econometric approaches:
  • Firm fixed-effects regressions to control for time-invariant firm heterogeneity.
  • Dynamic two-step System GMM to account for persistence in tax behaviour and potential endogeneity of regressors.
  • Robustness and selection-bias corrections: Heckman selection model, propensity score matching, and 2SLS instrumental-variable estimation.
• Additional checks: Alternative operationalizations of BGD and tax avoidance, heterogeneity analyses (e.g., by firm or country characteristics).

Implications for AI Economics

• AI as a governance complement: AI capability appears to enhance board effectiveness—especially the oversight contribution of female directors—by improving information processing, detection of aggressive tax strategies, and monitoring capacity.
• Policy relevance for developing economies: Investing in firm-level AI capabilities (and supporting digital adoption) can strengthen corporate governance and fiscal discipline, potentially improving tax revenue mobilization.
• Corporate governance design: Regulators and firms should consider complementarities between board composition policies (e.g., gender diversity) and technology adoption when designing interventions to curb tax avoidance.
• Research and measurement priorities: Future AI-economics work should unpack which AI capabilities (analytics, anomaly detection, automation, natural language processing) drive the moderation effect, and how AI interacts with institutional quality, auditor quality, and tax authorities’ capacity.
• Cautions and trade-offs: While AI can improve monitoring, it may also be used to design more sophisticated tax-planning strategies if aligned with managerial incentives. Policymakers should pair AI diffusion with transparency, accountability, and regulatory safeguards.
• Broader labor and distributional concerns: Greater AI adoption that strengthens governance may also change in-house skill demands and alter bargaining with tax advisors—areas for further economic assessment.