What question did this study set out to answer?

This study aims to identify the drivers of corporate greenwashing using the IMOC framework. It analyzes how institutional pressures and organizational opportunities influence this behavior.

April 27, 2026Open Access

What drives corporate greenwashing? A machine learning lens on pressure and opportunity in China’s A-share listed firms

Key Points

This study aims to identify the drivers of corporate greenwashing using the IMOC framework. It analyzes how institutional pressures and organizational opportunities influence this behavior.
Employs a comparative machine learning approach to identify factors influencing corporate greenwashing in China from 2011 to 2022.
Benchmarks LightGBM, XGBoost, and random forest algorithms against logistic regression for prediction effectiveness.
Uses SHAP values for feature interpretation to understand influences on greenwashing.
LightGBM outperforms other models in predicting greenwashing.
Institutional ownership concentration is identified as a strong positive predictor; moderate digital transformation and market competition show inhibitory effects.
Digital economy development enhances media monitoring and mitigates principal-agent conflict effects on greenwashing.

Abstract

Purpose This study aims to systematically identify the drivers of corporate greenwashing through the institution, market, organization and cognition (IMOC) framework. Grounded in institutional theory, resource-based view, agency theory and upper echelons theory, this research examines how factors across these four dimensions drive greenwashing through the dual mechanisms of institutional pressures and organizational opportunities, while establishing an effective prediction framework using machine learning techniques. Design/methodology/approach Employing a comparative ML approach, we comprehensively identify the factors affecting corporate greenwashing based on the pressure and opportunity dimensions and then explore how these factors influence corporate greenwashing by a sample of A-share listed firms in China from 2011 to 2022. The LightGBM, XGBoost and random forest algorithms are benchmarked against traditional logistic regression, with SHapley Additive exPlanations (SHAP) values used for feature interpretation. Findings The paper finds that (1) LightGBM outperforms other models in greenwashing prediction. (2) Institutional ownership concentration emerges as the strongest positive predictor, while moderate digital transformation and market competition exhibit inhibitory effects. (3) Digital economy development amplifies media monitoring efficacy and reduces principal-agent conflicts’ impact on greenwashing. Practical implications Regulators should prioritize institutional investors’ governance roles and establish digital transformation thresholds for environmental, social and governance (ESG) compliance. Investors can utilize our ML framework to assess greenwashing risks in portfolio companies. Originality/value This research pioneers the integration of institutional theory with explainable artificial intelligence in greenwashing detection, revealing the non-linear impacts of digital transformation. The proposed SHAP-empowered framework enables dynamic monitoring of emerging ESG risks.

What drives corporate greenwashing? A machine learning lens on pressure and opportunity in China’s A-share listed firms

Key Points

Abstract

Cite This Study