Scholarlink

<p>In the context of China’s intensified initiatives to combat environmental degradation while simultaneously pursuing rapid economic growth, the government’s increasing focus on ecological governance has emerged as a significant catalyst for sustainable corporate practices. This study investigates the impact of government environmental attention (GEA) on the Environmental, Social, and Governance (ESG) performance of A-share listed companies from 2013 to 2023. Unlike previous studies, this research employs a machine learning-enhanced environmental lexicon and natural language processing (NLP) techniques to quantify GEA based on over a decade of provincial-level government work reports. Utilizing panel regression models and enhanced ESG metrics derived from the Huazheng rating system, the empirical findings indicate that elevated GEA substantially enhances corporate ESG performance. Further analysis reveals that GEA fosters corporate ESG engagement not only directly but also indirectly by facilitating digital transformation and promoting investments in pollution mitigation. Additionally, the effects of GEA are found to vary among firms, influenced by factors such as firm size, exposure to environmental risks, and external regulatory pressures. These findings contribute novel insights into how algorithm-driven textual mining can enrich our understanding of policy signaling and its influence on sustainable corporate behavior at the firm level.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The impact of governmental ecological attention on corporate ESG—empirical evidence from machine learning and text analysis

  • 期刊论文

  • 发表日期：

    2026年3月3日

查看全文
  • Jian-Xiang Ge,
  • Hui Qu

摘要

In the context of China’s intensified initiatives to combat environmental degradation while simultaneously pursuing rapid economic growth, the government’s increasing focus on ecological governance has emerged as a significant catalyst for sustainable corporate practices. This study investigates the impact of government environmental attention (GEA) on the Environmental, Social, and Governance (ESG) performance of A-share listed companies from 2013 to 2023. Unlike previous studies, this research employs a machine learning-enhanced environmental lexicon and natural language processing (NLP) techniques to quantify GEA based on over a decade of provincial-level government work reports. Utilizing panel regression models and enhanced ESG metrics derived from the Huazheng rating system, the empirical findings indicate that elevated GEA substantially enhances corporate ESG performance. Further analysis reveals that GEA fosters corporate ESG engagement not only directly but also indirectly by facilitating digital transformation and promoting investments in pollution mitigation. Additionally, the effects of GEA are found to vary among firms, influenced by factors such as firm size, exposure to environmental risks, and external regulatory pressures. These findings contribute novel insights into how algorithm-driven textual mining can enrich our understanding of policy signaling and its influence on sustainable corporate behavior at the firm level.