Abstract
<jats:p>The paper aims to develop an approach for assessing the quality of corporate governance in ESG ratings for regional companies in the Northwestern Federal District (NWFD) of Russia. A distinctive feature of these companies is that they, firstly, operate within the regional economy, and secondly, are not always publicly listed entities that come under the scrutiny of federal rating agencies. Based on a literature review and existing methodologies for measuring corporate governance quality in commercial and academic ratings, a set of indicators has been proposed for this assessment. Using the proposed set, a comparative study was conducted on the quality of corporate governance in major federal companies and NWFD companies that are not included in federal ratings. The approach outlined in the paper is implemented in two stages. In the first stage, the rating assessment of corporate governance quality is based on the degree of data disclosure on company websites. In the second stage, artificial intelligence algorithms trained on various publicly available data (company websites, news articles, and reports) are utilized. Despite a small sample size and significant discrepancies identified in some cases between the assessment results in the first and the second stage, a sufficiently high accuracy in predicting corporate governance quality ratings was achieved when training artificial intelligence algorithms. Additionally, by comparing ratings obtained from both stages, insights into the quality of corporate governance in the evaluated companies and directions for its improvement can be gained. The conclusion drawn emphasizes the necessity of employing various tools for an accurate assessment of corporate governance quality.</jats:p>