ESG emphasis of top companies in East and Southeast Asian countries unveiled by deep learning

Environmental, social, and governance (ESG) considerations are becoming increasingly vital in evaluating corporate performance and addressing global challenges. This study examines the ESG emphasis of leading companies in eight East and Southeast Asian countries through the analysis of 480 corporate reports published in 2024. By employing the Text Match Pre-trained Transformer (TMPT) model, which features advanced zero-shot learning capabilities with 519,939,541 parameters for multi-language analyses, we investigate the relatedness of report contents to 13 ESG-related topics across environmental, social, and economic dimensions. This zero-shot framework enables accurate and language-independent ESG analysis without the need for task-specific fine-tuning, allowing consistent comparison across countries. Our findings reveal that economic topics, including governance risk, production cost, and economic ripple effect, dominate corporate emphasis, while significant regional variations exist. Specifically, South Korean companies prioritize environmental and social issues, whereas Chinese firms focus more heavily on economic aspects. Japanese corporations demonstrate a balanced pattern, emphasizing work environment and domestic job creation, while companies in Indonesia, Malaysia, and the Philippines emphasize community-related topics and human rights. This research highlights the capabilities of AI-driven models in investigating specific patterns in corporate ESG emphasis across multilingual datasets. By providing a detailed understanding of regional ESG priorities and tendencies, this study demonstrates emphasis variations among countries, sectors, and topics to support societal efforts towards accountability and sustainable development. • Deep learning maps ESG focus in 480 top Asian corporate reports. • Economic topics dominate; social and environmental vary by country. • Sector influence outweighs country in shaping ESG disclosure patterns. • Multilingual report consistency strong in China, Japan, Korea; weaker elsewhere. • TMPT model enables accurate, scalable ESG analysis across languages.

Authors

Institutions

Publication Details

Journal
Green Technologies and Sustainability
Published
2025-10-15
DOI
https://doi.org/10.1016/j.grets.2025.100296
Primary Topic
International Business and FDI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

ESG emphasis of top companies in East and Southeast Asian countries unveiled by deep learning

Shunsuke Managi, Alexander Ryota Keeley, Chao Li
Green Technologies and Sustainability
International Business and FDI
article

ESG emphasis of top companies in East and Southeast Asian countries unveiled by deep learning

Shunsuke Managi, Alexander Ryota Keeley, Chao Li
article en

Abstract

Environmental, social, and governance (ESG) considerations are becoming increasingly vital in evaluating corporate performance and addressing global challenges. This study examines the ESG emphasis of leading companies in eight East and Southeast Asian countries through the analysis of 480 corporate reports published in 2024. By employing the Text Match Pre-trained Transformer (TMPT) model, which features advanced zero-shot learning capabilities with 519,939,541 parameters for multi-language analyses, we investigate the relatedness of report contents to 13 ESG-related topics across environmental, social, and economic dimensions. This zero-shot framework enables accurate and language-independent ESG analysis without the need for task-specific fine-tuning, allowing consistent comparison across countries. Our findings reveal that economic topics, including governance risk, production cost, and economic ripple effect, dominate corporate emphasis, while significant regional variations exist. Specifically, South Korean companies prioritize environmental and social issues, whereas Chinese firms focus more heavily on economic aspects. Japanese corporations demonstrate a balanced pattern, emphasizing work environment and domestic job creation, while companies in Indonesia, Malaysia, and the Philippines emphasize community-related topics and human rights. This research highlights the capabilities of AI-driven models in investigating specific patterns in corporate ESG emphasis across multilingual datasets. By providing a detailed understanding of regional ESG priorities and tendencies, this study demonstrates emphasis variations among countries, sectors, and topics to support societal efforts towards accountability and sustainable development. • Deep learning maps ESG focus in 480 top Asian corporate reports. • Economic topics dominate; social and environmental vary by country. • Sector influence outweighs country in shaping ESG disclosure patterns. • Multilingual report consistency strong in China, Japan, Korea; weaker elsewhere. • TMPT model enables accurate, scalable ESG analysis across languages.

Green Technologies and SustainabilityVol. 4(2)
Kyushu University (JP)
Openalex Percentile: Top 57%
International Business and FDI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.