← 論文一覧に戻る

東南アジア銀行業におけるAI開示、ESGパフォーマンス、トービンのQ

AI Disclosure, ESG Performance, and Tobin’s Q in Southeast Asian Banking (原題)

Hugo Prasetyo Winotoatmojo, Meiryani Meiryani, Bambang Leo Handoko, Antonius Ary Setyawan

Inkubis Jurnal Ekonomi dan Bisnis📚 査読済 / ジャーナル2026-10-21#AI×ESG経営インパクト: 資金調達対象セクター: finance
DOI: 10.59261/inkubis.v8i4.494
原典: https://doi.org/10.59261/inkubis.v8i4.494

🤖 gxceed AI 要約

日本語

東南アジア10カ国・205行の2020〜2025年パネルデータ（1,230観測）を用い、AI開示とESG性能、トービンのQの関連を検証。AI開示は同時点のESG性能（特に社会・ガバナンス）と正に関連するが、ラグ付き・1年先では有意でなく、トービンのQとの一貫した関連やESG経由の間接効果も確認されなかった。AI開示は「シグナル」に留まり、検証可能な成果連動型開示の必要性を示唆する。

English

Using a balanced panel of 1,230 bank-year observations across 205 banks in 10 Southeast Asian countries (2020–2025), this study examines links between AI disclosure, ESG performance, and Tobin's Q. AI disclosure is positively associated with contemporaneous ESG performance (notably social and governance), but not in lagged or one-year-ahead models, and shows no consistent valuation relevance via ESG. Findings frame AI disclosure as a potential signal and call for verifiable, outcome-linked reporting.

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

📝 gxceed 編集解説 — Why this matters

日本のGX文脈において

SSBJ・有報でのAI関連開示が拡大する日本企業にとっても、AI開示が実装や成果を伴わなければ評価・企業価値に結びつかないという警鐘は実務的示唆が大きい。統合報告書でのAI記述の質的向上を検討する日本企業の参考になる。

In the global GX context

As ISSB/CSRD and SEC climate disclosure regimes expand, this paper adds rare empirical evidence that AI-related disclosure may function as a signal without valuation relevance unless tied to verifiable outcomes—relevant to global debates on greenwashing and disclosure quality.

👥 読者別の含意

🔬研究者:AI開示とESG・企業価値の関連を同時点・ラグ・間接効果まで分解した実証設計は、開示研究の因果推論設計の参考になる。

🏢実務担当者:AI関連の開示は成果や実装と結びつけない限り評価に反映されにくく、開示の質と検証可能性を高める必要がある。

🏛政策担当者:AI開示の枠組み設計では、シグナル偏重を避け成果連動型・検証可能な報告を促す仕組みが求められる。

📄 Abstract（原文）

Background: Banks increasingly disclose AI-related activities, but disclosure does not establish actual implementation or realized outcomes. Evidence remains limited regarding whether AI disclosure is associated with ESG performance and whether ESG performance subsequently has valuation relevance in Southeast Asian banking. Objective: This study examines the associations among AI disclosure, ESG performance, and Tobin’s Q. Methods: A balanced panel of 1,230 bank-year observations covering 205 banking entities in 10 Southeast Asian countries from 2020 to 2025 was analyzed using bank and country-year fixed effects with bank-clustered standard errors. Robustness checks included lagged specifications, alternative variable definitions, ESG-pillar models, country exclusions, and cluster bootstrap analysis of the indirect association. Results: AI disclosure was positively associated with contemporaneous ESG performance (β = 0.0139, p = 0.006; within R² = 0.007), with significant associations for social (β = 0.0181, p = 0.004) and governance (β = 0.0096, p = 0.018) performance. However, the disclosure–ESG association was not significant in lagged or one-year-ahead specifications. AI disclosure was not consistently associated with Tobin’s Q, and the bootstrap confidence interval for the indirect association included zero. Conclusion: AI disclosure is associated with ESG performance only contemporaneously and does not provide consistent evidence of valuation relevance through ESG. The findings support a cautious interpretation of AI disclosure as a potential signal and underscore the need for verifiable, outcome-linked reporting.

🔗 Provenance — このレコードを発見したソース

🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。

gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。