グリーン転換におけるグリーンファイナンス政策と企業生産性：中国パイロットゾーンからの証拠
Green Finance Policy and Firm Productivity in the Green Transition: Evidence from China’s Pilot Zones (原題)
Shaozhe Li, Hanxiao Li, Kunming Li
🤖 gxceed AI 要約
日本語
中国のグリーンファイナンス改革・イノベーション・パイロットゾーン政策を準自然実験として、DDDとDouble Machine Learningで企業生産性への影響を識別。政策はグリーン企業の全要素生産性を有意に高める一方、重汚染企業では低下させる。重汚染企業ではグリーン技術革新と資本配分効率の両方が低下し、デジタル転換が悪影響を緩和しうることを示す。
English
Using China's Green Finance Reform and Innovation Pilot Zones as a quasi-natural experiment, this study combines triple-difference and Double Machine Learning to identify productivity effects. The policy raises total factor productivity for green firms but lowers it for heavily polluting firms, partly by reducing their green innovation and capital allocation efficiency. Digital transformation mitigates the adverse effects, and the policy improves governance among green firms.
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
中国のグリーンファイナンス政策が企業生産性に与える非対称な影響を示し、日本のGX推進における移行金融や産業政策の設計（特に高排出企業への支援バランス）を考える上で示唆に富む。SSBJや有報での移行計画開示を検討する日本企業にとって、政策が生産性に及ぼす実証的エビデンスとして参照価値がある。
In the global GX context
Provides empirical evidence on how green finance pilot policies can create asymmetric productivity effects across firm types, informing global debates on transition finance design under ISSB/TCFD frameworks. It highlights the need for differentiated policy support and governance mechanisms to ensure an inclusive green transition, relevant to CSRD and SEC climate disclosure discussions on transition planning.
👥 読者別の含意
🔬研究者:グリーンファイナンス政策の企業生産性への異質な影響をDDDとDDMLで識別する実証手法と、そのメカニズム分析が参考になる。
🏢実務担当者:グリーンファイナンス政策が自社の生産性や資本配分に与えうる影響を理解し、デジタル転換やガバナンス強化を通じた対応を検討する材料になる。
🏛政策担当者:グリーンファイナンス政策の設計において、環境制約と企業支援のバランス、差別化された政策措置、ガバナンス強化の重要性を示唆する。
📄 Abstract（原文）
Green finance has become an important policy instrument for promoting sustainable development, but its effects on firm productivity may vary across firms with different environmental attributes. This study develops a theoretical framework in which green finance incorporates environmental performance into credit allocation and capital pricing, thereby reshaping firms’ financing conditions, innovation incentives, and resource allocation efficiency. The framework suggests that green finance may create productivity-enhancing incentives for green firms while imposing transition-related pressures on heavily polluting firms. Using data on Chinese A-share listed firms from 2009 to 2024, this study exploits China’s Green Finance Reform and Innovation Pilot Zones policy as a quasi-natural experiment and combines a triple-difference (DDD) model with a Double Machine Learning (DDML) framework to identify the policy’s productivity effects. The results show that the policy significantly increases the total factor productivity of green firms but reduces that of heavily polluting firms. Mechanism analysis indicates that the policy reduces both green technological innovation and capital allocation efficiency among heavily polluting firms. For green firms, the policy promotes green technological innovation but reduces capital allocation efficiency. Heterogeneity analysis shows that these effects vary with government subsidies, organizational slack, and industry asset intensity. Further analysis suggests that digital transformation may mitigate the policy’s adverse productivity effect on heavily polluting firms, while the policy improves governance among green firms by reducing earnings management and controlling-shareholder expropriation. These findings suggest that green finance policy should balance environmental constraints with support for firms’ green transition, adopt differentiated policy designs, and strengthen governance mechanisms to promote sustainable productivity growth and an inclusive green transition.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.3390/su182111036first seen 2026-11-01 04:51:07 · last seen 2026-11-06 04:45:07
🔔 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。