NDC不遵守の予測：経済構造、国家能力、気候資金の限界
Forecasting NDC non-compliance: Economic structure, national capacity and the limits of climate finance (原題)
Martins MMV, Cezar RF, Carrer DA, Leal A
🤖 gxceed AI 要約
日本語
本論文は、NDC遵守ギャップ指標（NCGI）を104カ国について2013〜2024年にわたり算出し、2030年目標の未達リスクを6年前に予測できることを示す。機械学習7手法とSHAP、二重機械学習を用い、気候資金の予測可能性・譲許性・断片化は未達を予測せず、ギャップへの識別可能な効果も見られないと結論づける。遵守と最も強く関連するのは生産構造、土地賦存、エネルギー構成、公的機関の質であり、政策の焦点を資金量から実施の構造的条件へ移すべきだと主張する。
English
This paper builds the NDC Compliance Gap Indicator (NCGI) for 104 countries (2013–2024), showing non-compliance risk is predictable six years ahead. Using seven ML algorithms, SHAP, and double machine learning, it finds climate finance volume, concessionality, and fragmentation neither predict nor measurably reduce the gap. Compliance correlates most with productive structure, land endowment, energy profile, and institutional quality—shifting policy focus from finance volume to structural implementation conditions.
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
日本はNDC（2030年度46%削減）とGX推進戦略の整合が問われる段階にあり、資金量よりも産業構造・制度品質が遵守を左右するという知見は、国内の脱炭素政策設計や国際協力（アジアのNDC支援）の議論に示唆を与える。
In the global GX context
Amid global debates on climate finance effectiveness and the post-2025 NDC cycle, this paper challenges the assumption that finance volume drives compliance, offering evidence that structural and institutional factors dominate—relevant to ISSB/TCFD-aligned transition planning and donor accountability discussions.
👥 読者別の含意
🔬研究者:ML・因果推論（DML/SHAP）を気候政策評価に応用する手法論的枠組みとして参考になる。
🏢実務担当者:自社のNDC関連目標やサプライチェーン排出目標の実現可能性を、資金調達だけでなく構造要因から再評価する視点を提供する。
🏛政策担当者:気候資金の設計・配分において、資金量偏重から実施能力・制度構築支援への転換を検討する根拠となる。
📄 Abstract（原文）
<title>Abstract</title> <p> Two thirds of the countries outside Annex I of the United Nations climate convention are not on track to meet the 2030 decarbonization targets declared in their Nationally Determined Contributions (NDCs), and this article identifies them before the deadline. We build the NDC Compliance Gap Indicator for 2030 (NCGI), the distance between projected emissions and each country’s target, recalculated yearly from 2013 to 2024 for 104 countries, and test whether the design of international public climate finance, not only its volume, helps anticipate who falls short. The forecast is validated outside the training period with seven machine learning algorithms; each variable’s contribution is decomposed through Shapley additive explanations (SHAP); and the effect of finance is estimated by double machine learning. Three results stand out: non-compliance risk is predictable six years ahead; including or excluding land use reclassifies 37% of countries between compliance and non-compliance, exposing the fragility of self-declared targets; and climate finance, decomposed into predictability, concessionality and fragmentation, neither anticipates non-compliance nor shows an identifiable effect on the gap. The two estimates that cross the conventional threshold in a single coverage reflect donor selection, shown through a placebo in which finance still to be received anticipates the present gap more strongly than finance already received. What relates best to compliance is productive structure, land endowment, the energy profile and the quality of public institutions, rather than the international support received, moving the focus of climate policy from the volume of finance to the structural conditions of implementation. <bold>JEL codes:</bold> C53; Q56; F35. </p>
🔗 Provenance — このレコードを発見したソース
- Research Square https://doi.org/10.21203/rs.3.rs-11011032/v1first seen 2026-10-07 04:23:42 · last seen 2026-10-11 04:20:56
🔔 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。