氣候相關財務風險影響路徑分析與互動式視覺化決策支援工具之研究

Abstract

隨著全球氣候變遷風險加劇，極端氣候事件逐漸成為經濟與財務系統中最具影響力的風險來源，使氣候風險不再僅限於環境或永續治理議題，而是直接關係國家經濟穩定與企業資本配置。國際永續準則理事會ISSB （International Sustainability Standards Board, ISSB）於 2023 年正式發布 IFRS S1 （一般要求）與 S2 （氣候相關）準則，促使氣候相關財務揭露 TCFD （Task Force on Climate-related Financial Disclosures, TCFD）由自願性揭露框架轉化為具強制力之財務報導準則，企業必須具體說明氣候風險對現金流量與資產價值之財務實質性影響。然而，現行研究與實務間仍存在落差：整合評估模式 IAMs （Integrated Assessment Models, IAMs）多停留於總體尺度，難以對應至企業層級之財務報表項目；傳統財務與資產定價模型則缺乏對氣候衝擊之經濟中介機制的描述，使企業在氣候財務評估過程中面臨「傳導機制缺口」，形成難以追溯與查核的分析黑箱，進而限制揭露品質與監管可信度。 為回應上述理論與實務缺口，本研究建構一套具理論基礎之「氣候相關財務風險傳導矩陣」，並據此發展「互動式視覺化決策支援系統」。本研究目的涵蓋：建立跨尺度、可追溯之氣候風險傳導邏輯，將氣候語言轉譯為企業可理解且可操作之財務語言，並透過軟體工程與情境參數，突破傳統靜態試算表在處理非線性與情境相依風險時之限制。最後建構可供第三方查核之邏輯鏈結構，以回應IFRS、監管機關與查核單位對氣候財務資訊可稽核性與一致性的要求。 本研究採用混合方法論（mixed methodology）作為總體研究設計，並以設計科學研究法（Design Science Research Methodology, DSRM）作為研究架構。首先，透過質化文本分析（Qualitative Textual Analysis）與矩陣分析（Matrix Analysis），並彙整 IPCC、NGFS 與 BIS 等國際氣候權威機構發布之報告與相關學術文獻，建構涵蓋「氣候風險、總體經濟、個體經濟、財務風險與財務報表」之五層次傳導架構，搭配三元編碼邏輯（正相關 +1、負相關 −1、情境相依 0）。其次，採用 Python 與 Streamlit 框架開發決策支援系統，以規則驅動（rule-based）推論引擎與深度優先搜尋 DFS（Depth-First Search, DFS）演算法，解析情境相依節點，並透過桑基圖（Sankey Diagram）技術將高維度風險傳導路徑進行視覺化呈現，白箱化設計亦可將每一條傳導路徑具體文字化，提升決策過程之透明度與可解釋性。 研究結果顯示，氣候實體風險分析結果驗證極端氣候事件經由「基礎設施破壞」與「營運中斷」等傳導路徑，引發全要素生產率 TFP （Total factor productivity, TFP）下滑，並反映於資產負債表中不動產、廠房及設備之報廢損失；氣候轉型風險分析結果顯示碳定價政策壓力促使企業進行防禦性投資，於短期內推升資本支出與折舊費用，並透過「合規對沖效應」降低長期作業風險與融資成本，並維持營業外收支穩定。 本研究補足氣候科學、經濟學、財務管理學與會計學之間的理論斷層，提出一套結構化且可操作之氣候財務風險傳導知識本體；實務上則提供企業一套涵蓋現況診斷、風險傳遞與策略轉型之整合性決策支援工具，藉由將氣候風險映射（Mapping）至財務三表，有助於企業將氣候議題納入預算編制與投資決策流程，降低永續部門與財務部門間之組織隔閡，並為會計師與第三方單位查核提供一致且可追溯之基礎，進而降低漂綠（greenwashing）風險，提升資本市場面對氣候衝擊之整體韌性。

With the increasing risks of global climate change, extreme weather events have gradually emerged as one of the most influential sources of risk within economic and financial systems. Climate risk is no longer confined to environmental or sustainable governance concerns, but has become directly associated with national economic stability and corporate capital allocation decisions. In 2023, the International Sustainability Standards Board (ISSB) formally issued IFRS S1 and IFRS S2, transforming the Task Force on Climate-related Financial Disclosures (TCFD) from a voluntary disclosure framework into a mandatory financial reporting standard. Under the requirements, firms are obligated to explicitly disclose the financial material impacts of climate risks on cash flows and asset values. Nevertheless, a substantial gap persists between existing academic research and real-world implementation. Integrated Assessment Models (IAMs) are predominantly developed at the macroeconomic level and remain difficult to reconcile with firm-level financial statement items, while conventional financial and asset pricing models lack explicit descriptions of the economic transmission mechanisms of climate impacts. Consequently, corporate climate-related financial assessments often face a “transmission mechanism gap,” resulting in an untraceable and unverifiable analytical black box that constrains disclosure quality and undermines regulatory credibility. To address the theoretical and practical gaps identified above, this study develops a theoretically grounded Climate-related Financial Risk Transmission Matrix and an interactive visual decision-support system based on this framework. This research aims to establish a cross-scale and traceable logic of climate risk transmission that translates climate-related information into financial constructs that are understandable and operationalizable for cooperate decision-making, and overcome the limitation of traditional static spreadsheets in capturing non-linear and scenario-dependent risks through software engineering and scenario-based parameterization. Finally, this study constructs an auditable logical-linkage structure to meet the requirements of IFRS, regulators, and assurance providers for the auditability and consistency of climate-related financial information.

This study adopts a mixed-method research design, with the Design Science Research Methodology (DSRM) serving as the overarching research framework. First, qualitative textual analysis and matrix analysis are applied to synthesize reports and relevant academic literature published by leading international climate and financial institutions, including the IPCC, NGFS, and BIS. This process yields a five-layer transmission framework encompassing climate risk, macroeconomics, microeconomics, financial risk, and financial statements, supported by a ternary coding logic (positive correlation +1, negative correlation −1, and scenario-dependent 0). Second, a decision support system is developed using Python and the Streamlit framework, incorporating a rule-based inference engine and a Depth-First Search (DFS) algorithm to resolve scenario-dependent nodes. High-dimensional risk transmission pathways are visualized through Sankey diagrams, while a white-box design explicitly documents each transmission path in textual form, thereby enhancing transparency and interpretability in the decision-making process. The results indicate that extreme climate events, transmitted through pathways such as infrastructure damage and operational disruption, lead to declines in total factor productivity (TFP), which are subsequently reflected in impairment. The climate transition risk analysis further indicates that carbon pricing pressures induce defensive corporate investment, increasing capital expenditure and depreciation expenses in the short term, while generating a “compliance hedging effect” that reduces long-term operational risk and financing costs and stabilizes non-operating income. This study bridges the theoretical fragmentation across climate science, economics, financial management, and operationalizable climate-related financial risk transmission ontology. From a practical perspective, the proposed framework provides firms with an integrated decision support tool encompassing current-state diagnosis, risk transmission analysis, and strategic transition planning. By systematically mapping climate risks onto the three primary financial statements, the framework facilitates the integration of climate considerations into budgeting and investment decision-making processes, reduces organizational silos between sustainability and finance functions, and provides auditors and third-party assurance providers with a consistent and traceable basis for verification. Collectively, this approach helps mitigate greenwashing risks and enhances the overall resilience of capital markets to climate-related impacts.