Climate risk news and banking industry: A natural language processing approach

Naysary, Babak and Edisen, Ali and Tarazi, Amine (2026) Climate risk news and banking industry: A natural language processing approach. Journal of Financial Stability, 84. p. 101549. ISSN 1572-3089

Preview

Text 1-s2.0-S1572308926000513-main.pdf - Published Version Available under License Creative Commons Attribution. Download (6MB)

Abstract

This study analyzes the evolution of climate-risk discourse in banking using 4887 news articles (2008–2024) collected from ProQuest. We apply Natural Language Processing and add two novel layers: (i) an event-alignment analysis that links coverage dynamics to dated policy and supervisory milestones, and (ii) a discourse-network analysis connecting banks and regulators. We document a marked post-2020 shift, with ESG emerging as the dominant framing (7860 mentions) alongside persistent geographic asymmetries (U.S.-led coverage) and uneven sectoral engagement (Risk Management highest salience; Fintech lowest). Sentiment skews positive (≈4000 positive vs. ≈1500 negative), and topic modeling identifies eight stable thematic clusters spanning operations, ratings, ESG assessment, disclosures, and market instruments. Event alignment shows media attention is typically anticipatory (median peak two months before an anchor), with COP26 producing a sustained level shift (+100% within a ± 6-month window) and the Bank of England’s CBES results generating the largest single spike (210 articles), whereas some 2022 rule-making announcements (e.g., SEC climate-disclosure proposal) exhibit sharper but less durable attention. The discourse network centers on two regulatory hubs (the Federal Reserve and the ECB) with key banks (e.g., Citigroup, JPMorgan, UBS) bridging into supervisory narratives. Collectively, the findings show climate risk becoming embedded in core banking practice while revealing structural, regional, and functional asymmetries that matter for policy design and implementation.

Item Type:	Article
Identification Number:	10.1016/j.jfs.2026.101549
Dates:	Date Event 6 May 2026 Accepted 9 May 2026 Published Online
Uncontrolled Keywords:	Banking, Climate risk, NLP, Text mining, Sentiment analysis, Topic modeling
Subjects:	CAH17 - business and management > CAH17-01 - business and management > CAH17-01-02 - business studies
Divisions:	Business School > Accountancy, Finance and Economics
Depositing User:	Gemma Tonks
Date Deposited:	12 Aug 2026 11:55
Last Modified:	12 Aug 2026 11:55
URI:	https://www.open-access.bcu.ac.uk/id/eprint/17158

Actions (login required)

View Item