Study: Cat Models Underestimate US Wildfire Exposure by 40%
A peer-reviewed study in Nature Climate Change finds that major catastrophe risk models systematically underestimate wildfire exposure in the wildland-urban interface by approximately 40%, based on 15 years of US wildfire data. The findings have direct implications for reserve adequacy and reinsurance pricing across Property and Energy books. This is a modelling research finding rather than an insured loss event, but it represents a structural challenge to underwriting assumptions market-wide.
AI-generated from linked source reports. See our correction policy.
Impact verdict
Medium impact. MEDIUM: A 40% systematic underestimation of US wildfire exposure in cat models has significant commercial implications for Property and Reinsurance books — affecting reserve adequacy, reinsurance pricing, and treaty structures at renewal. This is not an acute loss event, but a structural modelling finding that underwriters and reinsurers relying on these models will need to address. Market-wide pricing and capacity decisions for US wildfire-exposed risks may be affected.
View assessment methodologyHow we grade what we know -- Known · Reported · Uncertain. Methodology →
Intelligence ledger
Each line expands in place to its underlying sourced claim.
Known4 lines
Peer-reviewed study published in Nature Climate Change▾
Study analyzed 15 years of US wildfire data against modeled losses from three major cat modeling firms▾
Models found to underestimate wildland-urban interface wildfire exposure by approximately 40%▾
Findings have implications for reserve adequacy and reinsurance pricing▾
Reported2 lines
Three unnamed major catastrophe modeling firms were assessed in the study▾
The underestimation is described as systematic rather than isolated▾
Uncertain4 lines
Which specific cat modeling firms were included in the study▾
Whether underestimation is uniform across all geographic regions or concentrated in specific high-risk areas▾
Timeline for cat model updates or market response▾
Quantified dollar impact on industry reserves or pricing adjustments required▾
Geographic Zone Matches
1 active match
- TRIA Certified AreasRule-basedConfidence 100%
Geographic zone matches are RiskEvents spatial/analytical indicators, not coverage determinations or Lloyd's official classifications.
Affected countries
Timeline
Lifecycle changed
signal → closed
Event Closed
Seeded/test data cleanup: synthetic scenario row from 2026-05-24 demo batch; should not appear in the current public RiskEvents feed.
Initial Detection
A peer-reviewed study in Nature Climate Change finds that major catastrophe risk models systematically underestimate wildfire exposure in the wildland-urban interface by approximately 40%, based on 15 years of US wildfire data. The findings have direct implications for reserve adequacy and reinsurance pricing across Property and Energy books. This is a modelling research finding rather than an insured loss event, but it represents a structural challenge to underwriting assumptions market-wide.
Current catastrophe risk models systematically underestimate wildfire exposure in the wildland-urban interface by approximately 40%. The study analyzed 15 years of US wildfire data against modeled losses from three major cat modeling firms. The findings have implications for reserve adequacy and reinsurance pricing.
Lloyd's classifications
Tracking this kind of risk? Get an email when Natural Catastrophe events escalate.
Get alerts