Marriott International Inc
An asset is classified as uninsurable when 50% or more of its value could be lost within 5 years with at least a 5% probability.
Period
2025 to 2030
Insurable Assets
S&P 500
MSCI World
How RTai defines uninsurable risk
RTai measures when climate risk stops being manageable and starts becoming structural. Our methodology is fully stochastic, globally consistent, and built around one clear threshold.
The 50 × 5 × 5 Rule
An asset is classified as at risk when 50% or more of its value could be lost within 5 years with at least a 5% probability
50%
or more of an asset's value
5 years
could be lost within 5 years
5%
with at least 5% of probability
The line between insurable and uninsurable.
This threshold reflects the point at which losses become incompatible with traditional insurance models, and where risk begins shifting onto balance sheets and public systems.
Of course, different insurance companies might have different cutoff levels, which we can accommodate.
Step 1: Data
Map the assets
We identify real physical assets and locate them precisely where climate risk occurs. Regions only appear where assets exist. Every input is sourced, documented, and auditable.
Step 2: Modeling
Simulate what could happen
Using a fully stochastic engine, we run thousands of climate scenarios per asset. This captures rare but severe losses—not just average outcomes—using the same methodology worldwide.
Step 3: Output
Show the risk clearly
Results are translated into a simple classification: Green for low risk, orange for rising pressure, red where assets cross the uninsurable threshold under the 50 × 5 × 5 Rule.
The Probability
Gap
The first financial ranking built to correct a fundamental market error.
For decades, financial models have priced catastrophic climate risk at roughly five percent, even though scientific evidence shows the probability is far higher. This gap has left assets, portfolios, and entire sectors mispriced.
The Climate Insurability Rankings close this gap by measuring the real probability of severe loss across 13,000 global companies. By mapping physical assets and simulating climate extremes, we provide a forward-looking signal that reflects how the world is actually changing, not how legacy models assume it behaves.
How the
Rankings Work
A three-layer engine built for accuracy at scale.
01
Asset-level mapping
We identify and locate millions of company assets across 13,000+ publicly listed firms, creating the most complete picture of physical exposure available today.
02
Stochastic climate simulations
We run ensembles of global climate models to capture extremes, compounding hazards, and future uncertainty. This approach reveals risk profiles that annual averages cannot detect.
03
Financial probability scoring
We convert these simulations into a single metric: the probability that a company's material assets face severe loss. This probability is then translated into a rank comparable across sectors and indices.
Why This
Matters
Climate risk is now a direct driver of financial outcomes.
As climate extremes intensify, physical risk is moving from footnote to balance-sheet driver. Insurers are reassessing coverage and pricing. Lenders are tightening climate-related due diligence. Regulators are running system-wide climate stress tests across entire financial sectors.
The Climate Insurability Rankings provide a common language for these decisions: a science-based, forward-looking measure of how exposed a company's assets are to severe climate loss, and how that exposure compares within its sector and chosen index.
Trusted by global
financial leaders.
The exclusive physical risk engine for the Bloomberg Terminal, powering insights for 350,000+ market participants.
The chosen platform for Canada's mandatory, system-wide climate stress tests for 400+ financial institutions.
From the founder who built the global standard for enterprise risk, trusted by 70% of the world's top 100 banks.
Uninsurability is no longer theoretical.
RTai gives institutions a shared, measurable language to see where risk is accumulating and where intervention is still possible.
Our results are derived from open-source data and, in some cases, lack proprietary adaptation information, which could reduce asset risk. Naturally, this is addressed when applied in practice, where such information is available.