Climate change is repricing financial assets. Floods, wildfires, and rising seas destroy buildings, disrupt supply chains, and erode property values. But how do you put a dollar figure on that exposure? Climate value at risk (CVaR) answers that question. It translates physical climate hazards into estimated monetary losses, giving banks, insurers, and corporates a single metric for climate-driven financial exposure.
This guide covers what climate value at risk means, how the calculation works, where MSCI’s approach differs from physical damage curve methods, and how organizations use CVaR for portfolio screening, disclosure, and capital allocation.
What Is Climate Value at Risk?
Climate value at risk is a metric that estimates the financial loss a building, portfolio, or enterprise could suffer from future climate hazards. It borrows the concept from traditional Value at Risk (VaR) in finance, but replaces market volatility with climate exposure.
Traditional VaR asks: “What’s the most I could lose on this portfolio in a given period at a given confidence level?” Climate value at risk asks the same question, but the risk driver is a flood, a heatwave, or a hurricane instead of a market downturn.
The metric matters because regulators, investors, and lenders increasingly demand quantified climate risk. The Task Force on Climate-related Financial Disclosures (TCFD) specifically calls for scenario-based financial impact analysis. CVaR provides exactly that: a dollar-denominated climate risk figure tied to specific scenarios and time horizons.
Banks use it for mortgage portfolio screening. Insurers use it for underwriting and reserve pricing. Corporates use it to identify which facilities carry the highest flood or wildfire exposure. Asset managers use it to compare climate risk across holdings.
How Climate Value at Risk Is Calculated
The Core Formula
At the building level, climate value at risk starts with three inputs:
Expected Loss = Vulnerability x Hazard x Exposure
- Vulnerability is the damage ratio: what fraction of the building’s value gets destroyed at a given hazard intensity. This comes from depth-damage curves (for floods) or wind fragility functions (for cyclones). The output is a number between 0.0 and 1.0.
- Hazard is the intensity of the climate event at that location. For floods, it’s water depth in feet or meters. For heat stress, it’s temperature. Climate models like NASA NEX-GDDP-CMIP6 project these values under different scenarios (SSP2-4.5, SSP5-8.5) out to 2050 and beyond.
- Exposure is the asset value at risk: the building’s replacement cost, contents value, or both.
Multiply these three together and you get the estimated loss from a single event at a single location.
From Building Loss to Portfolio-Level CVaR
A single-event loss estimate is useful but incomplete. Climate value at risk for a portfolio requires integrating losses across multiple return periods (10-year flood, 50-year flood, 100-year flood, 500-year flood) to produce an Expected Annual Loss (EAL):
EAL = Sum of [Loss at each return period x Annual probability of that event]
A 100-year flood has a 1% annual probability. If a building would lose $500,000 in that flood, the contribution to EAL is $5,000 per year. Add up contributions across all return periods, and you get the average yearly loss the building faces from flooding alone.
For enterprise-level CVaR, the formula becomes:
CVaR = Present Value of Future Climate Costs / Market Value of Enterprise
This discounts future losses to present value, accounting for when they’re expected to hit and at what rate money is discounted.
Worked Example
Consider a $10 million commercial warehouse (HAZUS occupancy type COM2: Wholesale Trade) in a riverine flood zone. Climate projections under SSP2-4.5 indicate a 100-year flood depth of 4 feet at this location by 2050.
Using the HAZUS depth-damage curve for COM2, 4 feet of flooding produces approximately 30% structural damage and 35% contents damage. With contents valued at 100% of replacement (the HAZUS default for commercial buildings):
Damage ratios vary significantly across the HAZUS commercial occupancy classes — retail (COM1), warehouse (COM2), office (COM4), hospital (COM6), and six others each follow distinct depth-damage curves, so the same 4-foot flood can produce materially different losses depending on building type.
- Structural loss: $10M x 0.30 = $3.0 million
- Contents loss: $10M x 0.35 = $3.5 million
- Total single-event loss: $6.5 million
- Contribution to EAL (1% annual probability): $65,000 per year
Repeat this calculation across all return periods (10-year, 50-year, 100-year, 500-year), sum the contributions, discount to present value, and you have the climate value at risk for that single building. Scale to a portfolio of hundreds or thousands of buildings, and the same method applies.
Physical vs. Transition Climate Value at Risk
Physical CVaR
Physical climate value at risk captures losses from actual climate events. These split into two categories:
- Acute hazards: Floods, wildfires, cyclones, storm surge. Sudden events that cause immediate damage.
- Chronic hazards: Heat stress, water scarcity, sea level rise, subsidence. Gradual shifts that erode asset values over years.
Physical CVaR depends on location. A coastal warehouse and an inland warehouse face different physical risks even if they’re identical buildings owned by the same company.
Transition CVaR
Transition climate value at risk captures losses from the economic shift away from fossil fuels:
- Policy risk: Carbon pricing, emissions caps, and stranded fossil fuel assets.
- Technology risk: Renewable energy making carbon-intensive operations uncompetitive.
- Market and liability risk: Changing consumer preferences, climate litigation, and insurance withdrawal from high-risk areas.
Transition CVaR depends on sector. An oil refinery faces high transition risk regardless of its geographic location.
Why Both Matter
A complete climate risk picture requires both. A bank’s mortgage portfolio faces physical CVaR (flood damage to properties), while its corporate loan book faces transition CVaR (borrowers in carbon-intensive sectors losing value). TCFD and ISSB frameworks expect organizations to assess and disclose both.
MSCI Climate Value at Risk vs. Physical Risk Approaches
MSCI’s Portfolio-Level Standard
MSCI Climate VaR is the most widely adopted climate value at risk framework in institutional investing. It covers 10,000+ publicly traded companies and 9,000+ sovereign bonds. MSCI calculates CVaR by modeling how climate-related costs (carbon prices, technology shifts, extreme weather) affect a company’s projected cash flows under different warming scenarios.
MSCI’s strength is breadth: it provides a standardized, comparable metric across entire equity and bond portfolios. Its limitation is granularity. MSCI operates at the company level, not the building level. It cannot tell you which specific warehouse in your portfolio faces the highest flood risk.
Physical Risk CVaR Using Damage Curves
Physical risk CVaR takes the opposite approach: it starts at the individual building and works up. Using depth-damage curves from FEMA’s HAZUS framework (33 building types, 196 curves for the US) or the JRC Huizinga methodology (6 sectors, 214 countries globally), this approach calculates damage ratios at specific flood depths for specific building types.
The result is asset-level precision. You know exactly which buildings in a 500-property portfolio carry the most flood exposure, what the damage estimate is at different flood depths, and how that changes under different climate scenarios.
Continuuiti’s climate value at risk tool implements both HAZUS and JRC methodologies, allowing users to estimate flood damage for individual buildings or portfolios of up to 5,000 properties without installing GIS software or handling raw climate data.
How Damage Curves Power Climate Value at Risk
Depth-Damage Functions
Damage curves are the mathematical backbone of physical climate value at risk. They map a hazard intensity (flood depth, in this case) to a damage ratio (the fraction of the building’s value destroyed).
These curves follow a characteristic sigmoid shape. At shallow depths (below the first floor), damage is minimal. Between 0 and 8 feet above the first floor, damage rises steeply as water contacts flooring, electrical systems, HVAC equipment, and progressively higher contents. Above 8 feet, the curve flattens because the building’s structural frame survives even deep inundation.
HAZUS provides 196 individual curves covering 33 building types across three flood zones (riverine, coastal A, coastal V). Structure and contents damage are calculated separately. The free flood damage calculator lets you explore these curves interactively for any building type and flood depth.
JRC Huizinga provides global coverage with curves for 6 sectors across 7 continental groupings. JRC curves combine structural and contents damage into a single ratio and provide standard deviations where available.
Key Inputs for Damage Estimation
Calculating climate value at risk from damage curves requires four inputs per building:
- Flood depth above ground grade (from climate models or flood maps)
- Occupancy type (residential, commercial, industrial, with specific subcategories)
- First floor height (the vertical distance from ground to the first finished floor)
- Replacement value (structural cost, plus contents value)
First floor height is particularly important. Two identical buildings, one with a basement (4-foot FFH) and one on a slab (1-foot FFH), will show very different damage at the same flood depth.
From Damage Ratios to Expected Annual Loss
A single damage ratio answers: “If this building floods to 4 feet, how much damage occurs?” Expected Annual Loss (EAL) answers a harder question: “Across all possible flood events, weighted by their probability, what’s the average annual financial loss?”
EAL integrates damage across the full range of flood return periods. Each return period contributes its loss estimate multiplied by its annual exceedance probability. The sum is the expected yearly cost of flood risk at that location. EAL is the core output of physical climate value at risk and feeds directly into catastrophe modeling and climate risk modeling frameworks used for TCFD and ISSB reporting.
Climate Value at Risk for Financial Decision-Making
Climate value at risk has moved from academic concept to operational tool across financial services:
- Portfolio risk scoring: Rank buildings or loans by CVaR to identify concentration risk. A bank with $2 billion in coastal Florida mortgages needs to know its aggregate flood CVaR.
- Real estate due diligence: Before acquiring or refinancing commercial property, run a property risk assessment and estimate the climate value at risk to price in physical risk exposure.
- Insurance underwriting: Insurers use damage curves and CVaR to price premiums and set coverage limits. As climate risk intensifies, actuarial pricing increasingly incorporates forward-looking CVaR alongside historical loss data.
- TCFD and ISSB disclosure: Both frameworks require quantified scenario-based financial impact analysis. CVaR provides exactly the metric they call for.
- Regulatory capital: The ECB, Bank of England, and other central banks require climate stress testing. CVaR at the portfolio level feeds directly into these stress test scenarios.
Limitations
Climate value at risk is a screening-level metric, not a precision instrument. Key limitations:
- Damage curves are point estimates. HAZUS provides no confidence intervals. Research has found the upper bound of flood loss estimates can be 3x the lower bound depending on methodology choices (Tate et al., ASCE Natural Hazards Review).
- Geographic constraints. HAZUS covers the US only. JRC provides global coverage but uses continental averages, meaning a building in Germany and one in Greece get the same European curve.
- Temporal uncertainty. Climate projections extend decades into the future. The difference between SSP2-4.5 and SSP5-8.5 scenarios can double the projected flood depth at a given location.
- No adaptation modeling. Standard curves assume no flood-proofing, barriers, or elevation. Buildings that have been flood-proofed will sustain less damage than curves predict.
- Input sensitivity. CVaR is only as accurate as the flood depth input. Small changes in projected depth, especially in the steep 0-8 foot range of damage curves, can produce large changes in estimated loss.
These limitations don’t make climate value at risk useless. They make it a screening tool: appropriate for portfolio-level risk ranking and aggregate exposure quantification, not for individual property insurance pricing.
Frequently Asked Questions
What is climate value at risk?
Climate value at risk (CVaR) is a metric that estimates the financial loss a building, portfolio, or enterprise could suffer from future climate hazards like floods, wildfires, and sea level rise. It translates physical climate risk into a dollar figure that banks, insurers, and corporates use for risk management and regulatory disclosure.
How do you calculate climate value at risk?
Climate value at risk is calculated by multiplying three components: Vulnerability (damage ratio from depth-damage curves), Hazard (climate event intensity from climate models), and Exposure (asset replacement value). For portfolio-level CVaR, losses are integrated across multiple return periods to produce Expected Annual Loss, then discounted to present value.
What is MSCI climate value at risk?
MSCI Climate VaR is the most widely adopted portfolio-level climate risk metric in institutional investing. It covers 10,000+ companies and models how climate-related costs affect projected cash flows under different warming scenarios. MSCI focuses on enterprise-level exposure, while physical risk CVaR approaches use building-level damage curves for asset-specific granularity.
Why is climate value at risk important for banks?
Banks use climate value at risk to screen mortgage portfolios for flood and climate exposure, satisfy regulatory stress testing requirements from the ECB and Bank of England, meet TCFD and ISSB disclosure obligations, and identify concentration risk in coastal or flood-prone regions. CVaR provides the quantified, scenario-based metric that regulators expect.
What is the difference between CVaR and VaR?
Traditional Value at Risk (VaR) measures potential financial loss from market movements like stock price changes or interest rate shifts. Climate Value at Risk (CVaR) measures potential financial loss from climate events like floods, wildfires, and policy changes. Both express risk as a dollar figure, but the underlying risk drivers are completely different.
