Climate risk analytics & scenario modeling KPIs by sector (with ranges)

Climate risk analytics has moved from an academic exercise to a regulatory and fiduciary obligation across financial services, real estate, energy, and agriculture. Yet organizations deploying these tools frequently struggle to measure whether their investments are delivering actionable insights or generating elaborate reports that collect dust. A 2025 survey by the Network for Greening the Financial System (NGFS) found that 68% of financial institutions had implemented some form of climate scenario analysis, but only 23% could demonstrate that the results had materially influenced a lending, underwriting, or investment decision. The gap between deployment and decision-relevant output defines the KPI challenge in this domain: measuring what matters requires tracking not just technical model performance but operational integration and strategic impact.

Why It Matters

Regulatory mandates are the primary driver. The European Central Bank's supervisory expectations require banks to integrate climate risk into their Internal Capital Adequacy Assessment Processes (ICAAP) and demonstrate scenario analysis capabilities covering both physical and transition risks. The Bank of England's Climate Biennial Exploratory Scenario (CBES) tested major UK financial institutions against three climate pathways, revealing that cumulative losses could reach 10-15% of annual profits under disorderly transition scenarios. The Monetary Authority of Singapore mandates climate stress testing for domestically systemically important banks. In the US, the Federal Reserve's pilot climate scenario analysis exercise covered the six largest US banks in 2023, with expanded requirements expected.

Beyond financial services, the EU's Corporate Sustainability Reporting Directive (CSRD) requires all in-scope companies to disclose how climate-related risks and opportunities affect their business model, strategy, and financial planning under different climate scenarios. The European Sustainability Reporting Standards (ESRS E1) specifically mandate disclosure of financial effects from physical and transition risks, with quantitative estimates where feasible.

The business case extends beyond compliance. Munich Re estimated that global insured losses from natural catastrophes reached $145 billion in 2024, a 23% increase over the five-year average. Companies that integrate climate scenario analysis into capital allocation, supply chain planning, and real estate decisions demonstrably outperform peers in risk-adjusted returns. A 2025 study by the Cambridge Institute for Sustainability Leadership found that companies with mature climate risk analytics capabilities experienced 35% lower earnings volatility during climate-related disruption events compared to industry peers without such capabilities.

Key Concepts

Physical Risk Assessment quantifies the financial impact of acute climate hazards (hurricanes, floods, wildfires, heatwaves) and chronic shifts (sea-level rise, precipitation pattern changes, temperature increases) on specific assets, supply chains, and revenue streams. Leading approaches combine global climate model outputs (CMIP6 ensembles) with localized hazard models, asset-level exposure data, and vulnerability functions that translate hazard intensity into financial loss estimates. The critical technical challenge is downscaling: global climate models operate at 50-100 km resolution, while risk decisions require property-level granularity at 30-250 meter resolution.

Transition Risk Modeling estimates financial impacts from policy changes, technology disruption, market shifts, and reputational effects associated with the transition to a low-carbon economy. Common approaches include carbon price sensitivity analysis (assessing earnings impact under carbon prices of $50-250 per tonne), technology substitution curves (modeling the pace at which low-carbon alternatives displace fossil fuel assets), and stranded asset analysis (estimating write-down exposure for fossil fuel reserves, carbon-intensive real estate, and high-emissions industrial capacity).

Scenario Pathways are internally consistent narratives describing possible climate futures, typically anchored to the NGFS reference scenarios: Orderly (Net Zero 2050, Below 2C), Disorderly (Delayed Transition, Divergent Net Zero), and Hot House World (Current Policies, Nationally Determined Contributions). Each pathway specifies assumptions about carbon pricing, technology development, policy stringency, and physical climate outcomes that drive model inputs.

Climate Value-at-Risk (CVaR) applies financial risk measurement frameworks to climate-adjusted asset valuations, estimating the potential loss in portfolio value attributable to climate-related factors over specified time horizons. MSCI's CVaR model covers over 10,000 corporate issuers and disaggregates transition costs (policy, technology, and market risks) from physical risks (extreme weather and chronic changes) to enable targeted portfolio construction.

Climate Risk Analytics KPIs by Sector

Financial Services

KPI	Below Average	Average	Above Average	Leading
Portfolio Coverage (% AUM with climate risk scores)	<30%	30-60%	60-85%	>85%
Scenario Pathways Analyzed	1-2	3-4	5-6	>6 (incl. custom)
Physical Risk Asset Resolution	Country	Region/State	City/Postcode	Property-level
Decision Integration Rate (% decisions using climate data)	<10%	10-25%	25-50%	>50%
Time to Generate Scenario Report	>4 weeks	2-4 weeks	1-2 weeks	<1 week
Carbon Price Sensitivity Range Tested	Single price	2-3 prices	4-6 prices	Full curve
Data Refresh Frequency	Annual	Semi-annual	Quarterly	Monthly or better

Financial institutions leading in climate risk analytics, including ING, BNP Paribas, and Aviva, have achieved portfolio coverage exceeding 85% and integrate climate scenarios into credit committee processes for exposures above defined thresholds. ING's Terra approach aligns its lending portfolio to Paris Agreement pathways sector by sector, using technology-specific decarbonization curves to assess transition alignment. BNP Paribas reports that climate scenario analysis influenced over EUR 3 billion in credit decisions during 2024, including reduced exposure to thermal coal value chains and increased allocation to renewable energy infrastructure.

Real Estate and Infrastructure

KPI	Below Average	Average	Above Average	Leading
Asset-Level Physical Risk Assessment Coverage	<20%	20-50%	50-80%	>80%
Hazard Types Modeled	1-2	3-4	5-7	>7 (compound events)
Time Horizons Assessed	Single (2030 or 2050)	2 horizons	3-4 horizons	Decadal (2030-2100)
Financial Loss Quantification	Qualitative only	Order of magnitude	Asset-level estimates	Probabilistic distributions
Stranding Risk Assessment	None	Sector-level	Portfolio screening	Asset-specific modeling
Retrofit Prioritization Using Climate Data	None	Ad hoc	Systematic scoring	Optimized capital planning
Insurance Cost Projection Integration	None	Awareness	Scenario-based	Dynamic repricing models

Real estate investors and infrastructure owners face among the most direct physical risk exposures. Nuveen Real Estate applies property-level climate risk scoring across its $154 billion global portfolio, using Four Twenty Seven (now Moody's) data to assess flood, heat stress, hurricane, wildfire, and sea-level rise exposure for every asset. Properties scoring in the top risk decile undergo mandatory engineering assessments and adaptation planning. The Abu Dhabi Investment Authority (ADIA) integrates climate risk analytics into infrastructure due diligence, requiring climate scenario analysis for all acquisitions with expected hold periods exceeding 15 years.

Energy and Utilities

KPI	Below Average	Average	Above Average	Leading
Generation Portfolio Climate Stress Testing	Qualitative	Single scenario	Multi-scenario	Stochastic modeling
Transition Risk: Stranded Asset Exposure Quantified	None	Aggregate estimate	Asset-level	NPV under multiple pathways
Grid Resilience Modeling Coverage	None	Transmission only	Transmission + distribution	Full system with DER
Demand Scenario Horizon	<5 years	5-10 years	10-20 years	20-40 years
Renewable Integration Risk Assessment	Static capacity	Annual planning	Hourly modeling	Probabilistic weather-adjusted
Regulatory Scenario Tracking	Reactive	Annual review	Quarterly updates	Continuous monitoring

Enel has deployed climate scenario analysis across its global portfolio spanning 89 GW of capacity in 28 countries. The company models physical risks to generation assets (drought impacts on hydropower, temperature effects on thermal efficiency, extreme weather damage) alongside transition risks (carbon pricing trajectories, renewable cost curves, demand electrification pathways). National Grid ESO uses probabilistic climate-adjusted demand forecasting that integrates temperature projections under multiple NGFS scenarios to plan transmission investments with 30-40 year operational lifetimes.

Agriculture and Food

KPI	Below Average	Average	Above Average	Leading
Supply Chain Physical Risk Mapping	None	Country-level	Region/watershed	Farm-level
Crop Yield Scenario Modeling	None	Single crop/region	Multi-crop portfolio	Dynamic adaptation modeling
Water Stress Projection Integration	Awareness	Basin-level screening	Supply-specific analysis	Supplier water budgets
Commodity Price Volatility Under Climate Scenarios	None	Qualitative	Directional estimates	Quantified ranges
Adaptation Investment Planning Horizon	<3 years	3-5 years	5-10 years	>10 years

Nestle applies climate scenario analysis to its agricultural supply chains across 14 priority commodities, using crop modeling, water stress projections, and extreme weather frequency analysis to identify sourcing regions at highest risk of disruption. The company reported that climate risk analytics informed $1.2 billion in supply chain diversification and adaptation investments during 2023-2025, including relocation of sourcing regions for cocoa, coffee, and dairy.

What's Working

Integration Into Existing Risk Frameworks

Organizations that embed climate scenarios within existing enterprise risk management processes achieve higher decision integration rates than those maintaining climate risk as a standalone function. HSBC integrated climate scenario outputs into its credit risk appetite framework, establishing concentration limits for high-transition-risk sectors that trigger escalation when breached. This approach leverages existing governance structures and decision-making workflows rather than creating parallel processes.

Combining Physical and Transition Risk Analysis

Leading practitioners analyze physical and transition risks jointly rather than in isolation. A disorderly transition scenario, for example, implies both higher physical risks (due to delayed emissions reductions) and sudden transition costs (due to abrupt policy shifts). Zurich Insurance Group models these interactions explicitly, estimating that compound physical-transition scenarios produce losses 20-35% higher than the sum of independent assessments.

Sector-Specific Calibration

Generic climate risk scores provide limited decision value. Organizations achieving the highest impact calibrate scenarios to sector-specific value drivers: water availability for beverage companies, grid reliability for data center operators, transportation disruption for logistics firms. Unilever's climate risk analytics are tailored to the agricultural commodities, water basins, and distribution networks most material to its operations, enabling targeted adaptation spending rather than broad-based risk acknowledgment.

What's Not Working

Overreliance on Long-Term Scenarios

Most climate scenario analysis focuses on 2050 or 2100 endpoints, but capital allocation and strategic planning operate on 3-10 year horizons. Organizations struggle to translate multi-decade scenarios into near-term actionable insights. The gap between climate model timeframes and business planning cycles remains the single most cited barrier to decision integration in industry surveys.

Model Uncertainty Communication

Climate models produce wide uncertainty ranges that risk managers find difficult to operationalize. A physical risk estimate might span a factor of three depending on the climate model, emissions pathway, and vulnerability assumptions used. Many organizations default to central estimates that obscure tail risks, while others become paralyzed by the breadth of possible outcomes. Best practice is to present decision-relevant metrics (such as the probability of exceeding specific loss thresholds) rather than point estimates or full probability distributions.

Data Gaps in Emerging Markets

Climate risk analytics tools were developed primarily for OECD markets with comprehensive asset databases, weather station networks, and regulatory frameworks. Coverage of emerging and developing economies, where physical climate risks are often most severe, remains inadequate. Hazard models for Sub-Saharan Africa, Southeast Asia, and Latin America carry significantly wider uncertainty bounds, and asset-level exposure data is frequently incomplete or outdated.

Action Checklist

• Establish baseline metrics by auditing current climate risk analytics capabilities against the sector-specific KPI tables above
• Implement decision integration tracking by recording which lending, investment, underwriting, or capital allocation decisions incorporate climate scenario outputs
• Adopt at least three NGFS reference scenarios plus one custom scenario reflecting organization-specific risk factors
• Require property-level or asset-level physical risk resolution for material exposures, moving beyond country or regional aggregations
• Set quarterly data refresh targets for climate risk scores, replacing annual update cycles that miss emerging risks
• Bridge the planning horizon gap by developing near-term (3-5 year) climate risk indicators alongside long-term (2050) scenario analysis
• Communicate model uncertainty using decision-relevant formats such as exceedance probability curves rather than single point estimates
• Benchmark climate risk analytics spending against peer institutions to ensure investment levels are commensurate with regulatory expectations

FAQ

Q: What is a reasonable budget for implementing climate risk analytics capabilities? A: For mid-sized financial institutions (EUR 50-200 billion in assets), expect EUR 2-5 million for initial platform procurement, data acquisition, and model development, plus EUR 500,000-1.5 million in annual operating costs. Large global institutions spend EUR 10-25 million annually on climate risk analytics, including dedicated teams of 15-40 specialists. Real estate firms typically allocate EUR 0.50-2.00 per asset annually for portfolio-level climate risk screening, with EUR 5,000-25,000 per asset for detailed engineering-grade assessments.

Q: How should organizations select between climate risk analytics vendors? A: Evaluate vendors across five dimensions: hazard model transparency (can you inspect underlying assumptions?), asset resolution (property-level versus postcode), scenario flexibility (can you run custom pathways?), financial loss quantification methodology (probabilistic versus deterministic), and integration with existing risk systems (API availability, data format compatibility). Request validation studies showing model accuracy against historical loss events and conduct parallel assessments using multiple vendors for a subset of the portfolio.

Q: What is the minimum viable climate scenario analysis for CSRD compliance? A: At minimum, companies must disclose the resilience of their strategy under at least two scenarios: one consistent with limiting warming to 1.5C and one reflecting higher warming outcomes (3C or current policies). Disclosures must include the time horizons considered, key assumptions, and the financial effects of identified risks and opportunities. For most companies, this requires physical risk screening of material assets and operations, transition risk assessment of carbon-intensive activities, and qualitative or quantitative financial impact estimation.

Q: How frequently should climate risk analytics be updated? A: Leading practice is quarterly updates for portfolio-level metrics and annual deep-dive analysis for material exposures. Physical risk models should be refreshed when new climate model generations are released (approximately every 5-7 years for CMIP cycles) or when significant methodology improvements become available. Transition risk assumptions (carbon prices, policy changes, technology costs) should be reviewed at least semi-annually given the pace of regulatory development.

Q: Can smaller organizations without dedicated teams implement meaningful climate risk analytics? A: Yes, through tiered approaches. Tier 1 (minimal resources): use free tools such as the NGFS Climate Impact Explorer and the Climate Risk Dashboard from the Global Adaptation Index to screen physical risks at the country and regional level. Tier 2 (moderate resources): engage consultants or specialized platforms (Cervest, Jupiter Intelligence, XDI) for asset-level assessments of the most material exposures. Tier 3 (dedicated capability): build in-house analytical capacity with proprietary models calibrated to specific asset classes and risk factors. Most mid-sized companies should target Tier 2 within 12-18 months.

Sources

• Network for Greening the Financial System. (2025). Climate Scenario Analysis by Financial Institutions: Progress and Challenges. Paris: NGFS Secretariat.
• Cambridge Institute for Sustainability Leadership. (2025). Climate Risk Analytics Maturity and Financial Performance. Cambridge: CISL.
• Munich Re. (2025). Natural Catastrophe Review 2024: Global Insured Loss Trends. Munich: Munich Re Group.
• European Central Bank. (2025). Supervisory Expectations on Climate Risk: Updated Assessment. Frankfurt: ECB Banking Supervision.
• MSCI. (2025). Climate Value-at-Risk Methodology: Technical Documentation, Version 4.0. New York: MSCI Inc.
• Bank of England. (2024). Results of the Climate Biennial Exploratory Scenario: Follow-Up Assessment. London: Bank of England.
• Intergovernmental Panel on Climate Change. (2023). AR6 Synthesis Report: Physical Science Basis for Climate Risk Assessment. Geneva: IPCC.