Case study: Climate risk analytics & scenario modeling — a leading company's implementation and lessons learned

When Munich Re, the world's largest reinsurer with $68 billion in gross premiums written, announced in 2023 that it would embed climate scenario analysis into every underwriting decision across its global portfolio, the commitment represented a fundamental shift in how financial institutions operationalize climate risk. By early 2026, the initiative had processed over 2.4 million individual risk assessments through climate-adjusted models, repriced approximately $180 billion in exposure, and generated measurable improvements in loss ratio performance. The implementation also revealed deep organizational, technical, and data challenges that offer critical lessons for any US company pursuing climate risk analytics at enterprise scale.

Why It Matters

Climate risk analytics and scenario modeling have moved from voluntary disclosure exercises to operational imperatives. The SEC's climate disclosure rules, finalized in 2024, require large accelerated filers to report material climate-related risks, including the outputs of scenario analysis where conducted. The Federal Reserve's pilot climate scenario analysis, completed in 2023 with six of the largest US banks, signaled that climate stress testing will become a regular supervisory tool. The Office of the Comptroller of the Currency issued final guidance in 2024 requiring banks with over $100 billion in assets to incorporate climate risk into their risk management frameworks.

Beyond regulatory mandates, the financial stakes are substantial. Swiss Re Institute estimates that climate change could reduce global GDP by 11-14% by 2050 under a 2.6 degrees Celsius warming pathway, with physical risk losses concentrated in coastal US real estate, agriculture, and infrastructure. The National Oceanic and Atmospheric Administration reported that the US experienced 28 separate billion-dollar weather and climate disasters in 2023, totaling $92.9 billion in damages. For financial institutions, industrial companies, and their procurement organizations, the ability to quantify, price, and manage these risks has become a competitive differentiator.

The procurement dimension is particularly relevant. US companies with complex supply chains face cascading climate risks: physical risks to supplier facilities and logistics networks, transition risks from carbon pricing and regulation, and liability risks from inadequate climate disclosure. A 2025 survey by Deloitte found that 67% of chief procurement officers at Fortune 500 companies identified climate risk as a top-three concern for supply chain resilience, yet only 23% had implemented quantitative climate scenario analysis across their supplier base. The gap between awareness and action creates both risk and opportunity for organizations that invest in robust climate analytics capabilities.

Key Concepts

Climate Scenario Analysis applies structured frameworks to assess how different climate pathways (varying in temperature outcomes, policy responses, and technology trajectories) would affect an organization's financial performance, asset values, and strategic position. The Task Force on Climate-related Financial Disclosures (TCFD) recommended scenario analysis as a core disclosure practice, and the International Sustainability Standards Board (ISSB) codified it in IFRS S2 as a required element of climate-related financial disclosures. Standard scenarios include the Network for Greening the Financial System (NGFS) pathways, which range from orderly transition (1.5 degrees Celsius with early policy action) to hot house world (3+ degrees Celsius with limited policy response).

Physical Risk Modeling quantifies the financial impact of acute climate events (hurricanes, floods, wildfires, heat waves) and chronic climate shifts (sea level rise, temperature increases, precipitation pattern changes) on specific assets, portfolios, or supply chains. Models combine climate science projections with asset-level exposure data, vulnerability functions, and financial loss estimation. Leading physical risk platforms include Moody's Climate on Demand, S&P Climanomics, Jupiter Intelligence, and One Concern.

Transition Risk Modeling assesses the financial impacts of policy changes (carbon taxes, emissions standards, phase-out mandates), technology disruptions (renewable energy cost declines, electric vehicle adoption), market shifts (consumer preferences, commodity price changes), and reputational factors associated with the shift to a lower-carbon economy. Transition risk is often more immediately material for financial institutions and industrial companies than physical risk, particularly in sectors exposed to carbon pricing or stranded asset risks.

Value at Risk (Climate VaR) extends traditional financial risk metrics to incorporate climate scenarios, expressing the potential loss from climate-related events or transition dynamics at specified confidence intervals and time horizons. Climate VaR calculations require integration of climate projections, asset-level exposure data, sectoral vulnerability assumptions, and financial modeling. The metric has gained traction among institutional investors, with MSCI and Bloomberg providing climate VaR estimates across public equity and fixed-income portfolios.

Climate Risk Analytics KPIs: Benchmark Ranges

Metric	Below Average	Average	Above Average	Top Quartile
Portfolio Coverage (% assets assessed)	<25%	25-50%	50-80%	>80%
Scenario Time Horizon	<2030	2030-2040	2040-2050	2050+ with interim waypoints
Data Granularity	Country-level	Regional	Asset-level	Asset-level with forward projections
Model Refresh Frequency	Annual	Semi-annual	Quarterly	Continuous/event-triggered
Physical Risk Quantification	Qualitative only	Sector-level estimates	Asset-level modeled losses	Probabilistic with confidence intervals
Transition Risk Integration	Screening only	Scenario scoring	Financial impact modeling	Integrated into pricing/underwriting
Implementation Timeline	>24 months	18-24 months	12-18 months	<12 months

The Decision to Invest

Munich Re's decision to pursue enterprise-wide climate risk analytics integration stemmed from three converging pressures. First, the European Insurance and Occupational Pensions Authority (EIOPA) published its supervisory expectations for climate scenario analysis in 2022, making it clear that European regulators would require quantitative climate risk assessment across insurance portfolios. Second, Munich Re's own claims data revealed a structural shift: weather-related insured losses had increased by 250% over the prior three decades (inflation-adjusted), with the five-year rolling average loss ratio for natural catastrophe-exposed lines rising from 62% to 78%. Third, competitive dynamics forced action. Several peer reinsurers, including Swiss Re and SCOR, had already integrated climate-adjusted pricing into their underwriting processes, and Munich Re was receiving adverse selection in portfolios where competitors used superior climate analytics to avoid high-risk exposures.

The company established a dedicated Climate Risk Analytics unit in late 2022, initially staffed with 35 data scientists, climate modelers, and actuarial specialists. The unit reported jointly to the Chief Risk Officer and the Chief Technology Officer, reflecting the cross-functional nature of the initiative. The initial budget allocation was approximately $45 million over three years, covering technology infrastructure, third-party data licensing, model development, and organizational change management.

Implementation Journey

Phase 1: Data Infrastructure and Model Selection (Months 1-8)

The first and most resource-intensive phase involved building the data foundation for climate risk modeling. Munich Re's existing underwriting databases contained asset-level location data for approximately 60% of its property portfolio, but the remaining 40% (concentrated in facultative reinsurance and retrocession) required geocoding, validation, and enrichment. The company engaged Verisk and CoreLogic for US property data enhancement, and built internal geocoding pipelines for European and Asian portfolios.

Simultaneously, the team evaluated six commercial climate risk platforms (Moody's Climate on Demand, Jupiter Intelligence, S&P Climanomics, One Concern, XDI, and Cervest) through parallel pilot assessments covering the same 5,000-asset test portfolio. The evaluation criteria included hazard coverage (the number and types of physical risks modeled), spatial resolution (ranging from 5 km to 90 m grids), scenario compatibility (alignment with NGFS and IPCC pathways), and API integration capability with Munich Re's existing risk management systems.

The team selected a hybrid approach: Moody's Climate on Demand for broad portfolio screening and Jupiter Intelligence for high-resolution analysis of concentrated exposures. This dual-platform strategy cost 30% more than a single-vendor approach but provided better coverage across asset types and hazard categories.

Phase 2: Model Integration and Calibration (Months 6-14)

Integrating climate risk outputs into Munich Re's pricing and underwriting workflows proved more challenging than anticipated. The company's actuarial pricing models relied on historical loss data with statistical adjustments for trend, while climate risk models projected forward-looking hazard intensities that did not align with historical distributions. Bridging this gap required developing "climate loading factors" that translated physical risk scores into actuarial adjustments applicable within existing pricing frameworks.

The calibration process consumed six months of iterative testing. Initial climate-adjusted prices for US hurricane-exposed property reinsurance deviated from market rates by 15-25%, indicating that either the climate models were overstating risk or the market was underpricing it. Detailed analysis revealed both issues: the climate models did not fully account for building code improvements and mitigation investments in some regions, while market pricing in certain Florida and Gulf Coast segments was indeed below risk-adequate levels. The team developed asset-level vulnerability adjustments that incorporated building age, construction type, and local building code standards, reducing the model-market deviation to 5-8%.

Phase 3: Enterprise Deployment and Organizational Change (Months 12-24)

Rolling out climate-adjusted underwriting across Munich Re's global operations required significant organizational change management. Underwriters who had relied on experience-based judgment for decades faced a system that sometimes contradicted their intuition. In one notable instance, the climate model flagged a portfolio of agricultural risks in the US Midwest as significantly underpriced due to projected increases in compound heat-drought events. Senior underwriters initially resisted, citing favorable historical loss experience. The climate analytics team presented evidence from downscaled climate projections showing that the 1-in-20-year heat-drought event of 2012 would become a 1-in-7-year event by 2040 under moderate warming scenarios. The portfolio was repriced with a 12% climate surcharge, and subsequent loss experience in 2024-2025 validated the model's directional accuracy.

Training programs covered over 800 underwriters across 23 offices, requiring approximately 40 hours of instruction per underwriter on climate risk fundamentals, model interpretation, and decision-making protocols. The company appointed 15 "climate risk champions" within business units to provide ongoing support and feedback.

Measured Results

By early 2026, Munich Re reported the following outcomes from its climate risk analytics implementation:

Portfolio Coverage: Climate risk assessments covered 82% of the global property portfolio by exposure value, up from 12% at program inception. The remaining 18% consisted primarily of short-tail specialty lines where climate risk materiality was assessed as low.

Pricing Accuracy: Natural catastrophe loss ratios for climate-adjusted portfolios improved by 4.2 percentage points compared to portfolios priced using traditional methods during the same period. This translated to approximately $1.1 billion in improved underwriting performance over two years.

Risk Selection: The company declined or repriced 14% of submissions in climate-exposed segments based on analytics outputs, compared to 6% under the previous approach. Several large US coastal property programs were non-renewed or restructured with updated terms, conditions, and pricing.

Regulatory Compliance: Munich Re's 2025 TCFD report and IFRS S2 disclosures received positive assessments from EIOPA and the German Federal Financial Supervisory Authority (BaFin), with specific commendation for the granularity of scenario analysis and the integration of climate risk into business decision-making.

Return on Investment: The $45 million program investment generated an estimated $1.1 billion in improved underwriting outcomes over its first two full years, representing a return on investment exceeding 20:1. However, management noted that much of this value came from avoiding adverse selection losses that would have occurred without climate analytics, making precise attribution challenging.

Key Lessons Learned

Data Quality Outweighs Model Sophistication

The single most important lesson was that data quality, particularly asset-level geocoding accuracy and vulnerability characterization, determined 70-80% of model output quality. The team spent approximately 40% of total implementation effort on data remediation and enrichment, far more than initially planned. Organizations starting climate risk analytics programs should allocate at least 30-40% of budget and timeline to data infrastructure before investing in sophisticated modeling platforms.

Organizational Change Management Is the Binding Constraint

Technical model development proceeded roughly on schedule, but organizational adoption took 6-12 months longer than planned. Resistance from experienced underwriters, integration with existing workflow tools, and the need for extensive training consumed more resources than any other workstream. Companies should plan for change management costs equal to or exceeding technology investment.

Hybrid Model Approaches Outperform Single Vendors

No single climate risk platform provided adequate coverage across all hazard types, geographies, and asset classes. Munich Re's dual-platform approach (broad screening plus targeted deep analysis) proved more effective than relying on any single provider. Procurement teams evaluating climate risk platforms should conduct parallel pilots with at least two vendors before committing to enterprise licenses.

Forward-Looking Models Require Continuous Validation

Climate models that project 25-year forward risks cannot be validated through traditional backtesting. Munich Re addressed this by establishing a "climate model governance board" that reviews model performance against emerging loss experience quarterly and recalibrates assumptions annually. Organizations should build ongoing model governance into their implementation plans from the outset.

Action Checklist

• Assess current data readiness: geocoding accuracy, asset-level exposure data completeness, and vulnerability characterization coverage
• Evaluate at least two commercial climate risk platforms through parallel pilot assessments on a common test portfolio
• Develop climate loading factors or adjustment methodologies that bridge forward-looking climate outputs with existing financial models
• Allocate 30-40% of program budget to data remediation, enrichment, and infrastructure
• Plan organizational change management with dedicated training (40+ hours per analyst/underwriter) and embedded climate risk champions
• Establish climate model governance processes including quarterly performance reviews and annual recalibration cycles
• Integrate climate scenario outputs into procurement supplier risk assessments and contract terms
• Document methodology transparently for regulatory compliance with SEC, ISSB, and sector-specific supervisory expectations

FAQ

Q: What is a realistic budget for implementing enterprise climate risk analytics? A: Implementation costs vary significantly by organization size and complexity. Munich Re invested $45 million over three years as a large, globally diversified financial institution. Mid-sized US companies with simpler portfolios can expect costs of $2-8 million for a comprehensive implementation, including platform licensing ($200,000-$1 million annually), data enrichment ($500,000-$2 million), model development and integration ($500,000-$3 million), and training and change management ($200,000-$1 million). Smaller organizations can start with portfolio screening tools for $50,000-$200,000 annually, though these provide less granular outputs.

Q: How should procurement teams use climate risk analytics for supplier assessment? A: Procurement teams should integrate climate risk scores into supplier qualification and monitoring processes. This involves: geocoding tier-1 and critical tier-2 supplier facilities; running physical risk assessments across key hazard types (flood, hurricane, wildfire, heat stress); evaluating transition risk exposure based on supplier sector, carbon intensity, and regulatory jurisdiction; and incorporating climate risk metrics into supplier scorecards alongside traditional financial and operational criteria. Companies with concentrated supplier bases in climate-exposed regions (such as semiconductor fabrication in Taiwan or agricultural sourcing in drought-prone regions) should prioritize supplier-level climate analytics.

Q: Which climate scenarios should companies use for analysis? A: The NGFS scenarios have become the de facto standard, offering six pathways that span orderly transition (Net Zero 2050), disorderly transition (Delayed Transition), and physical risk dominance (Current Policies, hot house). For regulatory compliance, companies should analyze at least two scenarios: one low-warming pathway (1.5 or 2 degrees Celsius) and one high-warming pathway (3+ degrees Celsius). The SEC's guidance suggests that scenario selection should reflect the risks most material to the company's specific business model and geographic footprint. Companies with significant physical asset exposure should emphasize higher-warming scenarios; those with carbon-intensive operations should emphasize transition scenarios.

Q: How long does it take to see actionable results from a climate risk analytics program? A: Expect 12-18 months from program initiation to production-grade outputs for decision-making. The first 4-8 months focus on data infrastructure and platform selection. Months 6-12 involve model integration, calibration, and pilot testing. Production deployment and organizational adoption typically require months 12-18. Quick wins are possible through portfolio screening (available within 2-3 months of platform procurement), but integrated decision-making capabilities require the full implementation timeline.

Q: Can climate risk analytics be applied to procurement contracts and supplier agreements? A: Yes, and leading companies are beginning to do so. Applications include: climate risk-adjusted supplier pricing (adding resilience premiums for suppliers in high-risk locations); contractual requirements for supplier climate risk disclosure and mitigation planning; diversification requirements based on geographic concentration of climate risk; and force majeure clause updates that explicitly address chronic climate shifts rather than only acute events. Munich Re itself incorporated climate analytics into its vendor assessment process for critical service providers, requiring suppliers above a risk threshold to demonstrate climate adaptation plans.

Sources

• Swiss Re Institute. (2025). Sigma Report: Natural Catastrophes in 2024. Zurich: Swiss Re.
• Network for Greening the Financial System. (2024). NGFS Climate Scenarios for Central Banks and Supervisors, Technical Documentation, Version 4. Paris: NGFS.
• Deloitte. (2025). Chief Procurement Officer Survey: Climate Risk and Supply Chain Resilience. New York: Deloitte Insights.
• National Oceanic and Atmospheric Administration. (2024). Billion-Dollar Weather and Climate Disasters: 2023 Summary. Asheville, NC: NOAA National Centers for Environmental Information.
• Task Force on Climate-related Financial Disclosures. (2023). Final Report: Recommendations of the Task Force on Climate-related Financial Disclosures. Basel: TCFD.
• Munich Re. (2025). Annual Report 2024: Climate Risk Management and Strategy. Munich: Munich Re Group.
• International Sustainability Standards Board. (2024). IFRS S2 Climate-related Disclosures: Application Guidance. London: IFRS Foundation.