Introduction: The Emerging Systemic Frontier in Quantitative Risk

For decades, quantitative risk management focused exclusively on financial variables: interest rates, credit defaults, equity price volatility, and operational disruptions. However, in modern global markets, institutional stability is increasingly threatened by an exogenous force that traditional models ignored: Climate Change and Environmental, Social, and Governance (ESG) factors.

Climate risk is divided into two distinct categories: Physical Risk (the economic destruction caused by acute weather events like floods and hurricanes, and chronic shifts like rising sea levels) and Transition Risk (the financial shocks triggered by sudden policy changes, technological shifts, and carbon taxation as the global economy transitions to net-zero). Central banks and international regulators now mandate that commercial banks incorporate climate scenarios into macroprudential stress tests. This lesson deconstructs physical and transition risks, ESG quantitative scoring models, carbon footprint metrics, and climate stress testing frameworks.

Part 1: Physical Risk versus Transition Risk

Quantitative risk desks evaluate climate exposure across two primary dimensions:

1. Physical Risk (Acute and Chronic)

Acute Physical Risks: Event-driven disasters such as hurricanes, wildfires, and catastrophic floods that destroy physical assets, corporate factories, and real estate collateral.

Chronic Physical Risks: Long-term shifts in climate patterns, such as sustained temperature increases, water scarcity, and rising sea levels, which erode agricultural productivity and devalue coastal real estate portfolios.

Quantitative Impact: Modeled by linking geographic asset databases with geospatial climate hazard maps to estimate direct asset write-downs and collateral value drops.

2. Transition Risk (Policy, Technology, and Market Shifts)

Carbon Pricing and Regulation: The implementation of aggressive carbon taxes, emissions caps, and fossil fuel phase-outs that instantly penalize carbon-intensive industries (oil, gas, heavy manufacturing).

Stranded Assets: Investments in coal mines, oil reserves, and internal combustion manufacturing plants that risk becoming economically obsolete overnight, resulting in massive write-offs.

Quantitative Impact: Modeled using forward-looking macroeconomic scenarios (such as Network for Greening the Financial System – NGFS scenarios) that project carbon prices, energy demand shifts, and sectoral credit rating downgrades.

Part 2: ESG Quantitative Scoring and Portfolio Integration

Institutional investors and quantitative asset managers integrate ESG metrics directly into portfolio optimization models.

1. ESG Data Providers and Rating Discrepancies

Quantitative funds ingest raw ESG alternative data from specialized rating agencies (such as MSCI, Sustainalytics, or S&P). However, unlike financial accounting data, ESG ratings suffer from low cross-agency correlation due to subjective scoring methodologies. Quantitative researchers build proprietary data normalization pipelines to clean and reconcile conflicting ESG indicators.

2. ESG-Adjusted Portfolio Optimization

Negative Screening: Explicitly excluding companies involved in controversial sectors (weapons, tobacco, high-carbon fossil fuels).

Best-in-Class Selection: Selecting top-performing ESG companies within every individual industrial sector.

Optimizing the Efficient Frontier: Incorporating ESG scores as constraints or penalty terms directly into Modern Portfolio Theory covariance optimization, balancing risk-adjusted return maximization with carbon footprint reduction targets.

Part 3: Regulatory Climate Stress Testing

Central banks globally (including the Bank of England, the European Central Bank, and the Federal Reserve) conduct exploratory climate stress tests to assess the banking sector’s vulnerability to long-term climate trajectories.

1. The NGFS Scenarios

Regulators utilize standardized multi-decade forward-looking scenarios developed by the Network for Greening the Financial System (NGFS):

• Orderly Transition: Early, smooth policy implementation limiting global warming to 1.5°C with minimal transition friction.

• Disorderly Transition: Delayed, sudden policy implementation resulting in severe carbon price shocks and financial market stress.

• Hot House World: Minimal climate policy action, leading to catastrophic physical risks and severe long-term macroeconomic damage.

2. Integrating Climate Risk into Credit Models (PD/LGD)

Climate stress tests project how corporate borrowers’ revenues and default probabilities (PD) will evolve over 30-year horizons under these scenarios, calculating cumulative expected credit losses across bank loan books.

1. Climate Risk Taxonomy

Physical Risk Components:

Transition Risk Components:

2. Climate Risk Quantification

Physical Risk Modeling:

Physical Risk Assessment:

1. Asset Exposure:
   - Geographic location of assets
   - Climate hazard maps
   - Vulnerability assessment

2. Economic Impact:
   - Direct: Asset damage, business interruption
   - Indirect: Supply chain disruption, insurance costs

3. Valuation Impact:
   ΔV = -Σ(Hazard_i × Vulnerability_i × Exposure_i)

Where:
- Hazard_i = Probability of event i
- Vulnerability_i = Fraction of asset value at risk
- Exposure_i = Asset value exposed

4. Long-term Projections:
   - Climate model projections
   - RCP scenarios (2.6, 4.5, 6.0, 8.5)
   - Confidence intervals

Transition Risk Modeling:

Transition Risk Assessment:

1. Carbon Pricing:
   ΔCost = Emissions × Carbon_Price × (1 + Escalation_Rate)^t

2. Stranded Assets:
   Stranded_Value = Book_Value × Probability_of_Stranding

3. Credit Risk Adjustment:
   PD_Climate = PD_Base × (1 + Climate_Risk_Factor)

   Climate_Risk_Factor = f(Carbon_Intensity, Transition_Readiness, Regulatory_Exposure)

4. Scenario Analysis:
   - NGFS scenarios
   - Company-specific transition plans
   - Sectoral impacts

3. ESG Scoring and Integration

ESG Score Components:

ESG Rating Discrepancies:

Rating Agency Comparison:

| Company | MSCI | Sustainalytics | S&P | Correlation |
|---------|------|----------------|-----|-------------|
| Company A | AAA | 15 (Low Risk) | 85 | Low |
| Company B | BBB | 30 (Medium) | 65 | Low |
| Company C | AA | 20 (Low Risk) | 90 | Low |

Reasons for Discrepancies:
1. Different weighting of factors
2. Different data sources
3. Different interpretation of materiality
4. Methodological differences

ESG Integration into Portfolio Optimization:

ESG-Adjusted Optimization:

Maximize: (1 - λ) × μ_w - λ × Risk_w + η × ESG_w

Subject to:
Σ w_i = 1
w_i ≥ 0 (or with limits)

Where:
- μ_w = Expected portfolio return
- Risk_w = Portfolio risk (variance or VaR)
- ESG_w = Portfolio ESG score
- λ = Risk aversion parameter
- η = ESG preference parameter

Alternative Approach (Constraints):
Minimize: Risk_w
Subject to:
Σ w_i = 1
w_i ≥ 0
ESG_w ≥ ESG_min (minimum ESG score)

4. NGFS Climate Scenarios

NGFS Scenario Parameters:

Macroeconomic Variables by Scenario:

5. Climate Stress Testing

Credit Risk Adjustment:

Climate-Adjusted PD:

PD_t = PD_Base × (1 + α × Transition_Risk_t + β × Physical_Risk_t)

Where:
- Transition_Risk_t = Σ (Exposure_i × Carbon_Intensity_i × Carbon_Price_t)
- Physical_Risk_t = Σ (Exposure_i × Hazard_Probability_i × Vulnerability_i)
- α, β = Calibration parameters

Climate-Adjusted LGD:

LGD_t = LGD_Base × (1 + γ × Stranded_Asset_Risk_t)

Where:
- Stranded_Asset_Risk_t = Fraction of collateral becoming stranded
- γ = Calibration parameter

Portfolio Impact Analysis:

Climate Stress Test Process:

1. Identify exposures:
   - Geographic location
   - Sector concentration
   - Carbon intensity

2. Apply scenario:
   - Select NGFS scenario
   - Project carbon prices
   - Project physical risks

3. Calculate impacts:
   - PD adjustments
   - LGD adjustments
   - Valuation changes

4. Aggregate losses:
   - Expected losses
   - Unexpected losses
   - Capital impact

5. Report results:
   - CET1 impact
   - Risk profile changes
   - Mitigation strategies

6. ESG Regulatory Landscape

Key Regulations: