Key Concepts¶

This page introduces the fundamental concepts used throughout the RWA calculator. Understanding these terms is essential for working with the system effectively.

Risk-Weighted Assets (RWA)¶

Risk-Weighted Assets (RWA) are a measure of a bank's assets, adjusted for risk. They are used to determine the minimum capital a bank must hold to remain solvent.

Capital Ratio = Regulatory Capital / RWA

The higher the risk of an asset, the higher its risk weight, and the more capital required.

Example¶

Regulatory Frameworks¶

CRR (Capital Requirements Regulation)¶

The CRR implements Basel 3.0 standards in the UK. It is the current regulatory framework, effective until 31 December 2026.

Key features: - 1.06 scaling factor for IRB approaches - SME Supporting Factor (Article 501) - Infrastructure Supporting Factor - No output floor

Basel 3.1¶

Basel 3.1 (implemented via PRA PS1/26) introduces enhanced risk sensitivity and removes certain capital relief mechanisms. Effective from 1 January 2027.

Key features: - Removal of 1.06 scaling factor - Removal of supporting factors - Introduction of 72.5% output floor - Differentiated PD floors - A-IRB LGD floors

Calculation Approaches¶

Standardised Approach (SA)¶

The Standardised Approach uses regulatory-prescribed risk weights based on: - External credit ratings (where available) - Exposure class characteristics - Collateral type

RWA = EAD × Risk Weight

Advantages: Simple, transparent, consistent across institutions Disadvantages: Less risk-sensitive than IRB approaches

Foundation IRB (F-IRB)¶

The Foundation Internal Ratings-Based approach allows banks to estimate: - PD (Probability of Default) - Bank's own estimate - LGD (Loss Given Default) - Supervisory values - EAD (Exposure at Default) - Supervisory rules

RWA = K × 12.5 × EAD × Maturity Adjustment × [1.06 if CRR]

Where K is calculated using the IRB formula.

Advanced IRB (A-IRB)¶

The Advanced Internal Ratings-Based approach allows banks to estimate: - PD - Bank's own estimate - LGD - Bank's own estimate - EAD - Bank's own estimate

This provides the most risk-sensitive calculation but requires regulatory approval.

Slotting Approach¶

The Slotting Approach applies to Specialised Lending exposures where the bank cannot estimate PD: - Project Finance - Object Finance - Commodities Finance - Income-Producing Real Estate (IPRE) - High Volatility Commercial Real Estate (HVCRE)

Exposures are mapped to categories (Strong, Good, Satisfactory, Weak, Default) with prescribed risk weights.

Exposure Classes¶

Exposures are classified into regulatory categories based on the counterparty's entity type:

Entity Type Classification¶

The counterparty's entity_type field is the single source of truth for exposure class determination. The calculator supports 17 entity types that map to both SA and IRB exposure classes.

For example: - pse_sovereign maps to PSE for SA (uses PSE risk weight table) but CENTRAL_GOVT_CENTRAL_BANK for IRB (uses central govt/central bank formula) - mdb maps to MDB for SA (typically 0% RW) but CENTRAL_GOVT_CENTRAL_BANK for IRB

See Classification for the complete entity type mapping and classification algorithm.

Key Metrics¶

EAD (Exposure at Default)¶

The Exposure at Default is the expected exposure amount if the counterparty defaults.

For on-balance sheet items:

EAD = Drawn Amount

For off-balance sheet items:

EAD = Committed Amount × CCF

Where CCF (Credit Conversion Factor) converts undrawn commitments to on-balance sheet equivalents.

PD (Probability of Default)¶

The Probability of Default is the likelihood that a counterparty will default within one year. Used in IRB approaches.

• Range: 0.03% to 100%
• Floor depends on framework and exposure class

LGD (Loss Given Default)¶

The Loss Given Default is the percentage of exposure lost if default occurs, after recoveries.

CCF (Credit Conversion Factor)¶

The Credit Conversion Factor converts off-balance sheet exposures to on-balance sheet equivalents.

Credit Risk Mitigation (CRM)¶

CRM techniques reduce the capital required for an exposure:

Collateral¶

Physical or financial assets securing an exposure. Subject to haircuts:

Plus 8% currency mismatch haircut if applicable.

Guarantees¶

Credit protection provided by a third party. The exposure can be treated as if it were to the guarantor (substitution approach).

Provisions¶

Specific provisions reduce EAD for SA exposures or expected loss for IRB exposures.

Data Hierarchy¶

Exposures follow a hierarchical structure:

graph TD
    A[Counterparty] --> B[Facility]
    B --> C[Loan/Draw]
    A --> D[Collateral]
    A --> E[Guarantee]
    A --> F[Rating]
    B --> G[Collateral]
    C --> H[Collateral]

• Counterparty: The obligor (borrower)
• Facility: A credit arrangement (e.g., credit line)
• Loan: Individual draws or tranches

Ratings and collateral can be assigned at any level and inherit down the hierarchy.

Pipeline Stages¶

The calculation flows through distinct stages:

1. Loader: Reads raw data from files
2. Hierarchy Resolver: Resolves parent-child relationships and inherits attributes
3. Classifier: Assigns exposure classes and calculation approaches
4. CRM Processor: Applies collateral, guarantees, and provisions
5. Calculators: Compute RWA using SA, IRB, Slotting, or Equity
6. Aggregator: Combines results and applies floors/factors

Next Steps¶

• Regulatory Frameworks - Deep dive into CRR and Basel 3.1
• Calculation Methodology - Detailed calculation explanations
• Data Model - Input and output data structures