Skip to content

Basel 3.1 Framework Differences

Key differences from CRR including output floor, PD/LGD floors, and removal of supporting factors.

Regulatory Reference: PRA PS1/26


Overview

Basel 3.1 (effective 1 January 2027 in the UK) introduces significant changes to the credit risk framework. The calculator supports both regimes via a configuration toggle.

Key Differences

Parameter CRR (Current) Basel 3.1 Reference
RWA Scaling Factor 1.06 Removed
SME Supporting Factor 0.7619 / 0.85 Removed CRR Art. 501
Infrastructure Factor 0.75 Removed CRR Art. 501a
Output Floor None 72.5% of SA PRA PS1/26
PD Floor 0.03% (all classes) Differentiated CRE30.55
A-IRB LGD Floors None Yes (by collateral type) CRE30.41
Slotting Risk Weights Maturity-differentiated HVCRE + PF pre-op differentiated PRA PS1/26

Differentiated PD Floors (Basel 3.1)

Exposure Class PD Floor
Corporate 0.05%
Corporate SME 0.05%
Retail Mortgage 0.05%
Retail Other 0.05%
QRRE (Transactors) 0.03%
QRRE (Revolvers) 0.10%

A-IRB LGD Floors (Basel 3.1)

Collateral Type LGD Floor
Unsecured (Senior) 25%
Unsecured (Subordinated) 50%
Financial collateral 0%
Receivables 10%*
Commercial real estate 10%*
Residential real estate 5%*
Other physical 15%*

*Values reflect PRA PS1/26 implementation. BCBS standard values differ (Receivables: 15%, CRE: 10%, RRE: 10%, Other Physical: 20%).

F-IRB Supervisory LGD (CRE32)

Exposure Type CRR Basel 3.1
Corporate/Institution (Senior) 45% 40%
Corporate/Institution (Subordinated) 75% 75%
Secured - Financial Collateral 0% 0%
Secured - Receivables 35% 20%
Secured - CRE/RRE 35% 20%
Secured - Other Physical 40% 25%

Output Floor

The output floor ensures IRB RWA cannot fall below a percentage of what the SA would produce:

RWA_final = max(RWA_IRB, floor_percentage x RWA_SA)

Transitional Schedule

Year Floor Percentage
2027 50.0%
2028 55.0%
2029 60.0%
2030 65.0%
2031 70.0%
2032+ 72.5%

Output Floor Adjustment (OF-ADJ)

The full output floor formula from PRA PS1/26 Art. 92 is:

TREA = max(U-TREA, x × S-TREA + OF-ADJ)

Where:

  • U-TREA = un-floored total risk exposure (using internal models where permitted)
  • S-TREA = standardised total risk exposure (recalculated using SA only)
  • x = floor percentage (see transitional schedule above)
  • OF-ADJ = adjustment for the difference between IRB expected loss provisions treatment and SA general credit risk adjustments

OF-ADJ reconciles the different treatment of provisions: under IRB, expected loss shortfall adds to capital requirements while under SA, general credit risk adjustments reduce risk exposure. Without this adjustment, the floor comparison would not be on a like-for-like basis.

Supervisory Haircut Comparison

CRR Haircuts (3 maturity bands)

Collateral Type 0-1y 1-5y 5y+
Govt bonds CQS 1 0.5% 2% 4%
Govt bonds CQS 2-3 1% 3% 6%
Corp bonds CQS 1 1% 4% 8%
Corp bonds CQS 2-3 2% 6% 12%
Main index equities 15%
Other equities 25%
Gold 15%
Cash 0%

Basel 3.1 Haircuts (5 maturity bands)

Collateral Type 0-1y 1-3y 3-5y 5-10y 10y+
Govt bonds CQS 1 0.5% 2% 2% 4% 4%
Govt bonds CQS 2-3 1% 3% 4% 6% 12%
Corp bonds CQS 1 1% 4% 6% 10% 12%
Corp bonds CQS 2-3 2% 6% 8% 15% 15%
Main index equities 25%
Other equities 35%
Gold 15%
Cash 0%

Currency mismatch haircut remains 8% under both frameworks (CRR Art. 224 / CRE22.54).

Slotting Risk Weights (Basel 3.1)

Basel 3.1 (BCBS CRE33) introduces three distinct slotting weight tables:

Non-HVCRE Operational (OF, CF, IPRE, PF Operational)

Category Risk Weight
Strong 70%
Good 90%
Satisfactory 115%
Weak 250%
Default 0% (EL)

Project Finance Pre-Operational

Category Risk Weight
Strong 80%
Good 100%
Satisfactory 120%
Weak 350%
Default 0% (EL)

HVCRE

Category Risk Weight
Strong 95%
Good 120%
Satisfactory 140%
Weak 250%
Default 0% (EL)

Slotting Subgrades

Basel 3.1 allows residual maturity-based differentiation within the Strong and Good categories using subgrades:

Category Subgrade A (< 2.5yr residual) Subgrade B (≥ 2.5yr residual)
Strong A / Strong B 50% / 70% (PF Operational) 70% / 70%
Good C / Good D 70% / 90% (PF Operational) 90% / 90%

IPRE "Strong A" requires specific criteria: low LTV, adequate tenant income, and no ADC characteristics. These subgrades provide finer risk differentiation within the broader slotting categories.

Compare with CRR slotting weights in the Slotting Approach specification.

Financial Institution Correlation Multiplier (CRE31.5)

The 1.25x correlation multiplier applies to exposures to financial institutions only (not non-financial corporates): - Regulated financial institutions with total assets above the applicable threshold: - CRR: EUR 70bn (Art. 153(2)) - BCBS/Basel 3.1: USD 100bn (CRE31.5) - Unregulated financial institutions regardless of size

This multiplier is already implemented via the requires_fi_scalar flag in the classifier and _polars_correlation_expr() in the IRB formulas. It applies under both CRR and Basel 3.1 frameworks.

Note: There is no separate "large corporate" correlation multiplier for non-financial corporates in either the BCBS standard or PRA PS1/26.

A-IRB CCF Floor (CRE32.27)

A-IRB own-estimate CCFs must be at least 50% of the SA CCF for the same item type.

IRB Maturity Calculation Changes

Basel 3.1 refines the IRB effective maturity (M) calculation (PRA PS1/26 Art. 162):

Aspect CRR Basel 3.1
Cash-flow schedule Weighted average of cash flows Same: M = max(1, min(Σ(t×CF_t)/Σ(CF_t), 5))
Revolving exposures Repayment date of current drawing Maximum contractual termination date
Floor 1 year 1 year
Cap 5 years 5 years

The revolving maturity change is significant — it typically increases M for revolving facilities, leading to higher maturity adjustments and therefore higher capital.

Configuration

Switch between frameworks using the configuration factory:

from rwa_calc.contracts.config import CalculationConfig

# CRR (current)
config = CalculationConfig.crr(reporting_date=date(2026, 12, 31))

# Basel 3.1
config = CalculationConfig.basel_3_1(reporting_date=date(2027, 1, 1))