Slotting Approach Specification¶
Specialised lending slotting categories for project finance, object finance, commodities finance, and IPRE.
Regulatory Reference: CRR Articles 147(8), 153(5)
Test Group: CRR-E
Requirements Status¶
| ID | Requirement | Priority | Status |
|---|---|---|---|
| FR-1.6 | Specialised lending slotting with maturity band risk weights | P0 | Done |
| FR-1.7 | Equity risk weights: SA (Art. 133) and IRB Simple (Art. 155) | P1 | Done |
Overview¶
The slotting approach assigns risk weights based on qualitative category assessment rather than PD/LGD modelling. It applies to specialised lending exposures where banks cannot estimate PD using standard IRB models.
Specialised Lending Types¶
| Type | Abbreviation | Description |
|---|---|---|
| Project Finance | PF | Long-term financing of infrastructure/industrial projects |
| Object Finance | OF | Financing of physical assets (ships, aircraft, etc.) |
| Commodities Finance | CF | Structured financing of commodity inventories |
| Income-Producing RE | IPRE | Commercial real estate where repayment depends on rental income |
| High Volatility CRE | HVCRE | CRE with higher risk characteristics (PRA PS1/26 only — see below) |
CRR Slotting Risk Weights¶
Under UK CRR Art. 153(5), risk weights are assigned from a single table (Table 1) differentiated by remaining maturity. All specialised lending types (PF, OF, CF, IPRE) use the same table.
Table 1 (Art. 153(5))¶
| Category | Remaining Maturity >= 2.5yr | Remaining Maturity < 2.5yr |
|---|---|---|
| Strong | 70% | 50% |
| Good | 90% | 70% |
| Satisfactory | 115% | 115% |
| Weak | 250% | 250% |
| Default | 0% (EL) | 0% (EL) |
(EL) annotation — Default category
The 0% risk weight on the Default row is paired with a non-zero expected loss
rate of 50% of EAD. The capital impact for defaulted slotting exposures flows
through the EL shortfall/excess mechanism (Art. 159), not through the risk weight.
See Slotting Expected Loss Rates — Table B
for the full EL ladder; the same (EL) convention is used in the
Basel 3.1 slotting tables.
UK CRR Art. 158 was omitted by SI 2021/1078 — the 50% Default EL rate is sourced
from the PRA Rulebook (CRR Firms) equivalent, see
Provisions spec for the legal basis.
UK CRR Has No HVCRE Concept — No Table 2
The term "high volatility commercial real estate" does not appear anywhere in the UK onshored CRR (Regulation (EU) No 575/2013 as retained in UK law). Art. 153(5) contains only Table 1 — a single table covering all specialised lending types with no HVCRE distinction. Art. 147(8) defines specialised lending generically (no HVCRE sub-type).
The original EU CRR had a separate Table 2 with elevated HVCRE weights (Strong: 95%/70%, Good: 120%/95%, Satisfactory: 140%/140%, Weak: 250%/250%, Default: 0%/0%). This table was not retained in the UK onshored version. Under UK CRR, exposures that would be classified as HVCRE in the EU use the same Table 1 as all other SL types.
PRA PS1/26 (Basel 3.1, effective 1 Jan 2027) introduces HVCRE as a distinct sub-type in Table A with separate, higher risk weights. See Basel 3.1 section below.
Code divergence (D3.22): The calculator applies EU CRR Table 2 HVCRE weights
(rulepack lookups slotting_rw_hvcre / slotting_rw_hvcre_short, read in
engine/slotting/transforms.py) for CRR exposures with is_hvcre=True. This is more conservative than required by UK law (higher RWs)
but has no UK CRR legal basis. See acceptance tests CRR-E4, CRR-E7, CRR-E8.
Basel 3.1 Slotting Risk Weights¶
Under Basel 3.1, PRA PS1/26 Art. 153(5) Table A uses subgrade columns A/B (Strong) and C/D (Good) for maturity-based differentiation. Art. 153(5)(c) assigns column B (Strong) and column D (Good) as the default. Column A/C (lower weights) may be used in three cases:
- Art. 153(5)(d): remaining maturity < 2.5 years — optional for all SL types
- Art. 153(5)(e): IPRE Strong meeting all four sub-conditions — (i) substantially stronger underwriting, (ii) very low LTV, (iii) investment-grade income stream including tenant income ≥ 100% of the obligor's debt service obligations, (iv) no ADC characteristics
- Art. 153(5)(f): PF Strong with substantially stronger underwriting (single substance test; no quantitative sub-conditions)
Satisfactory/Weak/Default have no maturity split. See Basel 3.1 Slotting spec for full details.
PRA vs BCBS B31 Slotting Structure
BCBS CRE33 removes the maturity distinction and uses flat risk weights for B31 slotting. PRA PS1/26 Art. 153(5) Table A preserves maturity-based subgrade columns from CRR. Column A/C = short maturity (<2.5yr), Column B/D = standard (≥2.5yr). The values below reflect the PRA structure.
Non-HVCRE Operational (OF, CF, IPRE, PF Operational) — Table A¶
| Category | Remaining Maturity >= 2.5yr | Remaining Maturity < 2.5yr |
|---|---|---|
| Strong | 70% | 50% |
| Good | 90% | 70% |
| Satisfactory | 115% | 115% |
| Weak | 250% | 250% |
| Default | 0% (EL) | 0% (EL) |
Project Finance Pre-Operational¶
PRA Has No Separate Pre-Operational Slotting Table
PRA PS1/26 Art. 153(5) Table A does not contain a separate pre-operational PF table. All PF (including pre-operational) uses the standard Non-HVCRE weights above. The pre-operational distinction in PRA only applies under SA (Art. 122B(2)(c): 130%/100%/80%), not under slotting. BCBS CRE33 had separate higher weights (Strong=80%, Good=100%, Satisfactory=120%, Weak=350%) but PRA did not adopt this.
Pre-operational PF under PRA slotting uses the standard table: Strong=70%, Good=90%, Satisfactory=115%, Weak=250%, Default=0%.
HVCRE — Table A¶
HVCRE Introduced by PRA PS1/26
HVCRE is not a continuation from CRR — it is newly introduced by PRA PS1/26 Art. 153(5) Table A. UK CRR has no HVCRE concept (see CRR section above). The elevated weights below have no CRR predecessor in UK law.
| Category | Remaining Maturity >= 2.5yr | Remaining Maturity < 2.5yr |
|---|---|---|
| Strong | 95% | 70% |
| Good | 120% | 95% |
| Satisfactory | 140% | 140% |
| Weak | 250% | 250% |
| Default | 0% (EL) | 0% (EL) |
Slotting Expected Loss Rates — Table B (PRA PS1/26 Art. 158(6))¶
EL rates for slotting exposures are maturity-dependent, unlike the flat values in BCBS CRE33.
Non-HVCRE (OF, PF, CF, IPRE) — Table B¶
| Category | Remaining Maturity < 2.5yr | Remaining Maturity >= 2.5yr |
|---|---|---|
| Strong | 0% | 0.4% |
| Good | 0.4% | 0.8% |
| Satisfactory | 2.8% | 2.8% |
| Weak | 8% | 8% |
| Default | 50% | 50% |
HVCRE — Table B¶
| Category | EL Rate (< 2.5yr, col A) | EL Rate (>= 2.5yr, col B) |
|---|---|---|
| Strong | 0.4% | 0.4% |
| Good | 0.4% | 0.4% |
| Satisfactory | 2.8% | 2.8% |
| Weak | 8% | 8% |
| Default | 50% | 50% |
HVCRE EL Rates
HVCRE Table B EL rates are the same regardless of maturity band for most categories. Strong and Good both use 0.4% in all columns, unlike non-HVCRE where Strong short-maturity has 0% EL and Good short-maturity has 0.4% EL.
Previous Values Were Wrong
The EL rates previously documented here (Strong=5%, Good=10%, Satisfactory=35%, Weak=50%, Default=50%) were BCBS CRE33 values, not PRA PS1/26 values. The PRA Table B values above are dramatically lower for Strong/Good categories (e.g., Strong 0% vs 5%, Good 0.4% vs 10%). Using the BCBS values would massively overstate the EL shortfall for well-categorised slotting exposures.
These EL rates are used when calculating the IRB EL shortfall/excess for slotting exposures. The 0% risk weight for defaulted slotting categories means K=0, but the EL amount = EL_rate × EAD is still recognised for the EL shortfall calculation.
PRA vs BCBS Slotting Differences¶
No Separate PRA Pre-Operational Table (CRR or Basel 3.1)
BCBS CRE33 defines a separate pre-operational PF table, but PRA does not adopt this distinction under either CRR or Basel 3.1. PRA PS1/26 Art. 153(5) Table A uses a single table for all non-HVCRE SL types regardless of operational status. The pre-operational PF differentiation in PRA only applies under SA (Art. 122B(2)(c): 130%/100%/80% by quality).
Large FSE Threshold¶
A large financial sector entity (LFSE) is defined under CRR as having total assets ≥ EUR 70 billion (CRR Art. 142(1)(4)); under Basel 3.1 the threshold is GBP 79 billion (PS1/26 Glossary p. 78, Note "corresponds to Article 142(1)(4) of CRR"). This threshold determines eligibility for the 1.25x correlation multiplier (Art. 153(2)), not the Art. 147A approach restriction (which applies to all FSEs regardless of size). It is distinct from the Art. 147A(1)(e) large corporate revenue threshold (GBP 440m).
See Framework Differences for full Basel 3.1 detail.
Equity¶
SA (CRR Art. 133)¶
Risk weights by equity type (listed, unlisted, strategic holdings).
IRB Simple (CRR Art. 155)¶
- Exchange-traded / listed equity: 290%
- Private equity (diversified): 190%
- All other equity (unlisted, speculative, etc.): 370%
Basel 3.1¶
Removal of equity IRB — Art. 155 is entirely "[Provision left blank]" in PRA PS1/26. All equity falls to SA treatment under Art. 147A(1)(a).
Key Scenarios¶
Long Maturity (≥2.5 years)¶
| Scenario ID | SL Type | Category | Expected RW | Reference |
|---|---|---|---|---|
| CRR-E1 | Project Finance | Strong | 70% | Art. 153(5) Table 1 |
| CRR-E2 | Project Finance | Good | 90% | Art. 153(5) Table 1 |
| CRR-E3 | IPRE | Weak | 250% | Art. 153(5) Table 1 |
| CRR-E4 | HVCRE | Strong | 95% | EU CRR Table 2 (no UK basis — see D3.22) |
Short Maturity (<2.5 years)¶
| Scenario ID | SL Type | Category | Expected RW | Reference |
|---|---|---|---|---|
| CRR-E5 | Project Finance | Strong | 50% | Art. 153(5) Table 1 |
| CRR-E6 | Project Finance | Good | 70% | Art. 153(5) Table 1 |
| CRR-E7 | HVCRE | Strong | 70% | EU CRR Table 2 (no UK basis — see D3.22) |
| CRR-E8 | HVCRE | Good | 95% | EU CRR Table 2 (no UK basis — see D3.22) |
Coverage Notes
Object finance and commodities finance use the same non-HVCRE table as PF — validated through CRR-E1/E2/E5/E6 (same risk weight lookup). Defaulted slotting (0% RW) is validated through the CRR-I group (defaulted exposures). Satisfactory and Weak short-maturity scenarios (115%/250% — no maturity differentiation) are validated implicitly through E3 (same weight regardless of maturity band).
Acceptance Tests¶
| Group | Scenarios | Tests | Pass Rate |
|---|---|---|---|
| CRR-E: Specialised Lending | E1–E8 | 13 | 100% (13/13) |