Skip to content

Slotting Approach Specification

Basel 3.1 revised risk weights for specialised lending slotting categories, with subgrade maturity differentiation and no separate pre-operational project finance distinction.

Regulatory Reference: PRA PS1/26 Art. 147(8), Art. 153(5), CRE33 Test Group: B31-E

Requirements Status

ID Requirement Priority Status
FR-5.1 Revised slotting risk weight tables (subgrade maturity differentiation) P0 Done
FR-5.2 No separate pre-operational PF table (PRA deviation from BCBS) P0 Done
FR-5.3 HVCRE elevated risk weights P0 Done
FR-5.4 Default category EL treatment (0% RW) P0 Done

Overview

Basel 3.1 revises the specialised lending slotting approach risk weights. The key changes are:

  1. Maturity differentiation preserved via subgrades — PRA PS1/26 Art. 153(5) Table A uses A/B columns (Strong) and C/D columns (Good) to preserve maturity-based differentiation. Column A/C (short maturity < 2.5yr) may be used optionally; column B/D (standard, >= 2.5yr) is the default assignment per Art. 153(5)(c).
  2. No pre-operational PF distinction — PRA does not adopt the BCBS separate table for pre-operational project finance (CRE33.6 Table 6); all PF uses the standard non-HVCRE table
  3. HVCRE introduced — elevated weights for high-volatility commercial real estate

Correction: PRA Retains Maturity Differentiation

This section previously stated "maturity split removed" and "flat weights regardless of maturity". This was wrong — it described the BCBS CRE33 approach, not the PRA PS1/26 approach. PRA PS1/26 Art. 153(5)(c)-(d) preserves maturity-based differentiation through the subgrade column structure (A/B for Strong, C/D for Good). See Subgrade Treatment for full details.

Key Changes from CRR

Category CRR (< 2.5yr / ≥ 2.5yr) Basel 3.1 (B/D default) Basel 3.1 (A/C short maturity) Change
Strong 50% / 70% 70% 50% Same structure, subgrade columns
Good 70% / 90% 90% 70% Same structure, subgrade columns
Satisfactory 115% / 115% 115% 115% Unchanged
Weak 250% / 250% 250% 250% Unchanged
Default 0% (EL) / 0% (EL) 0% (EL) 0% (EL) Unchanged

PRA Deviation from BCBS — No Pre-Operational PF Table

BCBS CRE33.6 Table 6 defines separate elevated weights for pre-operational project finance (80%/100%/120%/350%). The PRA does not adopt this distinction — all project finance uses the standard non-HVCRE table regardless of operational status. This is confirmed by the absence of a separate pre-operational table in PRA PS1/26.

Risk Weight Tables

Non-HVCRE Specialised Lending (PF, IPRE, OF, CF)

Art. 153(5), Table A

Applies to: Project Finance, Income-Producing Real Estate, Object Finance, Commodities Finance. Default assignment uses column B (Strong) / D (Good) per Art. 153(5)(c). Short maturity (< 2.5yr) may use column A (Strong) / C (Good) per Art. 153(5)(d).

Slotting Category RW (>= 2.5yr, col B/D) RW (< 2.5yr, col A/C) EL (>= 2.5yr, col B) EL (< 2.5yr, col A)
Strong 70% 50% 0.4% 0%
Good 90% 70% 0.8% 0.4%
Satisfactory 115% 115% 2.8% 2.8%
Weak 250% 250% 8.0% 8.0%
Default 0% 0% 50.0% 50.0%

HVCRE Specialised Lending

Art. 153(5), HVCRE Table

HVCRE Introduced by PRA PS1/26

HVCRE is newly introduced by PRA PS1/26 Art. 153(5) Table A. The UK onshored CRR has no HVCRE concept — Art. 153(5) contains only Table 1 (a single table for all SL types). The original EU CRR had a separate Table 2, but it was not retained in UK onshoring. See CRR Slotting spec for details.

High-Volatility Commercial Real Estate receives elevated weights to reflect the higher risk profile. Default assignment uses column B (Strong) / D (Good) per Art. 153(5)(c). Short maturity (< 2.5yr) may use column A (Strong) / C (Good) per Art. 153(5)(d).

Slotting Category RW col A (Strong < 2.5yr) RW col B (Strong default) RW col C (Good < 2.5yr) RW col D (Good default) EL col A EL col B EL col C EL col D
Strong 70% 95% — — 0.4% 0.4% — —
Good — — 95% 120% — — 0.4% 0.4%
Satisfactory 140% 140% 140% 140% 2.8% 2.8% 2.8% 2.8%
Weak 250% 250% 250% 250% 8.0% 8.0% 8.0% 8.0%
Default 0% 0% 0% 0% 50.0% 50.0% 50.0% 50.0%

Read together with Table A (RW) and Table B (EL) verbatim from PS1/26 Appendix 1 pp. 103 and 108: HVCRE risk weights split A/B for Strong (70%/95%) and C/D for Good (95%/120%), but HVCRE expected loss is flat at 0.4% across all four subgrade columns (A=B=C=D=0.4%). The subgrade structure is therefore present in RW but collapses on the EL side for HVCRE only — non-HVCRE retains the EL split (col A=0%/0.4%, col B=0.4%, col C=0.4%, col D=0.8%).

HVCRE Short-Maturity Subgrades — Implemented (Art. 153(5)(d), P1.117 closed)

HVCRE Table A has distinct short-maturity values in columns A and C (Strong A = 70%, Good C = 95%) — see the same rows above. The calculator now honours these: the b31 pack supplies a distinct slotting_rw_hvcre_short entry, and engine/slotting/transforms.py::lookup_rw fires the weights["hvcre_short"] branch when is_short is set for an HVCRE exposure (with is_short_maturity derived from remaining_maturity < 2.5). Short-maturity HVCRE Strong exposures therefore receive 70% (col A) and Good exposures 95% (col C). See Subgrade Treatment below for the consolidated column A/C behaviour covering both HVCRE and non-HVCRE.

Art. 158(6) Table B — verbatim (PS1/26 Appendix 1 p. 108)

Rating grades: Strong (A | B), Good (C | D), Satisfactory, Weak, Default

Exposure A B C D Satisfactory Weak Default
Object finance 0% 0.4% 0.4% 0.8% 2.8% 8% 50%
Project finance 0% 0.4% 0.4% 0.8% 2.8% 8% 50%
Commodities finance 0% 0.4% 0.4% 0.8% 2.8% 8% 50%
IPRE 0% 0.4% 0.4% 0.8% 2.8% 8% 50%
HVCRE 0.4% 0.4% 0.4% 0.4% 2.8% 8% 50%

HVCRE EL — No Maturity Split AND No Strong/Good Differentiation

Unlike non-HVCRE, the HVCRE EL row in PRA PS1/26 Art. 158(6) Table B is flat (no < 2.5yr / >= 2.5yr distinction) and Strong and Good both carry the same 0.4% rate — see the verbatim quote above. The HVCRE row reads 0.4% / 0.4% / 0.4% / 0.4% / 2.8% / 8% / 50% across columns A | B | C | D | Satisfactory | Weak | Default. The subgrade differentiation collapses on the EL side even though risk weights (Table A) still carry distinct subgrade values (col A = 70%, col B = 95%, col C = 95%, col D = 120%).

Resolved 2026-04-18 (P1.150)

Prior versions of this table and the underlying constants used HVCRE Good = 0.8%, mirroring non-HVCRE long-maturity. That was wrong; PRA PS1/26 Table B HVCRE row is flat at 0.4% across both Strong and Good. The EL shortfall for HVCRE Good exposures was overstated by a factor of two before this fix.

HVCRE vs Non-HVCRE

HVCRE is distinguished from standard CRE by the volatility of the underlying property cash flows and the speculative nature of the development. The classification is determined during the hierarchy/classification stage based on the specialised_lending_type input field.

Subgrade Treatment (Table A Columns A/B/C/D)

PRA PS1/26 Art. 153(5) Table A uses subgrade columns for Strong (columns A and B) and Good (columns C and D). Satisfactory, Weak, and Default have single columns. The full Table A structure:

Exposure Type Strong A Strong B Good C Good D Satisfactory Weak Default
OF, CF, PF, IPRE 50% 70% 70% 90% 115% 250% 0%
HVCRE 70% 95% 95% 120% 140% 250% 0%

Column assignment rules (Art. 153(5)(c)–(f)):

  • (c) Default: Strong → column B; Good → column D; Satisfactory and Weak use their single columns. These are the risk weights in the non-HVCRE and HVCRE tables above.
  • (d) Short maturity: If remaining maturity < 2.5 years, firms may assign column A (Strong) or C (Good) instead — providing lower risk weights. Available for all specialised lending categories (OF, CF, PF, IPRE, HVCRE).
  • (e) IPRE enhanced: IPRE Strong exposures may use column A if all four sub-conditions (i)–(iv) are met — see verbatim quote below. The test combines (i) an underwriting-quality bar, (ii) a very-low-LTV bar, (iii) an investment-grade income stream bar that expressly includes tenant income ≥ 100% of the obligor's debt service obligations, and (iv) exclusion of ADC (land acquisition, development and construction of commercial real estate) characteristics.
  • (f) PF enhanced: PF Strong exposures may use column A if the institution's underwriting and the exposure's other characteristics are substantially stronger than required by the Strong rating grade. Unlike (e), Art. 153(5)(f) contains no quantitative sub-conditions — it is a single substance-over-form test.

Art. 153(5)(c)–(f) — verbatim (PS1/26 Appendix 1 pp. 102–103)

(c) subject to points (d) to (f) of this paragraph an institution shall:

  • (i) assign the relevant risk weight in column B of Table A to exposures assigned to the 'Strong' rating grade;
  • (ii) assign the relevant risk weight in column D of Table A to exposures assigned to the 'Good' rating grade;
  • (iii) assign the relevant risk weight in the 'Satisfactory' column of Table A to exposures assigned to the 'Satisfactory' rating grade; and
  • (iv) assign the relevant risk weight in the 'Weak' column of Table A to exposures assigned to the 'Weak' rating grade.

(d) an institution may, for all categories of specialised lending exposures, if less than 2.5 years remain until maturity of an exposure:

  • (i) for exposures assigned to the 'Strong' rating grade: assign the relevant risk weight in column A of Table A to the exposure instead of the risk weight in column B of Table A; and
  • (ii) for exposures assigned to the 'Good' rating grade: assign the relevant risk weight in column C of Table A to the exposure instead of the risk weight in column D of Table A;

(e) an institution may, for IPRE exposures assigned to the 'Strong' rating grade, assign the relevant risk weight in column A of Table A to the exposure instead of the risk weight in column B of Table A if:

  • (i) the institution's underwriting of the exposure and the exposure's other characteristics are substantially stronger than required by the 'Strong' rating grade;
  • (ii) the loan to value ratio is very low for the property type;
  • (iii) the income stream on which the repayment of the obligation depends is consistent with that which the institution would reasonably expect for an investment grade exposure, including that the tenant income from the property is at least 100% of the obligor's debt service obligations; and
  • (iv) the exposure does not finance the land acquisition, development and construction ('ADC') of commercial real estate;

(f) an institution may, for project finance exposures assigned to the 'Strong' rating grade, assign the relevant risk weight in column A of Table A to the exposure instead of the risk weight in column B of Table A if the institution's underwriting of the exposure and the exposure's other characteristics are substantially stronger than required by the 'Strong' rating grade;

Asymmetry between (e) and (f) — why IPRE carries four tests but PF only one

Art. 153(5)(e) overlays three additional quantitative tests on top of the substantial-strength bar in (i): LTV floor (ii), investment-grade income-stream / tenant-coverage test (iii), and ADC exclusion (iv). Art. 153(5)(f) stops at the substantial-strength bar alone. The difference reflects IPRE's dependence on identifiable property-level cash flows (tenant income, LTV) — which lend themselves to hard quantitative gates — whereas project-finance cash flows derive from a sponsor's completed project and are evaluated holistically in the slotting factor lists (Appendix 1 List 2). Firms cannot import the (e)(ii)–(iv) tests into (f) by analogy.

Art. 153(5)(e)(iii) — the income-stream test is not just 'tenant ≥ 100%'

A common paraphrase of (e)(iii) reduces it to "investment-grade tenant income (≥ 100% debt service)". That conflates two requirements. The primary test is that the income stream on which repayment depends is consistent with an investment-grade exposure — the tenant-income coverage at ≥ 100% of the obligor's (not the property's) debt service obligations is one component ("including that...") of that broader investment-grade test, not the whole test. A property with a single non-investment-grade tenant paying 110% of debt service would fail (e)(iii) despite clearing the numerical threshold.

PRA vs BCBS — Maturity Differentiation Preserved

BCBS CRE33 removes maturity-based differentiation entirely and uses flat risk weights. PRA PS1/26 retains the maturity-based differentiation from CRR by structuring Table A with A/B/C/D columns. The values are identical to the CRR maturity-split tables (CRR "≥ 2.5yr" = column B/D; CRR "< 2.5yr" = column A/C). Column A/C assignment is explicitly optional ("may") under Art. 153(5)(d)–(f).

Column A/C Concession — Maturity Subgrades Implemented; Enhanced-Underwriting Concession Outstanding

Maturity-based column A/C assignment is now implemented for Basel 3.1 slotting via engine/slotting/transforms.py::lookup_rw, which dispatches on is_short (is_short_maturity derived from remaining_maturity < 2.5) against distinct short-maturity weight entries supplied by the b31 pack:

  • Non-HVCRE short maturity (P1.97 closed): the b31 pack supplies a slotting_rw_short entry; lookup_rw fires weights["short"] for short-dated non-HVCRE exposures. Short-dated Strong receives 50% (col A) and Good 70% (col C).
  • HVCRE short maturity (P1.117 closed): the b31 pack supplies a slotting_rw_hvcre_short entry; lookup_rw fires weights["hvcre_short"] for short-dated HVCRE exposures. Short-dated Strong receives 70% (col A) and Good 95% (col C).
  • Enhanced-underwriting concessions (Art. 153(5)(e) IPRE, Art. 153(5)(f) PF) remain not implemented — no input field exists to mark an exposure as meeting the enhanced-criteria threshold.

CRR maturity-based differentiation is implemented the same way, via the crr pack's slotting_rw_short / slotting_rw_hvcre_short entries read by the CRR branch of lookup_rw.

Default Category (0% RW, EL Treatment)

For defaulted specialised lending exposures assigned to the Default slotting category:

  • Risk weight = 0% (no capital charge via RWA)
  • Expected loss = 50% of EAD
  • The capital impact is captured through the EL shortfall/excess mechanism (Art. 159), not through the risk weight

This treatment is unchanged from CRR.

Routing to Slotting

Under Basel 3.1, exposures are routed to slotting based on specialised lending type:

IPRE / HVCRE (Art. 147A(2)): Must use slotting. These sub-types cannot use F-IRB or A-IRB under Basel 3.1 (unless the firm has specific A-IRB approval, which is rare).

PF, OF, CF: Routed to slotting when:

  1. The exposure is classified as specialised lending
  2. The firm does not have A-IRB permission for the sub-class
  3. The firm has F-IRB or slotting permission — PF/OF/CF may use F-IRB if the firm has F-IRB permission and can estimate PD

If the firm has A-IRB permission for the specific SL sub-class, the exposure may use A-IRB instead of slotting (see Model Permissions).

Key Scenarios

Scenario ID Description Expected RW
B31-E1 PF Strong, operational 70%
B31-E2 IPRE Good 90%
B31-E3 IPRE Weak 250%
B31-E4 HVCRE Strong 95%

Acceptance Tests

Group Scenarios Tests Pass Rate
B31-E: Slotting Approach E1–E4 13 100% (13/13)