Prudential Regulation Authority Rulebook

6.1

This chapter sets out the methodology the PRA uses to inform the setting of a firm’s Pillar 2A capital requirement for single name, sector and geographical credit concentration risk in the banking and trading books.

6.2

Credit concentration risk is the risk of losses arising as a result of concentrations of exposures due to imperfect diversification. This imperfect diversification can arise from the small size of a portfolio or a large number of exposures to specific obligors (single name concentration) or from imperfect diversification with respect to economic sectors or geographical regions.

6.3

For the purposes of the methodology specified below, only wholesale credit portfolios are considered for single name and sector concentration risk (excluding securitisation, intra-group exposures¹⁵ and non-performing loans). All credit portfolios other than residential mortgage portfolios on the standardised approach are considered for geographic concentration risk.

6.4

Firms are required to calculate a credit concentration risk measure, the Herfindahl-Hirschman Index (HHI), for all relevant portfolios (single name, pre-defined industry sectors and geographic regions). The HHI is defined as the sum of the squares of the relative portfolio shares of all borrowers (these portfolio shares are calculated using risk-weighted assets (RWAs)). Well-diversified portfolios have an HHI close to 0, whilst the most concentrated portfolios have a number close to 1. The HHI is a good indicator of the level of credit concentration risk within a portfolio. Mapping models translate a firm’s HHI into a proposed capital add-on range. The table mapping the HHI for single name, sector and geographical credit concentration to capital add-on ranges is set out in Figure 1.

6.5

The mapping models for single name, sector and geographical credit concentration are described below.

6.6

The Gordy-Lütkebohmert (GL) methodology¹⁶ is an extension of the Basel risk-weight function and aims to quantify the undiversified idiosyncratic risk in a credit portfolio not considered to be sufficiently granular. The GL methodology uses credit risk parameters to quantify the single name risk in a portfolio and suggests the necessary capital add-on range to account for single name concentration risk.

6.7

When assessing the degree to which a firm might be subject to industry sector or geographical credit concentration risk, the PRA adopts a methodology based on published multi-factor capital methodologies (eg Düllmann and Masschelein).¹⁷

6.8

The PRA has constructed a benchmark portfolio based on the average lending distribution from a sample of well-diversified firms. The PRA developed a multi-factor capital model, which takes into account the default rate volatilities (intra-sector and intra-region correlation) of eight pre-defined geographic regions and industry sectors as well as default rate volatility correlations between pre-defined geographic regions and industry sectors (inter-sector and inter-region correlations).

6.9

Sectors are broadly aligned to standard industry classification (SIC) codes and NACE (Nomenclature of Economic Classification) codes (set out in Table B), while the geographical regions are based on the International Monetary Fund’s definition of the main global economic regions (set out in Table C). The United Kingdom is considered separately.

6.10

The multi-factor model is calibrated so that the capital requirement for a well-diversified lending portfolio (the benchmark portfolio) using the multi-factor model and a single risk factor model (on which the IRB framework is based) are equal. The PRA created a sequence of portfolios with increasing levels of concentration and compared the capital requirements derived from the multi-factor model with those derived from the single-factor risk model. The difference in the capital requirements between the multi-factor and single-factor risk model (capital add-ons) was compared to the HHI measures of concentration. The relationship between the two measures is strong. The PRA has therefore mapped the HHI measures to capital add-on ranges derived from its multi-factor capital model.

Figure 1 Concentration risk – mapping of capital add-on ranges to HHI

(*) 2.8% for CRE but 2% for financial.

Table B Breakdown of sectors

Agriculture, forestry and fishing

Construction

Financial industry (bank and non-bank)

Real estate (commercial)

Manufacturing

Mining and quarrying

Wholesale and retail trade

Services and other

Transport, storage and utilities

Table C Geographic breakdown

United Kingdom

North America

South/Latin America and Caribbean

European (west) area

Eastern Europe and Central Asia (including Russian Federation)

East Asia and Pacific

South Asia

Middle East and North Africa

Sub-Saharan Africa

6.11

Given a capital add-on range produced by the concentration risk models, the PRA exercises its judgement as to where within that range the capital add-on should be set. In order to promote consistency of judgement, the mid-point of the range acts as a starting point. When setting the Pillar 2A credit concentration risk capital add-on, the PRA may consider a range of factors including firms’ own concentration risk assessments; firms’ ability to manage concentration risk; the degree to which conservatism is reflected in firms’ Pillar 1 RWAs; instances where portfolio correlations are not adequately captured; any other factors not adequately captured under the quantitative assessment; and business models.

6.12

The PRA will continue to be proportionate in its approach to setting capital; supervisors may exercise judgement for small firms where they identify that the credit concentration risk methodology could overstate risks, or could incentivise risk-taking behaviour.

6.13

The quantitative methodologies informing the recommended capital add-on ranges have been constructed so as to apply independently of one another in order to avoid double counting. The capital add-on for credit concentration risk is therefore the sum of the respective add-ons for each credit concentration risk type.

6.14

The measure of credit concentration risk is based on the Pillar 1 risk assessment (ie the risk weighting of the obligor, sector or geographic regions). Exposures with low risk weights therefore attract a lower concentration risk add-on compared to exposures with higher risk weights, everything else constant.

6.15

Where the PRA considers that a firm’s credit risk RWAs do not accurately reflect the underlying credit risk within a portfolio, the Pillar 2A credit concentration risk capital add-on may be adjusted upwards.

6.16

Capital held against potential losses from credit valuation adjustments are excluded from the credit concentration risk assessment.

6.17

All firms must report the data contained in the credit concentration risk Pillar 2 data items in accordance with Reporting Pillar 2, 2.2. Firms are required to submit the data with their ICAAP submissions. These data items include information on the portfolio HHI for each of the concentration risk types and additional information on portfolio composition.