FINS3650 · International Banking
Credit Risk Management in Banks
Week 5's credit-risk half sets out the principles of sound bank credit-risk management from the BCBS 'Principles for the management of credit risk' — establishing an appropriate credit-risk environment, sound granting criteria, appropriate administration/measurement/monitoring, and adequate controls — and links these to provisioning, problem-asset treatment and how credit risk flows into Basel capital via expected loss. It is examined as a gap-checklist evaluation of a lending policy and as an expected-loss calculation.
What this chapter covers
- 01Credit risk defined: the potential that a borrower or counterparty fails to meet obligations on agreed terms; the goal is to maximise risk-adjusted return within acceptable limits
- 02Sources beyond loans: interbank, trade finance, FX, derivatives, guarantees, settlement (Herstatt/settlement risk)
- 03The BCBS 17 principles across five areas: appropriate environment, sound granting, administration/measurement/monitoring, adequate controls, supervisors' role
- 04Common causes of credit problems: concentrations (the single biggest cause), credit-process weaknesses, market- and liquidity-sensitive exposures (wrong-way risk)
- 05Expected loss and the IRB parameters: EL = PD × LGD × EAD; 12-month vs lifetime ECL under IFRS 9 by stage
- 06Loan policy, credit grading (pass/special-mention/substandard/doubtful/loss) and the loan-review system feeding provisions
- 07Provisioning and problem-asset treatment; how credit risk feeds RWA and capital
- 08International-lending overlay: country/sovereign and transfer risk on cross-border credit
Expected loss before and after a downgrade
- +1State the formula: expected loss EL = PD × LGD × EAD, where PD is the one-year probability of default, LGD the loss given default and EAD the exposure at default.
- +1Before: EL = 0.01 × 0.45 × 10,000,000 = $45,000. This is the 12-month expected credit loss the bank would provision for a performing (Stage 1) exposure.
- +1After the downgrade: EL = 0.04 × 0.45 × 10,000,000 = $180,000. Only the PD changed (1% → 4%), and expected loss scales linearly with PD, so it quadruples.
- +1Comment: the provision rises from $45,000 to $180,000, a four-fold increase driven entirely by the higher PD. If the deterioration also moves the exposure to a significant-increase-in-credit-risk stage under IFRS 9, the bank must switch from 12-month to lifetime ECL, raising the provision further and pressuring capital via higher RWA.
Key terms
- Credit risk
- The potential that a bank borrower or counterparty fails to meet its obligations on agreed terms; the goal of credit-risk management is to maximise the bank's risk-adjusted return by keeping exposure within acceptable limits at both the individual-credit and portfolio level.
- Expected loss (EL)
- EL = PD × LGD × EAD, the probability-weighted loss on an exposure, where PD is the probability of default, LGD the loss given default and EAD the exposure at default; it scales linearly with each input.
- Expected Credit Loss (ECL) staging
- Under IFRS 9, exposures sit in Stage 1 (12-month ECL, performing), Stage 2 (lifetime ECL after a significant increase in credit risk) or Stage 3 (lifetime ECL, credit-impaired); a downgrade can trigger a stage migration that sharply raises provisions.
- Concentration risk
- The single most important cause of major credit problems — excessive exposure to one borrower, connected group, sector or correlated risk factor; managed through single-name and portfolio limits and stress testing.
- Settlement (Herstatt) risk
- The risk that one side of a transaction delivers value while the counterparty fails before completing its leg; named after the 1974 Herstatt failure and a driver of the founding of the BCBS.
- Loan grading
- A granular internal classification of credit quality — pass, special mention, substandard, doubtful, loss — driven by cash-flow coverage, leverage and collateral, feeding the loan-review system, provisioning and loan pricing.
Credit Risk Management in Banks FAQ
What is expected loss and how is it calculated?
Expected loss is the probability-weighted loss on a credit exposure: EL = PD × LGD × EAD, where PD is the probability of default, LGD the loss given default (the fraction not recovered) and EAD the exposure at default. Because it is a simple product, expected loss scales linearly with each input, so doubling the PD doubles the expected loss. Under IFRS 9 banks provision 12-month ECL for performing exposures and lifetime ECL once credit risk has increased significantly.
What is the single biggest cause of major credit problems?
Concentration — excessive exposure to a single borrower, a connected group, a sector, or a correlated risk factor. The BCBS material identifies it as the most important cause, whether it is an obvious single-name concentration or a subtle correlated-risk-factor concentration that only reveals itself in a shock. It is managed with single-name and portfolio limits, granular management information, and stress testing.
How does a sound credit-risk management program look in practice?
The BCBS principles group into five areas: an appropriate credit-risk environment (board-approved strategy and tolerance), a sound credit-granting process (clear criteria and limits, arm's-length related-party lending), appropriate administration, measurement and monitoring (internal risk ratings, portfolio monitoring, stress testing), adequate controls (independent review, exception reporting, early remedial action), and the supervisor's role. A lending policy is evaluated by checking whether each of these is present.
Can AI help me with the credit-risk topic?
Yes, as a study aid. Sia can walk you through the expected-loss calculation, the IFRS 9 staging logic, the five areas of the BCBS principles and the common causes of credit problems, and help you turn a weak lending policy into a gap checklist. It teaches the method and checks your reasoning; it does not do your graded UNSW assessment, and the academic-integrity policy applies.
Exam move
Approach this chapter as one calculation and one checklist. The calculation is expected loss EL = PD × LGD × EAD — practise it before and after a downgrade so you can articulate that EL is linear in PD and that an IFRS 9 stage migration (12-month → lifetime ECL) amplifies the provisioning hit and feeds RWA. The checklist is the sound-credit-risk-management program: memorise the five areas of the BCBS principles (environment, granting, administration/measurement/monitoring, controls, supervisors) and the common causes of credit problems (concentrations first, then process weaknesses and market/liquidity-sensitive exposures), so you can evaluate a given lending policy for what is missing. Keep the loan-grading ladder (pass → special mention → substandard → doubtful → loss) and settlement/Herstatt risk ready as short-answer items. Practise a gap-analysis paragraph on a weak policy, since that is a common exam shape. Ask Sia to set fresh PD/LGD/EAD numbers and a policy to critique; confirm assessment details on the FINS3650 course outline.
Working through Credit Risk Management in Banks in FINS3650? Sia is AskSia’s AI Business and Economics tutor — ask any FINS3650 Credit Risk Management in Banks question and get a clear, step-by-step explanation grounded in how FINS3650 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
