Chapter 6: Credit Risk Measurement Flashcards
Credit Risk Measurement Qualitative Factors
RETAIL
* Historically assessed by relationship managers.
* Now involves credit scorecards and other objective measures
Factors considered: (the 5 C’s – also consisered for companies)
* Security (Collateral/Surety) – “Collateral”
* Borrower (Qualifiations/Job/Age/Criminal record/Credit record) – “Character”
* Financials (Income/Expenses/Assets/Liabilities) – “Capital” and “Capacity”
* Economy and Loan Purpose (GDP/Interest rate) – “Conditions”
NON-RETAIL
Extra Factors considered:
* Management (Experience, Background, Past actions, Strategy statements vs. progress)
* Disclosure (Timeous, Verified, Relevant, Reporting team skills)
* Corporate Governance (Checks and balances, Protection against unethical and illegal activities)
Quantitative Measuring Systems Under Basel
1. Standardized approach (SA):
Relies on external credit ratings.
2. Internal ratings-based approach (IRBA):
Allows banks to develop their own internal ratings.
Key Parameters in Credit Risk Measurement
Probability of default (PD):
* The likelihood of a borrower defaulting on a credit obligation over a specific time horizon.
* PD can be point-in-time (PIT) or through-the-cycle (TTC).
* Definition of default can vary, but common definitions include 90 days past due, backruptcy, insolvency or breach of contractual conditions (like making payments on time or loan covenant violation)
Loss given default (LGD):
* The percentage of exposure (the amount owed by the borrower at the time of default) that a lender is expected to lose if the borrower defaults.
* Accounts for factors such as collateral and effectiveness of recovery efforts
Exposure at default (EAD):
* The total value a lender is exposed to when a borrower defaults.
* For fixed-term loans, it’s the outstanding balance at default
* For revolving credit facilities, it includes the drawn amount plus an estimate of potential future drawdowns
Exposure at Default (EAD) Modelling
Foundation approach:
* Fixed facilities is calculated as the outstanding exposure
* Revolving facilities use prescribed Credit Conversion Factors (CCFs)
Advanced approach:
* Banks can use their own internal models to estimate EAD for revolving facilities, subject to regulatory approval and floors.
Loss Given Default (LGD) Modelling
Foundation approach (F-IRB)
* Flat LGDs assigned to different asset classes
* Treatment of collateral is outlined (certain collateral types are not considered)
Advanced approach (A-IRB)
* Internal estimates of the LGD are calculated by the bank using internal models
* Subject to approval by the relevant supervisory body
Subjective methods:
Primarily driven by expert judgment, these methods are often used when data is limited or for specific types of exposures.
Objective methods:
These methods rely on formal mathematical procedures and can be further divided into…
1. Explicit methods:
a. Market LGD:
Observed from market prices of defaulted bonds or loans.
b. Workout LGD:
Estimated based on the cash flows recovered during the workout process.
2. Implicit methods:
a. Implied market LGD:
Derived from the prices of bonds that are considered high risk but have not yet defaulted.
b. Implied historical LGD:
Estimated using long-term historical average losses for similar asset types.
Probability of Default (PD) Modelling
Internal default experience:
* Banks can use their own historical data on defaults to estimate the probability of default for different types of borrowers or loans.
Mapping to External data:
* Banks can use default data from external sources, such as credit rating agencies, to estimate PDs.
* This is particularly useful for smaller banks with limited internal data.
Statistical default models:
Banks can develop statistical models to predict the probability of default.
common factors (Inputs) used:
1. Borrower characteristics
2. Loan terms
3. Economic conditions.
common statistical Methods:
1. Linear regression
2. Logistic regression
3. Discriminant analysis.
4. Panel models
5. Cox proportional hazards models
6. Neural networks.
Credit Conversion Factor
- A percentage applied to the undrawn portion of a revolving credit facility to estimate the amount likely to be drawn in the event of default.
- This is used to calculate the EAD for such facilities.
EAD[current estimate] = (Exposure Drawn) + (CCF x (Exposure Undrawn)
IRB models vs IFRS9 models
SITBRO
Source:
* IRBA based on Basel by BIS
* IFRS9 based on accounting regulation by IASB
Impact:
* IRBA primarily affects the balance sheet
* IFRS9 will affect both the balance sheet and income statement, primarily P&L
Time Horizon:
* IRBA has combinations of PIT and TTC PDs for the next 12 months. Parameters refreshed annually.
* IFRS9 looks at PIT PDs for the next 12 months (Stage 1) and remaining life (Stage 2+3). Outputs are continuously done.
Basis:
* IRBA looks at downturn LGDs with conservative scenarios
* IRBA is more conservative and includes floors
* IFRS9 looks at PIT LGDs with a range of scenarios
* IFRS9 looks at best estimate basis over multiple scenarios
Requirements:
* IRBA has specific disclosure details (e.g., ICAAP) primarily within risk-based functions.
* IRBA has specific data requirements
* IFRS9 will affect both the balance sheet and income statement, primarily P&L
* IFRS9 is outcomes oriented and is not explicit
Output Focus:
* IRBA focused on regulatory capital requirements:
Identifying possible defaults on assets.
* IRBA ensures banks can estimate and prepare for unexpected losses
* IRBA models used to calculate RWAs and risk parameters
* IFRS9 focused on regulatory provision requirements:
Identifying impaired assets
* IFRS9 ensures banks can estimate and prepare for expected losses
* IFRS9 models ECLs
Credit Valuation Adjustment (CVA) Approaches
Basic Approach (BA-CVA) – Reduced
* Used for banks not actively using hedges against counterparty credit spreads (i.e., CVA risk)
* ALL banks need to calculate capital under this methodology, as capital output is used in the other approaches
* Capital requirements are calculated first by an individual counterparty level, allowing for netting sets. This prescribed formula includes defined parameters like risk-weights, EAD, discount factors, and effective maturities.
* The total value for the whole portfolio is then calculated. Formula for whole portfolio is below (C[reduced])
Basic Approach (BA-CVA) – Full
* This approach accounts for hedging against counterparty credit spreads (e.g., credit default swaps)
* K[reduced] basis, with adjustments to get the full value
Standardised Approach (SA-CVA)
* This approach is more complex than the BA
* Minimum requirements must be met before this approach can be used, and it requires regulatory approval
* Adapted from Basel’s Market RIsk SA (though simpler)
* Must be able to report CVA sensitivities to market risk factors at least monthly and must have a CVA desk that will manage and hedge CVA risk.
* Similar process to BA-CVA with differences:
1. Indvidual counterparty capital requirements are determined using internal estimates of PD (market implied from credit spreads), LGD (market-implied) and Discounted Future Exposure (future transactions discounted at risk free rate, allowing for margins/collateral)
Counterparty Credit Risk (CCR) Approaches
Standardised approach (SA-CCR)
* EAD = 1,4 × (RC + PFE)
* RC = Replacement Cost (cost incurred if transaction needs to be replaced immediately at point of default). Will depend if transactions were margined or unmargined. Greater CCR if unmargined since limited collateral)
* PFE = Potential Future Exposure (cost between point of default and the initiation of a new transaction)
Internal Model Method (IMM) approach
* Model not prescribed, but left to discretion of the bank
* Subject to regulatory approval
* EAD = 1.4 x EEPE
* EE = Expected Exposure (the average exposure at any future date, capped at the date the last transaction matures or 1 year).
* EPE = Expected Positive Exposure (the maximum EE for that date or any preceding it)
* EEPE = Effective Expected Positive Exposure (the weighted average over time of the EPE, weighted according to date at which the last transaction matures). Focuses on out-the-money positions that could become credit exposures. Addresses ‘wrong way risk’ meaning counterparites’ positions may be correlated to their probability of default.
* EEPE = Formula below
* Regulators have the discretion to require a higher factor than 1.4, based on the bank’s CCR exposures
Counterparty Credit Risk (CCR) Definition
- The risk that a counterparty to a trade defaults before the contract matures, leading to a loss for the other party.
- Risk type that borders between market and credit risk
- Subject to both the creditworthiness of the counterparty to the trade and general market changes which may affect the trade.
- CCR is applicable in cases where there is a bilateral risk of loss, i.e. either party to a transaction could default.
- Economic loss would occur if the transactions or portfolio of transactions with the counterparty has a positive economic value at the time of default (i.e. is in the money)
- Mitigated through collateral agreements, netting agreements and credit limits
Credit Valuation Adjustments (CVA) Definition
MR BADS I RAM
- CVAs are capital charges to cover Mark-to-market losses.
- forms part of the Risk management, alongside CCR.
- introduced as part of Basel III
- the true (risk-adjusted) Asset values generally decrease when counterparty creditworthiness deteriorates.
- the CVA incorporates this Default risk into the trading book capital calculations to account for possible loss.
- prevents banks from Shifting assets from the banking book to the trading book to benefit from the capital requirement arbitrage
- charges done first at Individual counterpary level, and then at an aggregate level.
- Recognised central counterparties (e.g., central clearing houses) are exempted from this adjustment by Basel.
-
Adjustments are allowed for the following…
a. Netting
b. Collateral
c. offsetting internal and external Hedges - this allowance is to encourage banks to follow efficient risk Management
Credit Spread Definition
The difference between the yield on a corporate bond and a risk-free government bond with the same maturity is called the credit spread.
This spread reflects the market’s assessment of the corporation’s credit risk.
Market-Implied PD Curve and Survival Curve
By analyzing credit spreads across different maturities, a market-implied PD curve can be constructed, showing how the market perceives the probability of default changing over time.
The survival curve is the complement of the PD curve, illustrating the probability of the borrower not defaulting (surviving) until a specific time.
Types of Guarantees
PAPI BOSS DP
-
Payment guarantee
Ensures the seller that the purchase price will be paid on the agreed date if all contractual obligations are met -
Advance payment guarantee
Ensures the buyer that the advanced payment will be reimbursed if the seller does not meet contractual delivery obligations in full
Performance bond
Serves as collateral for costs incurred by the buyer due to failure of the seller to provide goods and services promptly and as contractually agreed -
letter of Indemnity
Secures the shipping company against any claims if goods are delivered prior to receipt of the original bill of lading -
Bid bond (tender bond)
secures the organiser’s expenses in tenders by requiring participants to pay if their bid is accepted but withdrawn -
warranty Obligations guarantee
Secures any claims by the buyer for defects appearing after delivery -
credit Security bond
Serves as collateral for loan repayment. -
Sovereign guarantees
Type of guarantee that backs projects deemed in the public interest, supporting development of infratrstructure, new industries, regions and exports. Many sovereigns have
state-owned development and export-import banks to facilitate these guarantees. -
Director guarantees
Personal guarantees, where directors will provide a guarantee on an agreement and can be held personally liable in the event of default. This is usually used in cases where an entity has limited resources and cannot provide collateral or alternative guarantees and/or where the directors’ involvement is essential for success (promotes commitment in having “skin in the game”). -
Parent company guarantees
Provided by an entity’s holding company when the bank is lending to a subsidiary of the group.
Two main types of netting agreements
Payment netting:
This type of netting allows parties to offset payments due on the same day.
Close-out netting:
This type of netting allows parties to terminate all outstanding transactions with each other in the event of a default.
The transactions are then valued at their market value, and the net amount owed is calculated.
This can significantly reduce the amount of money that a bank would lose if a counterparty defaults.
Limitations of Face-to-Face Credit Rating
- Requires expertise (limited)
- Subjectivity/Bias/Errors
- Lack of Consistency between Managers
- Qualitative standards could change
