Credit, Counterparty and Settlement Risk

Introduction

Treasury activities connect institutions through loans, deposits, derivatives, securities transactions, foreign-exchange contracts and payment systems. These activities create the possibility that a borrower or trading counterparty will fail to perform, or that a payment will be made without receiving the expected counter-payment. Credit, counterparty and settlement risk therefore affect both profitability and financial stability.

Credit risk is the risk of loss arising from a borrower’s failure to meet contractual obligations. Counterparty credit risk is the risk that the other party to a financial contract will default before the final settlement of all contractual cash flows. Settlement risk is the risk that one party delivers cash or securities but does not receive the agreed counter-performance. The risks overlap, but they are not identical: credit risk usually concerns a funded exposure, while counterparty risk is often dynamic and depends on future market values; settlement risk focuses on the timing and mechanics of exchange.

A sound treasury function identifies exposures, measures them under normal and stressed conditions, mitigates them through collateral and netting, controls concentrations, and ensures that payments are released only within approved operational and legal safeguards.

Learning Objectives

After studying this chapter, students should be able to:

1. Distinguish credit, counterparty and settlement risk.
2. Explain ratings, probability of default, loss given default and exposure at default.
3. Calculate expected loss and interpret its limitations.
4. Assess current and potential future exposure in derivatives.
5. Explain collateral, close-out netting, limits and wrong-way risk.
6. Describe settlement controls and payment-versus-payment mechanisms.
7. Evaluate governance, regulatory and ethical issues in counterparty risk management.

1. The Nature of Credit and Counterparty Exposure

A bank’s exposure is the economic value that may be lost if an obligor defaults. For a loan, exposure is generally related to principal and accrued interest. For a derivative, exposure depends on whether the contract has positive replacement value to the bank. If the contract is out of the money, immediate counterparty exposure may be zero, although future exposure can still arise.

Credit exposure has four important dimensions:

• Identity: the legal entity that owes the obligation.
• Amount: the loss-relevant value at default.
• Maturity: the period during which default can occur.
• Dependence: the relationship between the counterparty’s credit quality and the value of the transaction.

Counterparty exposure is commonly separated into:

• Current exposure: the positive replacement cost at the measurement date.
• Potential future exposure (PFE): an estimate of how exposure may increase before maturity.
• Settlement exposure: the amount at risk during the interval between an outgoing and incoming payment.

The legal entity, branch, guarantor and parent company must be identified carefully. A group-wide name is not necessarily a single legal obligor. Guarantees, collateral agreements, bankruptcy regimes and resolution rules may differ by entity and jurisdiction.

2. Ratings and Credit Assessment

External credit ratings summarize an agency’s opinion about an issuer’s or instrument’s relative creditworthiness. Internal ratings are produced by a bank using financial analysis, sector information, behavioural data and quantitative models. Ratings may be expressed through letter grades or numerical grades, but they are ordinal categories rather than precise probabilities.

A professional credit assessment normally combines:

• financial strength, liquidity and leverage;
• cash-flow generation and debt-service capacity;
• business model, governance and management quality;
• industry, country and legal risks;
• collateral and guarantees;
• market indicators such as bond spreads or credit-default-swap spreads;
• transaction structure, maturity and covenants.

Ratings are useful for initial screening, pricing, limit allocation and portfolio monitoring. They should not be used mechanically. Ratings can be stale, may react slowly to deterioration, and can fail to capture concentration, contagion or a rapidly changing market environment. A high rating does not mean zero default probability.

A credit migration matrix describes the probability that an obligor moves from one rating grade to another over a specified horizon. It supports stress testing, economic-capital analysis and pricing of longer-term exposure. However, historical transition frequencies may not represent future conditions, particularly during crises.

3. The Expected-Loss Framework

The central credit-risk decomposition is:

where:

• EL is expected loss over the chosen horizon, expressed in currency;
• PD is probability of default over that horizon, expressed as a decimal;
• LGD is loss given default, expressed as a decimal of exposure after recoveries;
• EAD is exposure at default, expressed in currency.

The formula assumes that PD, LGD and EAD are defined consistently for the same obligor, facility, horizon and default event. It is a simplified expected-value relationship. In practice, the variables may be correlated, especially in a downturn, and cash-flow timing, discounting and recoveries must be treated consistently.

LGD can be represented as:

If recovery proceeds are received after default, a more refined analysis discounts expected recoveries and includes collection costs. Collateral can reduce LGD, but its value depends on enforceability, liquidity, haircuts, valuation frequency and the time required to realise it.

EAD for a funded loan may include drawn principal, accrued interest and expected additional drawings. For a revolving facility, a common conceptual expression is:

where CCF is the credit-conversion factor, expressed as a decimal. This relationship is an estimation method, not a universal law. It is appropriate only when the institution has a defensible basis for estimating future drawings.

Worked Example: Expected Loss

A bank has a corporate loan with a one-year EAD of $8,000,000. The estimated one-year PD is 2.5%, and LGD is 40%.

The result is an expected loss of $80,000 for the specified horizon. It is not a prediction that exactly $80,000 will be lost. Actual loss may be zero or much higher because default is uncertain and recovery is variable.

If the loan has $5,000,000 drawn and a $3,000,000 undrawn commitment, and the estimated CCF is 50%, then:

At the same PD and LGD, expected loss would be:

The example illustrates why undrawn commitments can create material future exposure.

Expected Loss and Unexpected Loss

Expected loss is normally incorporated into pricing, provisions or reserves, subject to the applicable accounting framework. Unexpected loss is the adverse deviation from expected loss and is addressed through capital, limits and risk appetite. A portfolio can have modest expected loss but substantial unexpected loss if defaults are highly correlated.

Expected loss is also horizon-dependent. A one-year PD cannot be substituted uncritically for a lifetime PD. The analyst must specify whether probabilities are point-in-time or through-the-cycle, and whether they are marginal or cumulative.

4. Measuring Derivative Counterparty Exposure

For a derivative portfolio, the bank’s current exposure is often approximated by positive mark-to-market value:

This expression ignores close-out costs and certain legal adjustments, but it captures the basic principle: only a positive replacement value is an unsecured claim on the counterparty.

Because future market prices are uncertain, a counterparty may become exposed even when current value is zero. Potential future exposure can be estimated through scenarios or simulation. A simple expected-positive-exposure measure is:

where V_t is the simulated portfolio value at future time t, T is the number of time points, and E[ ] denotes the expected value across scenarios. The formula assumes equally weighted time points; practical systems may use time-weighted integration or regulatory exposure methods.

A conservative percentile measure is potential future exposure at confidence level q:

For example, 95% PFE is the exposure not exceeded in 95% of modelled scenarios at a specified future date. PFE is not a loss probability; it is an exposure percentile conditional on market scenarios.

Exposure models should incorporate interest rates, foreign exchange, equity prices, commodities, volatility, correlations, contractual maturities, collateral calls and margin disputes. Model outputs require independent validation and back-testing. Simplified add-on methods can be useful for limits and initial screening, but they may be inadequate for complex portfolios.

5. Collateral and Margin

Collateral transfers credit risk into collateral-management risk; they do not eliminate risk. Common collateral includes cash, government securities and highly liquid marketable instruments. A collateral agreement should specify eligible assets, valuation rules, haircuts, margin frequency, thresholds, minimum transfer amounts, dispute procedures, custody and default rights.

A haircut protects the secured party against market-value decline and liquidation costs. If securities worth $10 million have a 5% haircut, their recognised collateral value is:

The haircut must reflect price volatility, liquidity, currency mismatch, concentration and liquidation time. Excessive reliance on collateral can create procyclical pressure: falling prices trigger margin calls, which may force sales and further depress prices.

Variation margin covers current mark-to-market exposure. Initial margin is intended to cover potential exposure during the margin period of risk. Collateral is effective only if it is legally enforceable, operationally available and correctly valued. Segregation and rehypothecation rules also matter, particularly when a counterparty defaults.

6. Netting and Close-Out

Without netting, a bank may owe a counterparty on one contract while being owed money on another. Close-out netting allows eligible transactions under a master agreement to be terminated and combined into one net claim after default.

For a set of transactions with values V₁, V₂, …, Vₙ, the unsecured net exposure is conceptually:

This is beneficial only where the legal agreement is enforceable in the relevant jurisdictions and the transactions fall within its scope. Netting sets must be defined correctly. Operational systems must link all transactions to the correct legal agreement, currency and collateral account.

Payment netting reduces the number and value of settlement payments. Close-out netting operates after default. These mechanisms should not be confused with accounting netting or informal bilateral offsetting.

7. Limits and Concentration Management

Limits convert risk appetite into enforceable operating constraints. Examples include:

• counterparty total exposure limits;
• settlement limits;
• product and tenor limits;
• country and sovereign limits;
• unsecured limits;
• intraday limits;
• wrong-way-risk limits;
• concentration limits for groups or correlated names.

A limit framework should specify the measure used, such as current exposure, PFE, stressed exposure or exposure after collateral. It should identify utilisation thresholds, escalation procedures, temporary excess approvals and consequences of breach.

Concentration is not eliminated merely because individual names satisfy their limits. A portfolio may be concentrated in one sector, currency, country, parent group or common funding source. Aggregation must consider legal entities and economically connected counterparties. A guarantee may transfer risk to a guarantor, but it does not necessarily remove it.

8. Wrong-Way Risk

Wrong-way risk occurs when exposure to a counterparty increases as that counterparty’s credit quality deteriorates. It is the opposite of useful diversification between exposure and default risk.

General wrong-way risk arises from broad relationships, such as exposure to highly leveraged firms when market volatility rises. Specific wrong-way risk arises from a direct relationship between the counterparty and the transaction: for example, purchasing protection on a counterparty’s own debt, or accepting shares issued by the counterparty as collateral.

Risk managers should identify causal links, not merely calculate historical correlation. Controls may include refusing certain collateral, applying higher haircuts, imposing separate limits, using stress scenarios, requiring initial margin and escalating approval for transactions with embedded dependence.

9. Settlement Risk and Settlement Controls

Settlement risk is especially important in foreign exchange, securities and payment transactions. In a principal-risk or “Herstatt” event, one party pays in one currency or asset but fails to receive the counter-payment because the other party defaults or a payment system closes before settlement.

The settlement process should include:

1. verified standing settlement instructions;
2. independent confirmation and matching;
3. segregation of trade capture, confirmation and payment release;
4. sanctions, fraud and anti-money-laundering screening;
5. cut-off and time-zone controls;
6. intraday liquidity monitoring;
7. reconciliation of nostro and custody accounts;
8. exception queues and aged-item escalation;
9. dual authorisation for high-value payments;
10. tested contingency and business-continuity procedures.

Delivery-versus-payment (DvP) links delivery of securities to payment so that one leg does not occur without the other. Payment-versus-payment (PvP) links payment in one currency to payment in another, reducing foreign-exchange principal risk. Central counterparties may interpose themselves between buyers and sellers and apply margin, but they introduce concentration and liquidity risks of their own.

Settlement exposure should be measured by amount and duration. A payment that remains irrevocable for several hours can create material intraday exposure even if the end-of-day balance appears small.

10. Governance, Regulation and Ethics

Counterparty risk management requires clear accountability among the front office, risk function, operations, legal department, finance and senior management. The first line originates and manages transactions; the second line sets policy, measures risk and challenges decisions; internal audit provides independent assurance.

Key governance practices include documented risk appetite, approved counterparties, independent limit monitoring, model validation, legal-opinion review, collateral dispute governance, data-quality controls and board reporting. Regulatory capital frameworks generally recognise exposure at default, risk weights, collateral, netting and maturity, but regulatory treatment does not replace economic risk management.

Ethical conduct is central. Staff must not conceal limit breaches, delay recognition of a deteriorating counterparty, manipulate valuations, misuse collateral or release payments without proper authority. Conflicts of interest can arise when revenue targets encourage excessive unsecured exposure. A culture that rewards escalation is therefore a control mechanism, not merely a human-resources preference.

11. Technology, Data and Artificial Intelligence

Modern treasury systems integrate trading, risk, collateral, settlement and accounting data. Essential controls include a single counterparty identifier, accurate legal-agreement mapping, timestamped valuations, audit trails, access management and reconciliations between front-office and independent systems.

Artificial intelligence can support early-warning indicators, document extraction, anomaly detection and payment-fraud screening. Its use requires explainability, data-quality testing, bias assessment, human oversight and controlled change management. An AI-generated risk score should not override legal analysis, expert judgement or approved limits without governance. Model drift is particularly dangerous when relationships change during crises.

12. Illustrative Case: Foreign-Exchange Counterparty Failure

Background

Bank Meridian enters a six-month foreign-exchange forward with Global Export Ltd. The forward has a current positive market value of $1.2 million to Meridian. The parties operate under a legally enforceable master agreement with daily variation margin. Global Export has posted $0.9 million in eligible government securities, subject to a 10% haircut.

Problem

Global Export experiences severe financial stress. At the time of a margin call, market volatility increases and the bank must estimate both current unsecured exposure and possible future exposure. The counterparty also earns most revenue in the same currency that drives the value of the forward, creating potential specific wrong-way risk.

Analysis

The recognised collateral value is:

Ignoring close-out costs, current unsecured exposure is:

The bank then runs stressed exchange-rate scenarios. In several scenarios, the forward’s value rises substantially while Global Export’s cash flow declines. This dependence means that the average exposure estimate understates risk. The risk committee also discovers that the collateral agreement permits a delay in substituting securities and that a margin dispute is unresolved.

Approach and Outcome

Meridian temporarily reduces the unsecured limit, requires additional cash margin, prohibits new trades with the counterparty, and escalates the dispute to legal and operations teams. It applies a separate wrong-way-risk add-on and monitors settlement instructions under dual control. When Global Export defaults, the enforceable close-out netting agreement reduces the portfolio to one claim, while the haircut and liquidation period prevent overstatement of collateral value.

Learning Points

The case demonstrates that marked-to-market exposure is only a starting point. Collateral quality, legal enforceability, future market movements, margin disputes, concentration and dependence between exposure and default all affect the true risk. Strong settlement and data controls are necessary even when the credit model appears satisfactory.

13. Limitations and Best Practices

No single metric captures counterparty risk completely. PD estimates can be unstable; LGD depends on uncertain recoveries; EAD models may fail during market discontinuities; ratings may lag; and historical correlations can break down. Netting and collateral benefits may be unavailable if legal documentation is defective. Stress testing should therefore complement expected-loss calculations.

Best practice is to:

• use multiple sources of credit information;
• distinguish legal, economic and operational exposure;
• combine current exposure with PFE and stress exposure;
• validate PD, LGD, EAD and collateral models;
• test wrong-way and concentration scenarios;
• maintain enforceable master agreements;
• reconcile collateral and settlement data daily;
• monitor intraday as well as overnight exposure;
• set limits that remain meaningful under stress;
• document overrides and escalate exceptions promptly.

Chapter Summary

Credit risk concerns failure to meet an obligation; counterparty risk concerns possible default during the life of a financial contract; settlement risk concerns failure of an expected exchange of cash or assets. Expected loss is commonly estimated as EL = PD × LGD × EAD, but the variables must share a consistent horizon and definition. Ratings support assessment but do not replace analysis.

Derivative exposure includes current positive replacement cost and uncertain potential future exposure. Collateral reduces exposure only when it is liquid, appropriately valued and legally enforceable. Netting can materially reduce exposure when supported by valid agreements. Limits should address counterparties, groups, products, tenors, countries, settlement and wrong-way risk. DvP, PvP, reconciliations, segregation of duties and payment controls reduce settlement losses. Effective risk management combines quantitative models, legal documentation, operational discipline, governance and professional judgement.

Key Terms

Credit risk: Risk that an obligor fails to meet contractual obligations.

Counterparty credit risk: Risk of loss from a counterparty defaulting before final settlement.

Settlement risk: Risk that one party delivers but does not receive the counter-performance.

Probability of default (PD): Likelihood of default over a specified horizon.

Loss given default (LGD): Percentage of exposure lost after recoveries.

Exposure at default (EAD): Exposure expected to be outstanding when default occurs.

Expected loss (EL): Average loss estimated from PD, LGD and EAD.

Current exposure: Positive replacement value at the measurement date.

Potential future exposure (PFE): A specified percentile of possible future exposure.

Collateral: Assets pledged to reduce credit loss.

Haircut: Reduction applied to collateral value to reflect risk.

Netting: Combining eligible amounts into a single net obligation.

Close-out netting: Netting transactions after default and termination.

Credit-conversion factor (CCF): Factor used to estimate future use of undrawn commitments.

Wrong-way risk: Risk that exposure rises as counterparty credit quality worsens.

Delivery-versus-payment (DvP): Linked exchange of securities and payment.

Payment-versus-payment (PvP): Linked exchange of payments in different currencies.

Settlement limit: Maximum approved amount exposed during settlement.

Margin: Collateral transferred to cover current or potential exposure.

Review Questions

Short Answer Questions

1. Distinguish credit risk from settlement risk.
2. Define PD, LGD and EAD.
3. Why is a rating not equivalent to a default probability?
4. What is the purpose of a collateral haircut?
5. Explain the difference between current exposure and PFE.

Long Answer Questions

1. Explain the expected-loss framework, including its assumptions and limitations.
2. Discuss how close-out netting and collateral reduce counterparty exposure.
3. Describe a comprehensive settlement-risk control framework.
4. Explain general and specific wrong-way risk with examples.
5. Evaluate the role of governance and the three-lines model in treasury counterparty risk.

Analytical / Application Questions

1. A facility has a drawn amount of $4 million, an undrawn commitment of $2 million, a CCF of 60%, PD of 3% and LGD of 45%. Calculate EAD and expected loss.
2. A derivative has a positive market value of $2.5 million and collateral with a market value of $2 million subject to a 15% haircut. Calculate current unsecured exposure and identify two additional risks that should be investigated.
3. Design a limit and monitoring response for a counterparty whose PFE is rising while its credit rating is falling and its collateral is highly correlated with its own share price.

Multiple Choice Questions with Answers

Question 1

Which expression represents the basic expected-loss relationship?

A. EL = PD × LGD × EAD

B. EL = PD + LGD + EAD

C. EL = EAD ÷ (PD × LGD)

D. EL = LGD − PD × EAD

Correct Answer: A. EL = PD × LGD × EAD

Question 2

A loan has EAD of $5,000,000, PD of 2% and LGD of 50%. What is expected loss?

A. $5,000

B. $50,000

C. $100,000

D. $250,000

Correct Answer: B. $50,000

Question 3

What does a collateral haircut primarily recognise?

A. The borrower’s accounting profit

B. Potential decline in value and liquidation costs

C. The probability that the bank earns interest

D. The legal maturity of an unrelated loan

Correct Answer: B. Potential decline in value and liquidation costs

Question 4

Which measure is generally the positive replacement value of a derivative at the measurement date?

A. Current exposure

B. Recovery rate

C. Credit-conversion factor

D. Settlement finality

Correct Answer: A. Current exposure

Question 5

What is wrong-way risk?

A. Exposure falls when credit quality deteriorates

B. Exposure and default risk are unrelated

C. Exposure increases as the counterparty’s credit quality worsens

D. A payment is settled earlier than scheduled

Correct Answer: C. Exposure increases as the counterparty’s credit quality worsens

Question 6

Which mechanism links delivery of securities to payment?

A. PFE

B. DvP

C. PD

D. LGD

Correct Answer: B. DvP

Question 7

A $10 million security has a 5% haircut. What is its recognised collateral value?

A. $9.5 million

B. $9.05 million

C. $10.5 million

D. $5 million

Correct Answer: A. $9.5 million

Question 8

What is the main purpose of a counterparty limit?

A. To guarantee that default cannot occur

B. To convert all unsecured exposure into collateral

C. To constrain exposure within approved risk appetite

D. To eliminate the need for settlement controls

Correct Answer: C. To constrain exposure within approved risk appetite

Question 9

Which statement about ratings is most accurate?

A. A high rating means zero probability of default

B. Ratings are useful but should be supplemented by independent analysis

C. Ratings measure only settlement-system capacity

D. Ratings automatically capture every wrong-way relationship

Correct Answer: B. Ratings are useful but should be supplemented by independent analysis

Question 10

Which control most directly reduces foreign-exchange principal settlement risk?

A. Payment-versus-payment settlement

B. Increasing the undrawn commitment

C. Removing payment authorisation controls

D. Ignoring time-zone differences

Correct Answer: A. Payment-versus-payment settlement