Credit Derivatives and Counterparty Risk

Introduction and context

Credit risk is the possibility that a borrower, issuer, or trading counterparty will fail to perform contractual obligations, producing a loss for the creditor. Traditional credit management relies on loan approval, covenants, diversification, provisions, and capital. Credit derivatives add a separate layer: they allow credit risk to be transferred, transformed, priced, or embedded in securities without necessarily transferring ownership of the underlying loan or bond.

The most important instruments are credit default swaps (CDSs), total-return swaps (TRSs), credit-linked notes (CLNs), and securitized credit products such as collateralized loan obligations (CLOs). Their economic usefulness is accompanied by counterparty risk. A protection buyer may correctly forecast a reference entity’s default and still suffer a loss if the protection seller fails at the same time. Consequently, modern valuation considers exposure, collateral, netting, margin, wrong-way risk, and valuation adjustments such as CVA and DVA.

This chapter treats credit derivatives as both financial contracts and risk-management systems. The central distinction is between the credit risk of the reference entity and the credit risk of the derivative counterparty. They are related, but they are not identical.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Explain the structure and economic purpose of CDSs, TRSs, CLNs, and securitized credit products.
2. Distinguish default probability, recovery rate, loss given default, exposure at default, and expected loss.
3. Interpret CDS premiums using hazard rates, recovery assumptions, discounting, and accrued-premium effects.
4. Estimate counterparty exposure under collateral and legally enforceable netting arrangements.
5. Explain CVA, DVA, funding and margin-related adjustments and their limitations.
6. Evaluate the governance, regulatory, liquidity, model, and ethical risks of credit derivatives.

1. Credit risk foundations

A credit instrument has at least three analytically distinct quantities:

• Probability of default (PD): the probability that a borrower defaults over a specified horizon.
• Recovery rate (R): the proportion of contractual claim value recovered after default.
• Loss given default (LGD): the proportion lost after default.

The basic identity is:

If a bond with a claim amount of 100 has a recovery rate of 40%, its loss given default is 60%, before considering costs, timing, and market-value effects. For a fixed exposure E, the simplified expected credit loss is:

Here, EL and E are measured in currency units, while PD and LGD are proportions. This formula assumes that exposure and recovery are represented consistently and that the relevant default event is defined clearly. It is a one-period approximation; multi-period valuation requires discounting and allowance for the timing of default and recovery.

Credit risk is commonly represented through a hazard rate or default intensity, denoted λ(t). Under a reduced-form model, conditional survival probability from time 0 to T is:

If the hazard rate is constant, λ(u) = λ, then:

The cumulative default probability is:

These expressions assume a non-negative intensity and a model in which default is represented by an arrival event. They do not imply that the hazard rate is directly observable; it is usually inferred from bond prices, CDS spreads, or both.

A useful approximation links a CDS spread s to hazard rate and recovery:

Thus, with a 5% annual CDS spread and a 40% recovery rate, the approximate hazard rate is 5% ÷ 60% = 8.33% per year. This is only an approximation. Exact pricing incorporates discount factors, payment dates, survival probabilities, accrual on default, stochastic recovery, liquidity premiums, and the contractual definition of credit events.

2. Credit default swaps

A CDS is a bilateral contract in which the protection buyer pays a periodic premium to the protection seller. In return, the seller compensates the buyer if a specified reference entity experiences a contractually defined credit event. The reference entity is not necessarily a counterparty to the CDS.

The main terms include:

• Reference entity: borrower or issuer whose credit event determines settlement.
• Reference obligation: debt instrument used to establish deliverability or settlement terms.
• Notional amount: contractual amount used to calculate premium and protection payment.
• CDS spread: annual premium rate, commonly quoted in basis points.
• Credit events: events such as bankruptcy, failure to pay, or restructuring, as defined by documentation.
• Recovery or auction price: amount used to determine the protection payment.
• Maturity and payment frequency: dates on which premium is paid.

At default, a cash-settled CDS generally pays approximately:

where N is notional and the recovery price is expressed as a fraction of par. If N = 10 million and the auction recovery price is 35%, the payment is 6.5 million. The actual amount depends on contract terms, settlement procedures, accrued premium, and whether the contract is physically or cash settled.

Premium and protection legs

The value of a CDS is the present value of expected protection receipts minus the present value of expected premium payments:

For a simplified annual-payment model with payment dates tᵢ, discount factor D(0,tᵢ), survival probability S(0,tᵢ), notional N, and constant recovery R, the premium leg is approximately:

The protection leg can be approximated as:

Here, s is the annual CDS spread, αᵢ is the accrual fraction for period i, and t₀ = 0. The second expression approximates default as occurring at payment dates or within discretized intervals. A more accurate implementation integrates over default time and includes premium accrued up to default.

At inception, a par CDS has a premium rate such that the two legs have approximately equal value. The market spread is therefore a risk-neutral price, not a pure forecast of the real-world default frequency. It can include risk premia, liquidity, funding effects, counterparty adjustments, and technical supply-demand effects.

CDS uses and risks

A protection buyer may hedge a bond, loan, or portfolio; express a negative credit view; reduce concentration; or separate interest-rate risk from credit risk. A protection seller may earn premium income, diversify credit exposure, or take a view that realized losses will be lower than implied by the spread.

Important risks include basis risk between the hedged asset and CDS, jump-to-default risk, liquidity risk, documentation risk, settlement risk, and counterparty risk. A CDS hedge can fail if the reference obligation differs materially from the hedged asset or if the hedge notional and maturity are poorly aligned.

3. Total-return swaps

A total-return swap transfers the economic total return of a reference asset. The total-return payer transfers price appreciation, coupon or interest income, and usually price depreciation to the total-return receiver. In exchange, the receiver pays a financing rate, commonly a floating benchmark plus a spread.

For a reference asset with beginning value V₀, ending value Vₜ, income I, and financing rate k over time interval Δt, the asset-side total return is:

The net cash flow to the total-return receiver, before collateral and other adjustments, is approximately:

where N is the contractual reference amount. The receiver obtains synthetic exposure without purchasing the asset. The payer retains legal ownership but transfers economic performance.

A TRS differs from a CDS because it transfers total economic performance rather than only a specified credit-event loss. A TRS receiver therefore bears spread widening, mark-to-market, interest-rate, liquidity, and potentially default risk. It may be used to finance a bond position, provide leveraged exposure to a loan portfolio, or transfer a bank’s credit exposure to an investment fund.

The principal risks are leverage, collateral calls, valuation disputes, wrong-way risk, and termination risk. If the reference asset falls sharply, the receiver may owe a large mark-to-market payment while also facing reduced liquidity.

4. Credit-linked notes

A credit-linked note combines a debt security with an embedded credit derivative. The investor lends money to the issuer and receives interest, but repayment is reduced if a specified reference credit event occurs. The investor therefore assumes both the issuer’s default risk and the reference entity’s credit risk.

Suppose a CLN has principal N and recovery-linked payment R × N after a reference default. The investor’s default-related loss is:

The note may offer a higher coupon than an otherwise comparable ordinary bond because the investor is selling credit protection. A CLN can be attractive when investors seek enhanced yield or targeted credit exposure, while an arranger may use it to transfer risk and obtain funded protection.

CLNs introduce structural complexity. Investors must analyze the issuer or special-purpose vehicle, collateral arrangements, subordination, trigger definitions, calculation-agent powers, liquidity, and the interaction between issuer default and reference-entity default. Correlation between these defaults can create wrong-way risk.

5. Securitized credit products

Securitization pools loans, bonds, receivables, or other assets and finances them through tranched liabilities. Cash flows are allocated according to a contractual waterfall. Senior tranches receive priority, while junior or equity tranches absorb losses first.

If a pool has asset value A and losses L, tranche losses depend on attachment and detachment points. For a tranche with attachment point a and detachment point d, expressed as fractions of pool notional, the approximate loss fraction is:

This formula assumes losses are measured as a fraction of pool notional and ignores timing, recoveries after delay, and structural features. The tranche notional is (d − a) times pool notional.

For example, consider a 100 million pool and a tranche attaching at 3% and detaching at 7%. If pool loss is 5 million, the loss above attachment is 2 million. The tranche absorbs 2 million of loss out of its 4 million notional, or 50%. If pool loss is 10 million, the tranche is fully exhausted because losses above 3 million exceed the 4 million tranche size.

Important securitization risks include asset correlation, model risk, servicer performance, prepayment, extension, liquidity, tranche subordination, and adverse selection. Diversification reduces idiosyncratic risk only when defaults are not highly correlated. Systemic stress can cause apparently diversified assets to default together.

6. Counterparty exposure, collateral, and netting

Counterparty risk is the risk that the other party to a derivative fails to pay when the contract has positive replacement value. The relevant quantity is not simply the notional. It is the amount that would be lost after considering current value, collateral, legally enforceable netting, close-out, and recovery on the counterparty.

For a single transaction, a simplified current exposure is:

where V is the market value owed to the institution and C is collateral held, with consistent sign conventions. For a netting set containing transactions j:

Netting is valid only where enforceability has been established, particularly under insolvency in relevant jurisdictions. Gross notional should not be confused with exposure: two offsetting trades may have a large gross notional but small net replacement cost.

Collateral reduces exposure but does not eliminate risk. Residual risks include margin-call delay, collateral price volatility, wrong-way collateral, disputes, operational failure, rehypothecation, and the period between the last margin exchange and close-out. Initial margin is designed primarily to cover potential future exposure during the margin period of risk; variation margin generally covers current mark-to-market exposure.

Potential future exposure (PFE) estimates possible future positive exposure at a specified confidence level. Expected exposure at time t is:

The expectation is taken under a specified market-risk simulation and is not the same as expected loss. Exposure describes the amount at risk; default probability and recovery determine how much of that exposure may be lost.

7. CVA, DVA, and related valuation adjustments

Credit valuation adjustment

CVA is an adjustment for the possibility that a counterparty will default before a derivative portfolio is terminated. A simplified unilateral CVA formula is:

LGD_C is counterparty loss given default, DF is a risk-free discount factor, EE is expected positive exposure, and ΔPD_C is the marginal default probability over interval i. This formula assumes exposure and default probability can be combined in this way and generally treats their dependence approximately. More advanced models integrate discounted exposure over default time and model dependence between exposure and default.

If expected exposure is 2 million, counterparty LGD is 60%, the marginal default probability over a one-year interval is 2%, and the discount factor is 0.98, the interval CVA contribution is:

2,000,000 × 0.60 × 0.02 × 0.98 = 23,520.

CVA is not merely an accounting number. It affects pricing, limits, collateral decisions, hedging, client selection, and capital allocation.

DVA

Debit valuation adjustment reflects the institution’s own default risk in the value of liabilities. In a bilateral framework, an institution’s liability may appear less costly because it may fail to pay its full obligation. DVA can therefore produce gains when the institution’s own credit spread widens, even though this reflects deteriorating financial condition.

DVA is conceptually controversial. It may be difficult to monetize through hedging because hedging one’s own default risk can be circular or impractical. It may also create counterintuitive reported gains during distress. Institutions must distinguish accounting measurement from economic risk management.

Other adjustments

Related adjustments may include:

• FVA: funding valuation adjustment for funding costs and benefits.
• MVA: margin valuation adjustment for the cost of funding initial margin.
• KVA: capital valuation adjustment for the cost of regulatory or economic capital.
• ColVA: collateral valuation adjustment for collateral remuneration and currency effects.

These adjustments are not universally additive in a simple independent way. Their treatment depends on the pricing framework, funding policy, collateral agreement, accounting rules, and whether a consistent no-arbitrage framework is being applied.

8. Worked integrated example

An institution enters a five-year CDS with a notional of 50 million. Assume a constant hazard rate of 4% per year, recovery rate of 40%, flat risk-free discounting at 3%, annual premium payments, and a simplified approximation that ignores accrual on default.

The approximate CDS spread is:

Thus the annual spread is approximately 2.40%, or 240 basis points. The annual premium on 50 million is:

Survival to year 5 is:

The cumulative default probability is approximately 18.13%. A simple undiscounted expected protection amount by year 5 is therefore:

This is not the exact present value of the protection leg because default timing and discounting have been simplified. It illustrates the relationship between notional, default probability, and recovery. In production pricing, the premium and protection legs would be calculated on each payment interval, with discount factors and default accrual.

9. Illustrative case: a leveraged loan hedge

Background

A regional bank holds 120 million of leveraged loans to companies in the communications sector. The loans generate attractive income, but their credit quality is highly correlated with advertising and consumer demand. The bank wants to reduce concentration without selling the loans, because selling would realize discounts and disrupt client relationships.

Problem

The bank considers buying CDS protection on a broad sector index and entering a TRS with an asset manager. The index hedge is cheaper than buying single-name CDSs, but it introduces basis risk. The TRS would transfer the economic return of a loan portfolio but could generate large collateral calls under stress.

Analysis

The bank identifies four risks. First, index constituents may not match its borrowers. Second, a broad market sell-off may cause the TRS receiver to owe variation margin precisely when liquidity is scarce. Third, the asset manager’s default could leave the bank with an unhedged portfolio. Fourth, legal netting may differ across jurisdictions and subsidiaries.

The bank estimates expected loss using scenario analysis. In a benign scenario, PD is 3% and LGD is 45%, producing expected loss of:

In a stress scenario, PD is 15% and LGD is 60%, producing:

These are portfolio approximations and do not capture default correlation, migration losses, market-value changes, or liquidity costs.

Approach and solution

The bank uses a layered hedge. It buys index CDS protection for systematic sector risk, retains a smaller amount of single-name protection for its largest exposures, and limits the TRS to a collateralized netting set with daily variation margin. It establishes an independent collateral valuation process, minimum transfer amounts, eligible collateral rules, close-out procedures, and counterparty limits.

The bank also calculates stressed expected exposure rather than relying solely on current exposure. It tests whether collateral remains adequate when spreads widen, interest rates move, and the counterparty’s credit quality deteriorates simultaneously.

Outcome

The hedge reduces systematic credit exposure but does not eliminate basis, liquidity, or counterparty risk. Senior management accepts the residual exposure because it is measured, limited, and supported by liquidity resources. The bank reviews the hedge when the loan portfolio changes and when index composition or market liquidity deteriorates.

Learning points

1. A hedge can transfer credit risk without eliminating total risk.
2. Legal netting and collateral terms are as important as the derivative’s quoted spread.
3. Stress testing must combine market movements, default, collateral calls, and counterparty deterioration.
4. Basis risk should be reported explicitly rather than hidden in an aggregate hedge ratio.

10. Governance, regulation, and ethical considerations

Credit derivatives require strong documentation, valuation control, model governance, and operational discipline. Master agreements, credit support annexes, definitions of credit events, settlement protocols, and netting opinions should be reviewed by specialists rather than treated as administrative details.

Regulatory frameworks generally seek to improve transparency, central clearing where appropriate, margining, trade reporting, capital adequacy, and resolution of failing institutions. Central clearing can reduce bilateral counterparty exposure through a central counterparty, but it concentrates risk in the clearing system and creates liquidity demands through margin calls.

Ethically, credit derivatives raise questions about incentives and information. A protection buyer may profit from a borrower’s failure, which can create reputational or contractual concerns even when the trade is legal. Complex securitizations can obscure risk transfer from investors. Institutions should disclose material assumptions, avoid misleading yield comparisons, and ensure that sales practices reflect client suitability and capacity to understand losses.

Model risk is particularly important. Small changes in recovery assumptions, default correlation, hazard-rate calibration, or exposure simulation can materially alter valuation. Independent validation should examine conceptual soundness, data quality, sensitivity, back-testing, stress testing, and limitations. Artificial intelligence may assist with document extraction, anomaly detection, collateral forecasting, and early-warning systems, but it should not replace expert judgment, legal review, model validation, or accountability. Explainability, data lineage, privacy, and bias controls are essential.

11. Contemporary developments and best practices

Current practice increasingly emphasizes integrated counterparty-credit and market-risk management. Important developments include real-time collateral optimization, portfolio-level XVA, central clearing, initial-margin models, standardized credit indices, climate-related credit scenarios, and machine-assisted surveillance.

Best practices include:

• Separate reference-entity risk from counterparty risk in reports and limits.
• Use legally enforceable netting sets rather than informal portfolio aggregation.
• Reconcile valuations, collateral, and dispute inventories frequently.
• Model exposure under stressed market conditions and wrong-way risk.
• Validate recovery, hazard-rate, correlation, and simulation assumptions.
• Include liquidity and margin-call capacity in hedge decisions.
• Report gross notional, replacement cost, potential future exposure, and expected loss separately.
• Maintain independent price verification and model governance.
• Document the economic purpose, hedge effectiveness, and residual basis risk of every material trade.

Chapter Summary

Credit derivatives transfer or reshape credit risk but do not make risk disappear. CDSs transfer specified credit-event losses; TRSs transfer total economic performance; CLNs embed funded credit protection in securities; and securitizations allocate pooled credit losses through waterfalls and tranches. Expected loss depends on exposure, probability of default, and loss given default, while market pricing additionally reflects discounting, liquidity, risk premia, and contract design.

Counterparty analysis focuses on positive replacement value after collateral and legally enforceable netting. Expected exposure and potential future exposure must be distinguished from expected loss. CVA prices counterparty default risk, while DVA reflects the institution’s own default risk; FVA, MVA, KVA, and collateral adjustments may also be relevant. These adjustments are model-dependent and should not be combined mechanically.

Effective risk management integrates legal documentation, collateral, valuation, stress testing, liquidity, governance, and ethical conduct. A credit hedge is successful only when its reference risk, basis risk, counterparty strength, collateral mechanics, and operational requirements are understood together.

Key Terms

Credit default swap: Contract transferring protection against defined credit events of a reference entity.

Protection buyer: Party paying the CDS premium and receiving protection upon a credit event.

Protection seller: Party receiving the CDS premium and making the protection payment.

Reference entity: Borrower or issuer whose credit event determines CDS settlement.

Recovery rate: Proportion of claim value recovered after default.

Loss given default: Proportion of exposure lost after default; generally 1 − recovery rate.

Hazard rate: Conditional instantaneous default intensity in a reduced-form credit model.

Total-return swap: Contract transferring price and income performance of a reference asset.

Credit-linked note: Funded security whose repayment depends partly on a reference credit event.

Securitization: Financing structure that pools assets and issues claims with specified priorities.

Tranche: Class of securitized claim with defined attachment, detachment, and payment priority.

Netting set: Transactions subject to an enforceable close-out netting agreement.

Variation margin: Collateral exchanged to cover current mark-to-market exposure.

Initial margin: Collateral intended to cover potential future exposure during close-out.

Expected exposure: Expected positive future replacement value of a derivative portfolio.

Potential future exposure: High-confidence estimate of possible future positive exposure.

CVA: Valuation adjustment for counterparty default risk.

DVA: Valuation adjustment associated with the institution’s own default risk.

Wrong-way risk: Situation in which exposure increases as counterparty credit quality deteriorates.

Basis risk: Risk that a hedge and the exposure do not respond identically to credit events or market movements.

Review Questions

Short Answer Questions

1. Distinguish probability of default, recovery rate, and loss given default.
2. Why is a CDS spread not necessarily equal to the expected real-world default loss rate?
3. What economic performance is transferred by a total-return swap?
4. How does collateral reduce counterparty exposure?
5. What is the conceptual difference between CVA and DVA?

Long Answer Questions

1. Explain the premium and protection legs of a CDS and discuss the assumptions behind simplified spread-to-hazard-rate approximations.
2. Compare CDSs, TRSs, and CLNs in terms of risk transferred, funding, leverage, and counterparty exposure.
3. Explain how attachment and detachment points determine losses in a securitized tranche.
4. Discuss the role of legally enforceable netting, variation margin, and initial margin in counterparty-risk management.
5. Critically evaluate the limitations of CVA and DVA as measures of economic risk.

Analytical / Application Questions

1. A portfolio has exposure of 80 million, PD of 4%, and LGD of 50%. Calculate simplified expected loss and explain two reasons why realized loss may differ.
2. A derivative netting set has transaction values of 9 million, −4 million, and 2 million, with 3 million of eligible collateral. Calculate simplified current exposure and explain the legal assumption required for netting.
3. A securitized tranche attaches at 2% and detaches at 6%. If pool loss is 5%, calculate the tranche loss as a percentage of tranche notional and explain the result.

Multiple Choice Questions with Answers

Question 1

A protection buyer under a CDS generally receives payment when:

A. The reference entity experiences a contractually defined credit event

B. Interest rates fall below the benchmark

C. The protection seller earns a trading profit

D. The reference entity’s share price rises

Correct Answer: A. The reference entity experiences a contractually defined credit event

Question 2

If recovery rate is 40%, the loss given default is:

A. 40%

B. 50%

C. 60%

D. 140%

Correct Answer: C. 60%

Question 3

Using the approximation s ≈ λ × (1 − R), a 6% CDS spread and 40% recovery imply a hazard rate of approximately:

A. 3.6%

B. 6.0%

C. 10.0%

D. 16.7%

Correct Answer: C. 10.0%

Question 4

A total-return swap normally transfers:

A. Only legal title to the reference asset

B. Price changes and income of the reference asset

C. Only the issuer’s voting rights

D. Only the counterparty’s funding cost

Correct Answer: B. Price changes and income of the reference asset

Question 5

A netting set has transaction values of 8 million and −5 million and eligible collateral of 1 million. Simplified current exposure is:

A. 1 million

B. 2 million

C. 3 million

D. 4 million

Correct Answer: B. 2 million

Question 6

CVA primarily compensates for:

A. The institution’s own operational expenses

B. Counterparty default risk on a derivative portfolio

C. Changes in the reference entity’s tax rate

D. The coupon paid on a government bond

Correct Answer: B. Counterparty default risk on a derivative portfolio

Question 7

A tranche attaches at 3% and detaches at 8%. If pool loss is 6%, what percentage of tranche notional is lost?

A. 30%

B. 50%

C. 60%

D. 100%

Correct Answer: C. 60%

Question 8

Wrong-way risk exists when:

A. Exposure decreases as counterparty credit quality worsens

B. Exposure is unrelated to market conditions

C. Exposure increases as counterparty credit quality worsens

D. Collateral is always greater than exposure

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

Question 9

Which adjustment reflects the institution’s own default risk?

A. CVA

B. DVA

C. PFE

D. LGD

Correct Answer: B. DVA

Question 10

For an exposure of 20 million, PD of 5%, and LGD of 40%, simplified expected loss is:

A. 200,000

B. 400,000

C. 800,000

D. 10,000,000

Correct Answer: B. 400,000