FRM® credit risk mitigation: what does each tool change?

On this page

Credit risk mitigation works by reducing the exposure at risk, improving recovery after default, or transferring loss to a protection provider. Collateral, enforceable netting and a guarantee therefore do different jobs; none makes the borrower inherently more reliable. For an FRM® Part II answer, identify the loss component affected and then ask whether the protection will still work at default.

Locate the risk before naming a product

A lender first needs to distinguish the chance of borrower default from the amount exposed and the fraction ultimately lost. Expected loss can be expressed as probability of default times exposure at default times loss given default. That is an analytical framework, not a statement that a bank knows any input with certainty. A pledge of valuable collateral can improve recovery and lower loss given default while the borrower's own probability of default remains the same. Netting can reduce a positive exposure across a set of contracts. A guarantee substitutes a claim on another party, introducing the quality and correlation of that party into the analysis.

The tool's legal and operational conditions matter as much as its name. A promise written in a contract is useful only if the bank can enforce it, identify the covered exposure and collect or liquidate when needed. The Basel Committee's credit risk mitigation chapter sets legal certainty and operational conditions for regulatory recognition. Those rules are a practical reminder that economic protection and recognised capital relief require separate tests. GARP's FRM study materials locate credit risk measurement and management within the Part II curriculum.

Collateral changes recovery, subject to value and control

Collateral is an asset the lender can claim under an enforceable security interest if the borrower fails. Its immediate economic purpose is to reduce the amount left unrecovered. That improvement depends on priority, the asset's realisable value, the costs and time of sale and whether the lender can take possession. A valuation made on a calm day can overstate what a distressed sale will yield. A cash deposit and specialised equipment can both be described as collateral, yet their liquidity and valuation uncertainty differ sharply.

Consider an entirely illustrative loan with exposure at default of 10 million units, a borrower default probability of 2% over the measurement period and an unsecured loss given default of 60%. Its illustrative expected loss is 120,000 units: 10 million multiplied by 0.02 and 0.60. Suppose enforceable collateral would produce 7 million units after a deliberately assumed stress adjustment and all sale costs. The residual loss would be 3 million units, giving a secured loss fraction of 30% and expected loss of 60,000 units if the borrower probability remains 2%. This example does not estimate an actual loan, collateral haircut or regulatory capital charge. It shows why reducing loss severity is different from reducing the chance of default.

A candidate should challenge the recovery assumption. If the pledged asset is closely linked to the borrower's fortunes, it may lose value when the borrower deteriorates. The Basel chapter specifically addresses correlation between counterparty credit quality and collateral value. Security over an asset is not the same as cash in hand. Check who owns it, whether another creditor has a senior claim, how quickly it can be sold and whether a margin process updates protection when market values move. These facts can change the effective recovery without changing the nominal pledged amount.

Netting changes the amount at risk

With several derivatives against one counterparty, the bank may owe money on some contracts and be owed money on others. An enforceable close-out netting arrangement can bring covered contracts into a single net claim at termination. For a simple illustration, positive mark-to-market values total 9 million units and negative values total 4 million units. Gross positive replacement exposure is 9 million units; the simple net positive amount is 5 million units. The reduction is 4 million units, not 9 million, and future market moves can change the figures. This is illustrative arithmetic, not a regulatory exposure-at-default calculation.

Netting is conditional on which trades and entities the agreement actually covers and on its enforceability in relevant jurisdictions. A trade with a different legal entity cannot simply be put into the same net set because the group name looks familiar. If an insolvency court does not uphold close-out rights, an assumed offset may disappear exactly when it matters. The Basel Committee requires legally binding, enforceable documentation for recognised netting and other credit mitigants. On an exam case, therefore, the phrase 'subject to enforceable netting' is a material condition rather than a decorative qualification.

Guarantees and credit derivatives move the claim

A guarantee gives the lender a claim against a protection provider for a specified obligation. A credit default swap is a contract under which the buyer pays for defined credit protection and the seller owes a payment after a covered credit event, subject to its terms. In either case, the original borrower remains capable of defaulting. The lender has changed who must absorb the relevant loss if the protection performs. A stronger and independent provider may make this valuable; a weak provider whose finances move with the borrower may add little protection.

To reason through a guarantee, ask four questions. What amount and maturity are covered? What event triggers payment? Can the guarantor dispute or cancel the obligation? Can it pay at the time the borrower cannot? A guarantee from an affiliate funded by the same stressed business may fail the last test. A credit derivative needs similar scrutiny of reference obligation, covered events, settlement and seller counterparty risk. The Basel chapter specifies that eligible guarantees and credit derivatives must represent direct, clearly defined claims and sets further recognition conditions. Do not leap from 'guaranteed' to 'no credit risk'.

Suppose a lender has an illustrative 10 million unit claim, of which a third party guarantees 6 million units. If the guarantee fully performs when called, the lender still has uncovered borrower exposure on the remaining 4 million units. It also has a contingent claim on the guarantor for the protected part. Treating all 10 million as risk free would omit both the uncovered slice and the guarantor's ability to pay. The simplified split helps identify claims; actual loss and capital treatment depend on contract details, recoveries, maturity and applicable rules.

Covenants and sales affect the path of exposure

Covenants can create reporting duties or restrict actions such as additional borrowing. A maintenance test is assessed according to its agreed schedule; an incurrence test becomes relevant when the specified action occurs. A breach may give the lender rights to intervene, but those rights and remedies depend on the documents. An early warning can help preserve value or stop exposure from increasing. It does not mechanically lower an already measured probability of default. A candidate should distinguish the contractual control from a realised reduction in loss.

A loan sale transfers a claim to a buyer, and a securitisation can move or redistribute credit risk through a defined structure. These measures need care because selling a claim, transferring risk synthetically and retaining a first-loss interest are different outcomes. The originator may still bear servicing obligations, residual positions or other exposures. The useful question is which risk remains with the original holder after the transaction, not whether the transaction has a familiar label.

A compact way to solve a mitigation case

Write down the starting borrower, exposure and expected recovery. Name the proposed tool, then state exactly what changes: exposure, loss severity, or the party from whom payment is expected. Apply any illustrated arithmetic only after setting assumptions. Finally test legal enforceability, valuation and wrong-way dependence. This sequence prevents the frequent error of saying that collateral lowers the borrower's probability of default, or that netting eliminates every positive claim.

The FRM programme page puts Part II preparation in context, while the Omni methodology page describes the study approach. For exam work, the central skill is to connect an instrument to the relevant loss channel and then articulate the condition that could make the apparent hedge fail. A product description without residual risk is only half an answer.