FRM® credit derivatives: where did crisis risk return?

On this page

Credit derivatives did not make a borrower's default disappear during the global financial crisis. They moved particular losses to another party, while leaving the buyer exposed to the seller's ability to perform, the contract's terms, and risks elsewhere in the lending chain. For FRM® Part I, trace each exposure after a transfer: who owns the loan, who bears its loss, who must pay, and what happens if several positions fail together.

Start with the risk being transferred

A lender can hold a loan and buy credit default swap protection on a reference borrower. If the specified credit event occurs and the contract conditions are met, the protection seller owes a settlement amount; the buyer pays premiums for that protection. The lender still has the loan. It has changed the distribution of its credit loss, not turned the loan into cash or removed every risk from its balance sheet. BIS research on credit risk transfer makes the distinction especially useful: an unfunded promise to absorb a loss creates exposure to the provider of that promise.

A cash securitisation works differently. Loans are sold into a structure that issues claims to investors, with cash flows and losses allocated by the security terms. A synthetic risk transfer can leave the loans at the bank while shifting a defined slice of credit risk through a derivative or guarantee. The Basel Committee's account of synthetic risk transfers describes that separation between ownership and risk. For an exam scenario, do not infer loan ownership merely from the statement that credit risk has been transferred. Ask what was sold, what was guaranteed, and which exposure remains.

Why the transfer chain became fragile

The danger is easiest to see when each participant considers only its immediate contract. An originator may expect a loan sale, an investor may rely on a security's position in a payment waterfall, and a protection buyer may rely on a swap. Those positions can all depend on the same underlying borrowers. Packaging claims changes who receives losses first, but cannot improve the borrowers' capacity to pay. When defaults rise together, diversification that looked convincing under ordinary conditions can prove weak because losses are correlated. The same shock may impair the reference assets and the firms that promised protection.

That last link is counterparty risk. A credit default swap can be valuable when the reference borrower deteriorates, precisely when the protection buyer most needs payment. If the seller's finances also deteriorate, the buyer may face an unpaid claim or costly replacement protection. This is a form of wrong-way exposure, not simply an arithmetic error in the expected loss on the loan. The BIS history of the CDS market records rapid pre-crisis growth and later reforms aimed at transparency and resilience. It does not imply that all protection sellers failed or that every contract lacked collateral.

Concentration matters because many apparently separate hedges can point to a small group of protection providers. A lender may see a diversified set of borrowers on its own books, yet have concentrated its credit hedge on one counterparty. At the system level, several lenders may have made the same choice. A stress to that seller then weakens many hedges together. The BIS credit risk transfer report had identified both concentration and rapid changes in counterparty exposure as concerns before the crisis. The exam lesson is a dependency map: identify repeated names behind positions that appear diversified.

An illustrative loss map

Consider an invented portfolio of loans with a face value of 100 monetary units. A lender buys protection covering the first 20 units of defined credit loss from one seller. Suppose eligible borrower losses amount to 12 units and the seller fully performs. Ignoring premiums and recovery details for this simple illustration, the seller pays 12 and the lender's direct borrower loss is offset by 12. Now suppose the seller can pay only 5. The lender has received 5, and the remaining 7 of its expected protection has become a counterparty shortfall. The borrower loss did not vanish when the swap was signed.

Change one assumption at a time to understand the same example. If borrower losses are 25 and the seller performs, protection stops at the agreed 20; the lender still bears 5. If the contract covers only a named reference pool while the lender has other loans, losses outside that pool are not protected. If the contract expires before the loans, the lender again has an uncovered period. These are different failure modes: limit risk, basis or scope mismatch, and maturity mismatch. Do not combine them under a vague claim that the derivative 'failed'. Identify the clause or exposure that produces each residual loss.

A tranched security requires a similar sequence. Determine which cash flows enter the structure, which tranche absorbs initial losses, and which tranche is reached later. Seniority tells you the order of loss allocation, not whether the total pool is safe. If many borrowers are exposed to the same economic shock, losses may move through several layers faster than a model based on independent defaults suggests. The point is conceptual here: there is no universal safe tranche and no single correlation number to memorise from this article. Work from the transaction's stated waterfall in the question.

What the crisis demonstrates, and what it does not

The global financial crisis exposed weak underwriting, opaque structures, correlated mortgage losses, funding pressure and counterparty dependencies. A credit derivative could legitimately hedge a specified exposure, but a hedge was only as effective as its contract, collateral and counterparty. Treat the crisis as a test of how risk moved through institutions rather than a verdict that every derivative was intrinsically harmful. The BIS account of post-crisis securitisation reforms describes reforms to disclosure, risk sensitivity and reliance on ratings, which helps separate the instrument from the surrounding incentives and controls.

Avoid a second oversimplification: selling a loan or transferring a slice of its credit risk does not necessarily end an originator's economic relationship with it. It may continue to service the loan, hold another tranche, provide liquidity, or face reputation pressure to support an associated vehicle. The precise obligations depend on documents, but the analytical habit is constant. Write down contractual obligations and then ask about possible exposures beyond them. A bank that expected to distribute loans can also be left holding inventory if investor demand disappears. The crisis made that warehousing risk visible.

Modern structures should not be read as identical to pre-crisis arrangements. The Basel Committee reports that post-crisis synthetic transfers face more scrutiny and that today's structures can use collateral or funded notes. Its separate analysis of the economics of synthetic transfers explains how collateral can address investor counterparty exposure and why rollover risk can still matter to issuing banks. These are useful distinctions, not a claim that all present structures are safe. For an FRM answer, date the institutional setting and inspect the actual safeguards described in the case.

How to answer a first encounter question

Begin with the underlying asset: identify the borrower and the event that produces an economic loss. Next draw each transfer arrow, including a sale, guarantee or swap, and mark whether the transaction moves an asset, a defined credit loss, or both. Then name the party responsible for settlement and the source of its payment capacity. Finally check limits, coverage dates, collateral, concentration and correlation. This order prevents the common mistake of treating 'hedged' as a complete description. It also prevents counting the same loss twice when a loan and its related derivative appear on different lines of a case.

For practice, use the FRM programme overview to place this concept within the wider risk curriculum, then apply the Omni study methodology to reconstruct the exposure map without looking at notes. Try a variation in which the hedge is partial, another in which the seller fails, and a third in which the loans are actually sold. Explain each residual exposure in a sentence before calculating anything. That is more useful than memorising crisis anecdotes whose figures may belong to a different source, date or market definition.

The distinction to retain

Credit risk transfer is a change in who bears a specified loss under specified conditions. The financial crisis showed how difficult it can be to see the final bearer when loans, securities, swaps and institutions are linked. On an exam, follow the loss from borrower to investor or protection seller, and then test whether the promised transfer would still work under stress. A correct answer usually identifies the exposure that remains after the apparent hedge, rather than declaring the entire transaction either risk-free or useless.