Wrong-Way Risk: a Worked Example

A loan's exposure is known: the balance drawn, plus whatever headroom the borrower can still draw. A derivative's exposure is not. It depends on where markets go, so a bank measures it as a range of future values, the potential future exposure that a derivative credit limit is sized on. That range is usually built as if the market and the customer were unrelated. Wrong-way risk is the case where they are not, and the high end of the range arrives in the same months that the customer's probability of default rises.

The literature separates two kinds. General wrong-way risk comes from the economy: a market factor such as interest rates, a currency or a commodity price moves in a way that both enlarges the exposure and weakens the customer. Specific wrong-way risk comes from the transaction itself, where the structure ties the exposure to the customer's own credit, such as protection bought from a company on its own group or collateral issued by the borrower. The UK's capital rules name both in article 291 of the retained Capital Requirements Regulation, in the part written for banks that model their derivative exposures, and require them to be identified, stress-tested and reported to senior management.

There is no single formula. Banks with models compare the exposure they expect in scenarios where the counterparty is close to default with the exposure they expect on average, and treat a large gap as the warning. The standardised measures have no such step. Australia's counterparty credit standard, APS 180, has no internal-model route: banks measure derivative exposure with the standardised approach (SA-CCR) or, where not approved to use it, an adjusted current exposure method, both built from fixed supervisory factors that do not change with the counterparty's own credit, and the standard does not use the term. For a mid-market lender the work is therefore mostly judgement, and it starts with three questions.

The same reading applies to support. A guarantee, a pledge or a piece of collateral reduces risk only if it holds its value when the borrower does not. UK capital rules make the point for financial collateral: the obligor's credit and the collateral's value must not have a material positive correlation, and securities issued by the obligor or a related group company are not eligible, outside a narrow exception for its own covered bonds posted in a repurchase transaction.

A scaffolding hire business, constructed for this page like every figure in it, borrows US$12m on a five-year floating-rate term loan and fixes the rate with a pay-fixed swap at 4.40% on the full amount. Its earnings follow building activity. A year later construction has slowed, the central bank has cut, and the swap rate for the four years left has fallen to 2.40%. The rates and the business's figures are the example's, and the swap is valued with annual payments for simplicity.

Every line moved against the bank at once. The swap did its job for the borrower's interest cost, but it took away the relief that falling rates would otherwise have given, and it added a liability of US$0.9m that would be payable in full if the swap were closed out on a default. Meanwhile the earnings and the resale value of the security fell for the same reason the rates did. A limit sized on the swap alone, without asking what the business would look like when the swap was worth most, would have read this as a hedge working as intended.

The business's sister company, which makes formwork for the same builders, had guaranteed the loan. In the year the bank might need it, its order book was falling too.

Right-way risk is the opposite pairing: the exposure shrinks as the customer weakens. A hedge of the business's own trading risk tends to work this way. An importer that buys foreign currency forward is hurt when its home currency falls, but in that case the forward is worth money to the importer, so the bank's claim on it falls just as its costs rise. An exporter that sells currency forward is the mirror image. The scaffolding swap differs because the rate cut is a symptom of the downturn, so the hedge pays the bank in the year the customer is weakest.

It also differs from connected exposure and concentration. Those ask whether several exposures will fail together; wrong-way risk asks whether one exposure grows as its counterparty fails. A file can show both, as the sister-company guarantee does.

DEFINED IN ONE PARAGRAPH → · SWAP MARK-TO-MARKET → · DERIVATIVE CREDIT LIMIT →

The first place is the derivative limit. A credit officer who sees wrong-way risk in a hedge can size the limit on a stressed scenario in which the borrower is also weak, shorten the tenor, or hedge less than the full loan, which is one reason a bank's hedging policy and the borrower's hedge ratio are read together. The second is the security. Facility documents commonly make the swap obligations secured money ranking with the loan, so that a close-out amount does not fall outside the security when it is largest.

The third is support. A guarantee from a company in the same group or the same market, or a share pledge over the borrower's own group, is counted for what it would be worth on the day it is called, and in a wrong-way file that is less than its value today. The borrower's side of the reading is in the interest rate swap guide: the break cost that makes a swap expensive to leave is the same amount the bank is owed.

HOW BANKS MANAGE CREDIT RISK → · INTEREST RATE SWAP, DEFINED →