Charta is on Google Play: practise on the go. Get the app
Charta

Outline the different credit risk mitigants, set them side by side and show what part they play in credit analysis.

FRM® Part II · Credit Risk Measurement and Management · Credit Risk · FRM2.CR.12.d

The idea

Credit risk mitigants reduce the loss from a given counterparty without changing how likely it is to default. Close-out netting lets all contracts with one counterparty be terminated and combined into a single net claim, so positive and negative values offset. Collateral, usually cash or securities under a margin agreement, lowers the unsecured exposure, after a haircut on the collateral's value. Downgrade triggers let you demand collateral or terminate if the counterparty's rating falls below a level. In credit analysis they act on exposure at default and loss given default, not on the default probability. Let E be the unsecured exposure after netting and collateral, and R the recovery rate. Then: Loss = E × (1 − R). Each mitigant has residual risks: legal enforceability, collateral volatility and delay, and wrong-way risk when collateral is tied to the counterparty. Exam point: apply collateral first, then recovery to what remains.

Formulas

Loss after collateral and recovery

loss = (EAD − C × (1 − h)) × (1 − R)

EAD
the exposure at default, after any netting
C
the market value of the collateral
h
the haircut on the collateral
R
the recovery rate on the unsecured claim

Collateral comes off first, then recovery applies to the unsecured remainder. Applying recovery to the whole exposure first understates the loss.

A practice question

From the question bank, with the reason each wrong option is wrong.

Under an enforceable close-out netting agreement, what happens when a counterparty defaults?

  1. Only the contracts with positive value to the surviving party are closed out; the negative ones continue to be performed.Why not: Describes cherry-picking, which netting agreements are designed to prevent.
  2. Only payments falling on the same date are offset, and each contract otherwise stays separate.Why not: Describes payment netting, not close-out netting of contract values.
  3. The surviving party takes ownership of a pledge of assets equal to its exposure.Why not: Confuses netting with collateral; netting is a legal offset, not a pledge of assets.
  4. All covered transactions are terminated and their values combined into a single net amount owed by one party to the other. - correct

Close-out netting turns many contracts into one claim, so the surviving party's exposure is the net value rather than the sum of the positive values.

A harder one

A bank has an uncollateralised-equivalent exposure of USD 12 million to a counterparty and holds USD 7 million of cash collateral with no haircut. Recovery on any unsecured claim is 40%. The loss if the counterparty defaults is closest to:

  1. USD 3.0 million. - correct
  2. USD 7.2 million.Why not: Applies recovery to the full exposure and ignores the collateral.
  3. USD 2.0 million.Why not: Uses the unsecured amount of 5 but multiplies by the recovery rate instead of 1 - R.
  4. USD 0.2 million.Why not: Applies recovery to the full exposure first (12 x 0.6 = 7.2) and then subtracts the collateral.

Collateral is applied first: unsecured exposure = 12 - 7 = 5. Loss = 5 x (1 - 0.40) = 3.0.

The hardest tier

Exposure at default is USD 20 million. The bank holds securities with a market value of USD 15 million as collateral, subject to a 20% haircut. Recovery on any unsecured claim is 25%. The bank's loss is closest to:

  1. USD 3.75 million.Why not: Ignores the haircut and treats the collateral as worth its full 15 million market value.
  2. USD 2.0 million.Why not: Uses the recovery rate (25%) as the loss rate on the unsecured amount of 8.
  3. USD 3.0 million.Why not: Applies recovery to the whole exposure first (20 x 0.75 = 15) and then subtracts the haircut collateral (12).
  4. USD 6.0 million. - correct

Collateral credit = 15 x (1 - 0.20) = 12. Unsecured exposure = 20 - 12 = 8. Loss = 8 x (1 - 0.25) = 6.0.

In the app: 6 practice questions on this outcome, spaced review that brings it back before you forget it.

Start free →

More in Credit Risk