Counterparty Credit Risk: Exposure That Moves with the Market

Key Takeaways

• Counterparty credit risk is forward-looking: a swap at zero mark-to-market today can drift millions in either direction over its life, so potential future exposure (PFE) matters more than current exposure.
• The SA-CCR formula multiplies an EAD estimate by alpha of 1.4 to account for correlation between exposure and default, the regulatory recognition that the two are not independent.
• Three mitigants compress CCR: ISDA netting (collapsing many trades to one net number), daily variation margin (reducing exposure to one-day gap), and central clearing (replacing bilateral exposure with a heavily margined CCP).
• CVA (credit valuation adjustment) is the market price of counterparty credit risk embedded in a derivative's fair value; CVA volatility is itself a separate Basel capital charge.

What It Is

CCR is bilateral and forward-looking. If you are in the money on a swap and the counterparty fails, you must replace the position at current market levels and may recover only a fraction of what you were owed. If you are out of the money, you owe them, but their failure does not usually create a loss for you beyond operational cleanup.

Regulators treat CCR as a separate capital charge from classical credit risk because exposure changes minute to minute. The Basel framework uses SA-CCR (the standardised approach for measuring counterparty credit risk exposures, BCBS 279) as the default calculation, with internal models (IMM) permitted for banks that have been approved.

The Intuition

A five-year interest rate swap has no loan balance. At origination, its mark-to-market is usually zero. Over time, rates move, one side goes into the money, and that mark-to-market is what is at risk if the losing side defaults. The exposure at default is therefore the positive market value plus the potential future exposure over the remaining life.

This is why CCR lives in the space between market risk and credit risk. You need a market model to project future exposures and a credit model to price the probability and loss on default.

How It Works

The core quantity is Expected Exposure (EE) at future time t, the average of the positive mark-to-market distribution. Integrating EE across time gives Expected Positive Exposure (EPE), and the regulatory figure is typically effective EPE multiplied by an alpha of 1.4.

The capital charge for a counterparty follows the standard IRB shape:

CVA capital ~ f(EAD, PD, LGD, maturity)

Where EAD is exposure at default, PD is probability of default, and LGD is loss given default. For senior unsecured derivatives claims, LGD is typically assumed at 40 to 60 percent.

Three mitigants compress exposure:

• Netting. An ISDA Master Agreement lets you offset in-the-money and out-of-the-money trades with the same counterparty, collapsing many positions into one net number.
• Collateral under a Credit Support Annex (CSA). Daily variation margin, often with a threshold and minimum transfer amount, plus initial margin for uncleared derivatives per the BCBS-IOSCO framework and ISDA SIMM.
• Central clearing. Novating a trade to a central counterparty (CCP) replaces bilateral exposure with exposure to the CCP, which is mutualised and heavily margined.

Banks also carry a Credit Valuation Adjustment (CVA), the price of counterparty credit risk embedded in a derivative's fair value. CVA volatility is itself a capital charge under the Basel CVA framework.

Worked Example

A bank holds a 10-year pay-fixed interest rate swap with a corporate, notional 100 million, currently 3 million in the money. There is a CSA with a 1 million threshold and daily variation margin, no initial margin.

• Current exposure: 3 million mark-to-market, minus 2 million collateral received (held above threshold), equals 1 million uncollateralised.
• Potential future exposure under SA-CCR: 100 million notional times a supervisory factor of 0.5 percent for interest rate swaps over one year, scaled for residual maturity, gives roughly 3.5 million.
• EAD after alpha: 1.4 x (1.0 + 3.5) equals about 6.3 million.

If the corporate has a 2 percent one-year PD and a 55 percent LGD, expected loss is 6.3 x 0.02 x 0.55, roughly 69 thousand per year. The unexpected loss piece, plus CVA volatility, drives the Basel capital number well above that.

Frequently Asked Questions

Q: What is counterparty credit risk in simple terms? CCR is the risk that the firm you traded with fails before your contract is settled, and you have to replace it at today's market price. Unlike a loan, both sides can be owed money at different times as prices move, so the exposure is not fixed.

Q: How does counterparty credit risk affect investment decisions? Banks set per-counterparty CCR limits based on expected positive exposure and credit quality. When a counterparty's rating falls, new trades are refused, existing trades are priced with higher CVA, and collateral requirements tighten, all of which can strain the counterparty further in a stress event.

Q: What is a real-world example of counterparty credit risk? A bank holds a 10-year interest rate swap 3 million in the money with a corporate. Under the CSA, 2 million of variation margin has been posted, leaving 1 million uncollateralised. If rates move sharply over a weekend, mark-to-market could jump to 5 million, making the unsecured gap 3.2 million before the next margin call.

Q: How can institutions reduce counterparty credit risk? Require daily variation margin under a tight CSA, use initial margin for uncleared derivatives per the BCBS-IOSCO framework, clear standardised products through a CCP, and net all trades with the same counterparty under an ISDA Master Agreement.

Q: How is counterparty credit risk different from standard credit risk? Standard credit risk (on a bond or loan) has fixed principal and one-directional exposure. CCR has fluctuating, bilateral exposure that moves with market prices every day. The tools are different, netting, margining, PFE modelling, because the risk structure is fundamentally different from a static loan balance.

Common Mistakes

1. Treating current exposure as the full risk. A swap at zero mark-to-market today can drift tens of millions in or out over its life. Potential future exposure is the number that matters for limits.

2. Netting across the wrong legal entity. Netting only works where a legal opinion confirms enforceability in the counterparty's jurisdiction. Assuming a global net when the opinion is limited to one entity is a standard audit finding.

3. Ignoring wrong-way risk. Exposure that rises exactly when the counterparty's credit deteriorates amplifies loss. The 2008 monoline insurers are the textbook case.

4. Underestimating CCP concentration. Clearing reduces bilateral risk but piles everyone into a few CCPs. Default fund contributions and stress-loss waterfalls are the offsetting exposure.

5. Stale collateral haircuts. A CSA written in 2014 with generous haircuts on non-cash collateral can understate true exposure in stressed markets. Annual CSA review is the discipline.

Sources

1. Basel Committee on Banking Supervision. "The standardised approach for measuring counterparty credit risk exposures." BCBS 279. https://www.bis.org/publ/bcbs279.htm
2. Basel Committee on Banking Supervision. "Basel III counterparty credit risk and exposures to central counterparties." https://www.bis.org/bcbs/publ/d325.htm
3. Federal Reserve. "Regulation Q, Risk-Based Capital Requirements." https://www.federalreserve.gov/supervisionreg/reglisting.htm
4. ISDA. "ISDA SIMM Methodology." https://www.isda.org/a/CeggE/ISDA-SIMM-v2.6-PUBLIC.pdf

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.