How do the threshold and minimum transfer amount in a CSA create residual unsecured exposure, and how is this quantified?
For FRM Part II, I understand that even with a CSA in place, exposure isn't fully collateralized because of the threshold and MTA. My question is: how do you actually calculate the maximum unsecured exposure that exists despite having a CSA? And how does this feed into the CVA calculation?
The threshold (H) and minimum transfer amount (MTA) in a CSA create a gap between the actual derivative exposure and the collateral held. This gap represents residual unsecured exposure — the amount at risk if the counterparty defaults at the worst possible moment within the CSA parameters.
Maximum Unsecured Exposure:
The worst-case unsecured exposure under a CSA is:
Max Unsecured Exposure = H + MTA + Potential market move during MPOR
This is because:
- The exposure can reach the threshold before any collateral is required
- The MTA means even when the threshold is breached, small excesses are not called
- During the MPOR, the exposure can move further without being collateralized
Step-by-Step Breakdown:
| Component | Source of Residual Exposure |
|---|---|
| Threshold (H) | Contractual permission to remain unsecured up to H |
| MTA | Operational minimum — calls below MTA are not made |
| MPOR move | Market movement between last clean call and close-out |
| Settlement lag | 1-2 day delay in receiving called collateral |
Worked Example: Fairpoint Securities has a CSA with Northgate Investments:
- Threshold (H) = 750,000
- MPOR = 10 business days
- Portfolio volatility (10-day, 97.5%) = 8M + 3.4M = 12.15 million loss if Northgate defaults at the worst moment. Compare this to a zero-threshold CSA: = 0.75M + 4.15 million
Reducing the threshold from 0"] --> B["MTM rises to 8.6M<br/>Excess = 0.75M)<br/>Still no margin call"] C --> D["MTM rises to 1.2M > MTA<br/>Margin call issued for 3.4M more<br/>before close-out"] E --> F["Worst-case unsecured:<br/>0.75M + 12.15M"]
**Impact on CVA:**
The expected exposure profile under a CSA with non-zero threshold is modeled as:
EE_CSA(t) = E[max(V(t) - C(t), 0)]
where C(t) is the collateral held at time t. The collateral is bounded by:
C(t) = max(V(t - delta) - H - MTA, 0) for V(t - delta) > H + MTA
Here delta represents the MPOR. The CVA under this CSA:
CVA_CSA = integral of DF(t) x EE_CSA(t) x PD(t) x LGD dt
is substantially lower than uncollateralized CVA but strictly greater than zero due to the threshold, MTA, and MPOR contributions.
Practice CVA calculations with CSA parameters in our FRM Part II question bank.
Master Part II with our FRM Course
64 lessons · 120+ hours· Expert instruction
Related Questions
Why is DV01 so much smaller than dollar duration if both are supposed to measure rate risk?
When should I stop using modified duration and switch to effective duration?
How should I think about the relationship between Macaulay duration and modified duration instead of memorizing two separate definitions?
Why do hedge calculations often use dollar duration or DV01 instead of just modified duration?
When should I prefer historical simulation VaR over delta-normal VaR?
Join the Discussion
Ask questions and get expert answers.