Decomposing Credit Default Swap Returns

Summary

The document considers how to calculate returns for a portfolio combining equities and credit, using an equity index and an investment-grade credit index as examples. It cautions that a credit default swap position does not have a meaningful return from spread changes alone. Its change in value has at least two components: the effect of paying the contractual premium accrual and the mark-to-market effect of spread movements.

For the spread component, the answer proposes multiplying the spread change by spread duration, with a convexity adjustment for greater precision. The response is brief and does not specify sign conventions, the exact premium accrual calculation, or the assumptions needed to translate these components into daily portfolio returns. It therefore gives a useful first-order decomposition, rather than a full return methodology or a worked calculation.

Key ideas

• A credit default swap position's value change includes premium accrual as well as spread movements.
• Spread duration provides a first-order approximation of the mark-to-market effect of a spread change.
• A convexity adjustment can improve the spread-based approximation.
• Portfolio return calculations require consistent treatment of both components and their sign conventions.

Tags

# Equity and Credit Portfolio Return


# Equity and Credit Portfolio Return












This might sound like a trivial question but would appreciate the answer. How would you calculate the return of the portfolio consisting of only equity and credit instruments? For example, consider only two assets S&P 500 and CDX IG and assume that they have equal weighting at 50%. In order to get daily portfolio returns we need daily returns for equity and credit. For equity it is simple (just daily price returns) but it is confusing how to do that for credit. I am assuming credit is quoted in spreads according to the convention. Would you just take a daily change in the credit spread and multiply that by spread duration? This wouldn’t be entirely correct as that’s not the only component of credit return but do you think it is a good approximation?

Thank you.

## Answer by Tarun Bhasker L (score 0)

https://quant.stackexchange.com/a/73727

Return on CDS alone doesn't make sense. \$ change in CDS position has 2 sources.

- 100bps accrual. This is a negative \$ impact.

- Duration*(change in spread). You can use 2nd order convexity as well to be a bit more precise.

Thanks,

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.