Choosing a Duration-Matched Benchmark for Bond Credit Spreads

Summary

The document asks how to construct a credit spread for a fixed-income bond by subtracting a risk-free benchmark yield. It describes two measures from policy papers: yield to maturity less a similar-duration overnight index swap rate, and effective yield less a German government zero-coupon rate described as having similar duration. The author questions whether the second measure requires bootstrapping a zero curve and selecting a spot rate, and whether the first requires choosing an OIS tenor matched to the bond.

No answer or calculation is provided, so the text does not settle how duration matching should be implemented or whether maturity, modified duration, or another measure is intended. Its useful point is the methodological ambiguity: a spread comparison depends on the benchmark curve, the yield measure, and how “similar duration” is defined. The document offers questions about curve construction and tenor selection, rather than evidence that one particular matching procedure is correct.

Key ideas

• The document compares bond yields with OIS and government zero-coupon benchmarks to define credit spreads.
• One cited approach uses an OIS rate with duration similar to the bond.
• Another subtracts a German government zero-coupon rate described as duration matched.
• The author asks whether zero-curve bootstrapping and OIS tenor selection are required.
• No response resolves how duration should be matched or how the proposed spreads should be calculated.

Tags

# Duration adjusted credit spread


# Duration adjusted credit spread












I am currently reading two policy papers that construct, for a fixed-income bond, a following credit spread:

- This one, p. 12, computes the yield to maturity minus the Overnight Index Swap (OIS) rate of a similar duration (let's assume that the underlying Interest Rate of the OIS is EONIA)

- This one (paywalled), p. 125, subtracts "from the effective yield the German bund zero coupon interest rate of a similar duration"

It is unclear to me what the authors did precisely in terms of the subtrahend, no further explanations are given.

For the second measure, would I have to bootstrap the zero curve and take the n-year spot rate where n equals the duration of the bond? However, duration is not the same as maturity.

For the first measure, OIS are bilateral agreements where the floating rate is based on the EONIA, under the above assumption. Would I have to select the rate of the agreement that has a duration which matches the one of the bond?

Thank you for your help!

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.