Skip to content
All library documents

Interpreting CDX Investment-Grade and High-Yield Quotes

Article Quant Q&A · Author: beeba

Summary

The document explains why North American investment-grade and high-yield CDX index series can appear to have unexpectedly different levels and volatility. The key is that the contracts use different quote conventions. Investment-grade CDX is quoted as a fair spread, while high-yield CDX is quoted as a price derived from the upfront payment. The contracts also carry fixed coupons, so their upfront payments generally move away from zero as market spreads differ from those coupons.

The response gives a rough relationship between upfront amount, tenor, market spread, and fixed coupon, then illustrates how to convert a high-yield quote into an implied spread. This conversion helps compare the indices on a common basis and interpret historical data. The relationship is explicitly approximate; the investment-grade conversion relies on the ISDA standard model, and the excerpt does not fully address whether older observations are suitable for a particular analysis. Raw quoted levels should therefore not be compared as though they were the same measure.

Key ideas

  • CDX investment-grade and high-yield indices use different quoting conventions, so raw levels are not directly comparable.
  • Investment-grade quotes represent fair spreads, while high-yield quotes are based on upfront value.
  • Fixed contractual coupons mean an index contract can require a positive or negative upfront payment.
  • Converting high-yield quotes into implied spreads makes the two index series easier to compare.
  • The stated conversion is approximate, and historical-data suitability is not fully resolved.

Tags

Full text
# Reliability of CDS indices?


# Reliability of CDS indices?












Looking at a time series two North American CDX indexes, one high yield and one investment grade, shows somewhat unexpected trends. CDX.NA.IG seems to be much higher and more volatile than CDX.NA.HY, which is the opposite of what I would expect. From Bloomberg:

The ticker for the IG index is IBOXUMAE, and for the HY index is IBOXHYAE. Is there any reason why the IG index was so much higher pre-2013? Can I still use pre-2013 data or would that be problematic?

## Answer by Chris Taylor (score 12, accepted)

https://quant.stackexchange.com/a/31918

The indices have different quoting conventions.

The way that a CDS index is traded is that you pay a fixed amount per year for protection in case of default (100 bps for IG, 500 bps for HY) and therefore the contract does not have a zero present value (as it would have if you paid the par spread, like in a fixed for floating interest rate swap). The amount you pay to enter the contract is the "upfront". If the fair spread is above the annual coupon, you will need to pay to enter the contract, i.e. the upfront is positive. If the fair spread is below the annual coupon, you will be paid to enter the contract, i.e. the upfront is negative.

When you see a quote for CDX IG, you are seeing the fair spread being quoted. There is a model used to turn this into an upfront payment (the ISDA CDS Standard Model).

CDX HY is quoted directly as the upfront (or rather, it is quoted as 100 minus the upfront, since the upfront can be negative, and dealers want to avoid using negative prices).

Very roughly, the relationship between the upfront and the spread is

$$ {\rm Upfront} = {\rm Tenor} \times ({\rm Spread} - {\rm Coupon}) $$

or

$$ {\rm Spread} = {\rm Coupon} + \frac{\rm Upfront}{\rm Tenor} $$

So when you see the CDX HY being quoted as 106, that means that the upfront is -6. Converting this to basis points gives -600. The coupon for CDX HY is 500 basis points, and the tenor is 5 years, so the spread is 500 + (-600)/5 = 380 basis points, i.e. buying protection on HY issues is cheap at the moment (relative to the CDS coupon).

Converting the CDX HY quote to a spread gives the expected relationship between the contracts -

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.