Skip to content
All library documents

Premium Accrual for Short-Tenor CDS Under the 2015 Roll Convention

Article Quant Q&A · Author: Francis

Summary

The document explains how premium accrual works for short-dated, SNAC-like credit default swaps under the 2015 roll convention. It distinguishes a three-month contract, whose premium leg has one coupon accruing from the prior standard date through maturity, from a six-month contract, whose premium leg has two coupons, with the first paid on the intervening standard date and the second ending at maturity.

The examples concern trades quoted on March 19, 2020, and clarify that a coupon payment date falling after maturity does not imply an extra coupon for the three-month contract. Protection can extend through the maturity date, but the response says there is no corresponding extra day of accrued premium. These are convention-specific examples; implementation should follow the applicable CDS documentation and contract terms.

Key ideas

  • A three-month SNAC-like CDS has one premium coupon accruing through its maturity date.
  • A six-month SNAC-like CDS has two coupons, with the first paid on the intervening standard date.
  • Protection applies on the maturity date, but that date does not add an extra day of premium accrual.
  • Short-tenor accrual should be interpreted using CDS conventions rather than coupon-date logic alone.

Tags

Full text
# CDS - 2015 roll convention and short CDS tenors


# CDS - 2015 roll convention and short CDS tenors












I am reading the ISDA document here regarding the change to roll conventions on CDS that came in at the end of 2015 and in particular section 11 relating to the short end tenors. Additionally, I am reading the ISDA standard CDS examples document here around the standardisation of CDS coupons.

I think that my question is best asked in the context of an example. On Thu 19 Mar 2020, we receive both 3M and 6M CDS quotes from our market data provider.

For the 3M quote, using the CDS 2015 roll conventions, the maturity date is Fri 20 Mar 2020. The second document referenced above has the following statement in point 5 in the Standardizing coupon dates section:

> A trade's first coupon payment date is determined by the trade date (T): it's the first coupon payment date after T+1 (calendar, unadjusted). This is consistent with the first coupon dates of, eg, CDX.

If I took this literally, I would get a first coupon payment date of Mon 22 Jun 2020 on the 3M CDS above which would not make sense. Is it correct to ignore this here and to assume that the 3M CDS has a fee leg, with a single coupon, with accrual starting on Fri 20 Dec 2019 and ending on Fri 20 Mar 2020 (inclusive of Fri 20 Mar 2020)?

For the 6M quote, the CDS 2015 unadjusted maturity is Sat 20 Jun 2020. Using the statement quoted above, the first and only coupon payment would be on Mon 22 Jun 2020. In other words, is it correct to assume that the 6M CDS traded on Thu 19 Mar 2020 has a fee leg, with a single coupon, with accrual starting on Fri 20 Mar 2020 and ending on Sat 20 Jun 2020 (inclusive of Sat 20 Jun 2020) with payment on Mon 22 Jun 2020?

## Answer by Dimitri Vulis (score 1, accepted)

https://quant.stackexchange.com/a/59395

You are correct. The premium leg of a 3 months SNAC-like CDS has one coupon. In your example, it accrues since December 20 2019 and ends at maturity on March 20, 2020.

(Please observe that SNAC-like CDS (any, not just 3 months) provides protection on maturity date (if a credit event happens exctly on that day, which sometimes happens) but there isn't a corresponding extra day of accrued premium.)

Likewise a 6 months SNAC-like CDS has two coupons. In your example, the first coupon accrues since December 20 2019 and is paid on March 20, 2020; the second coupon accrues since March 20, 2020 and ends at maturity on June 20, 2020.

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.